No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Exposing digital forgeries in video by detecting duplication
Abstract
With the advent of high-quality digital video cameras and sophisticated video editing software, it is becoming increas- ingly easier to tamper with digital video. A common form of manipulation is to clone or duplicate frames or parts of a frame to remove people or objects from a video. We describe a computationally ecient technique for detecting this form of tampering.
... In 2007, Wang et al. [20] discovered the region duplication and frame duplication in the digital videos in which a region of frame and a sequence of frame have copied and moved at other locations. In 2009, Wang et al. [21] found the double quantization while different quantization scale factors are used to compress the digital video two times. Su et al. [22] detected the frame deletion in the digital videos. ...
... In this technique, different quantization scales are used for the first and second compression. Moreover, Wang et al. [21] have analyzed the histogram of double quantized DCT coefficients of each macroblock. A detection function was defined in this technique using experimentally selected threshold of 0.1. ...
... Wang et al. [20] √ × × √ × × × × × Hsu et al. [12] √ × × × × × × × × Wang et al. [21] √ × × × × × × × × Su et al. [22] × × √ × × × × × × Kobayashi et al. [23] √ × × × × × × × × Su et al. [24] × × × × × × × × √ Xu et al. [25] × × × × × × × × √ Lin et al. [26] × × × √ × × × × × Subramanyam et al. [28] √ × × × × × × × × Bestagini et al. [90] √ × × × × × × × × Liao et al. [29] × × × √ × × × × × Karthikasini et al. [30] √ × × × × × × × × Pandey et al. [19] √ × × × × × × × × Wu et al. [31] × × × × √ × × × × Zhang et al. [32] × √ × × × × × × × Wang et al. [33] × √ × × × × × × × Gironi et al. [34] × √ × × × × × × × Anshida et al. [6] √ × × × × × × × × Bidokhti et al. [35] √ × × × × × × × × Zheng et al. [36] × √ × × × × × × × Singh et al. [37] × × × √ × × × × × Su et al. [93] √ × × × × × × × × Bagiwa et al. [38] × × × × × × × × √ Yang et al. [39] × × × √ × × × × × Aghamaleki et al. [40] × √ × × × × × × × Yu et al. [41] × × √ × × × × × × Ulutas et al. [42] × × × × × × × √ × Bozkurt et al. [43] × × × √ × × × × × Liu et al. [44] × × × √ × × × × × Kingra et al. [45] × × × × × √ × × × Liu et al. [46] × × × × × × × × √ Yao Y et al. [47] √ × × × × × × × × Long et al. [48] × × × √ × × × × × Ulutas et al. [49] × √ × × × × × × × Li et al. [50] × √ × × × × × × × Zhao et al. [51] × × × × × √ × × × Su et al. [53] √ × × × × × × × × Long et al. [54] × √ × × × × × × × Singh et al. [56] √ × × √ × × × × × Saddique et al. [57] √ × × × × × × × × Bakas J et al. [58] × × √ × × × × × × Kharat et al. [59] × × × √ × × × × × Fadl et al. [60] × × × × × × √ × × Fadl et al. [61] × × × × × √ × × × Kaur et al. [78] √ × × × × × × × × Ren et al. [62] × × × √ × × × × × ...
In the real-world scenario, the digital videos are captured to keep in records the life related memories, security purpose, truthful evidence etc. Due to rise in multimedia technology, the digital videos are mostly posted over social websites and applications such as Facebook, Instagram, What’s App, YouTube, etc. through the internet in daily routine. The contents of these social digital videos are manipulated by easily available editing tools and software. Therefore, the security of digital videos over social media are the most important requirement of people. This paper presents a systematic survey on Copy-Move forgery and its detection with several passive techniques in the digital videos. At first, Copy-Move forgery is introduced with its various types in this paper. Then, a survey is provided on the existing passive techniques for detecting Copy-Move forgery in digital videos. Here, brief review of techniques is also presented in the tabular forms with their used features, datasets, parameters, type of forgery and their limitations. Furthermore, this survey provides the parameters and datasets in detail which are used for the evaluation of existing techniques. Besides, the detecting tools for Copy-Move video forgery and its future directions are detailed in this survey. This paper also provides various new challenges for automatic detection of Copy-Move video forgery in the realm of deep learning.
... However, few algorithms have been suggested for detecting spatial tampering in digital video. The Wang and Farid (2007) are the first those who addressed video tampering detection problem. They have proposed a method in which each frame is divided into overlapping blocks of size 16 by 16. ...
... The correlation coefficient of each block is calculated and the blocks whose correlation is above specified threshold are considered as candidate of duplicated region. However, the detection accuracy is very low for small forged region [18]. The Subramanyam and Emmanuel (2012) have used HOG features to detect forgery, in which first the frames are allocated into suitable block size and HOG descriptors of each block are generated. ...
... The frame number and details of the attack are as listed in Table 3. The Figure 6 shows, comparison of the results for proposed method in terms of detection accuracy with respect to existing methods reported in [18] and [20]. From the graph it is clear that the detection accuracy of the proposed method is high (99.5%) ...
This paper presents passive blind forensic scheme to detect spatial tampering in MPEG-4 (Moving Picture Experts Group-4) digital video. In spatial tampering, small region of frame is copied and pasted at some other location in same frame. A proposed algorithm uses SIFT (Scale Invariant Feature Transform) and RANSAC (Random Sample Consensus) to detect the tampering. In this local features from each frame are extracted using SIFT and those features are matched to identify forged area. At the end RANSAC homography is used to remove the false matching to increase the detection accuracy. The proposed method performance is measured with respect to detection accuracy and computational time and verified on compressed and uncompressed videos. To create test data various geometric alterations used in forgery such as scaling, rotation are considered. The simulation results proves that the proposed method finds the forged area efficiently for all the above mentioned cases with average detection accuracy of 99.5%. The algorithm is tested for various compression rates to check its robustness. The detection accuracy of the algorithm increases as the compression rate increases. The performance of the proposed algorithm is compared with two other methods reported in literature which shows that the proposed scheme has higher detection accuracy compared to other methods. The average computational time observed is 0.56 seconds.
... Wang W. and Farid H. [10] introduced the first forgery detection algorithm for interframe duplication forgery in digital videos based on spatial and temporal correlation coefficients between frames. However, this method failed to find inter-frame duplication forgeries when highly similar frames were used for frame duplication and also when postprocessing attacks (e.g., noise addition) were applied on the duplicated frames. ...
... Their algorithm was also unable to detect frame duplication in videos taken by a stationary fixed camera and also when the duplication forgery occurred in a different order [14]. Yang et al. [15] improved the forgery detection method in [10] by proposing a two-stage similarity analysis approach. First, the features from each frame in the tampered video were extracted via singular value decomposition and the distance similarity measure was computed between the extracted features by using Euclidean distance. ...
... To evaluate the performance of the proposed approach, we compare the results of the proposed approach with the results of Wang and Farid [10], Lin et al. [11], Li and Huang [12], Singh et al. [13], Yang et al. [15], Fadl et al. [16], Shelke and Kasana [17], and Bozkurt et al. [18]. The measures of Precision, Recall, and F1 score rates are computed for all the tampered videos in the two established datasets, for the proposed method, and for comparing the other SOTA techniques. ...
Frame duplication forgery is the most common inter-frame video forgery type to alter the contents of digital video sequences. It can be used for removing or duplicating some events within the same video sequences. Most of the existing frame duplication forgery detection methods fail to detect highly similar frames in the surveillance videos. In this paper, we propose a frame duplication forgery detection method based on textural feature analysis of video frames for digital video sequences. Firstly, we compute the single-level 2-D wavelet decomposition for each frame in the forged video sequences. Secondly, textural features of each frame are extracted using the Gray Level of the Co-Occurrence Matrix (GLCM). Four second-order statistical descriptors, Contrast, Correlation, Energy, and Homogeneity, are computed for the extracted textural features of GLCM. Furthermore, we calculate four statistical features from each frame (standard deviation, entropy, Root-Mean-Square RMS, and variance). Finally, the combination of GLCM’s parameters and the other statistical features are then used to detect and localize the duplicated frames in the video sequences using the correlation between features. Experimental results demonstrate that the proposed approach outperforms other state-of-the-art (SOTA) methods in terms of Precision, Recall, and F1Score rates. Furthermore, the use of statistical features combined with GLCM features improves the performance of frame duplication forgery detection.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
In the previous chapters, multimedia forensics techniques based on the analysis of the data stream, i.e., the audio-visual signal, aimed at detecting artifacts and inconsistencies in the (statistics of the) content were presented. Recent research highlighted that useful forensic traces are also left in the file structure, thus offering the opportunity to understand a file’s life-cycle without looking at the content itself. This chapter is then devoted to the description of the main forensic methods for the analysis of image and video file formats.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Every camera model acquires images in a slightly different way. This may be due to differences in lenses and sensors. Alternatively, it may be due to the way each vendor applies characteristic image processing operations, from white balancing to compression.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Since the advent of smartphones, photography is increasingly being done with small, portable, multi-function devices. Relative to the purpose-built cameras that dominated previous eras, smartphone cameras must overcome challenges related to their small form factor. Smartphone cameras have small apertures that produce a wide depth of field, small sensors with rolling shutters that lead to motion artifacts, and small form factors which lead to more camera shake during exposure. Along with these challenges, smartphone cameras have the advantage of tight integration with additional sensors and the availability of significant computational resources. For these reasons, the field of computational imaging has advanced significantly in recent years, with academic groups and researchers from smartphone manufacturers helping these devices become more capable replacements for purpose-built cameras.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Chapters 3 and 4 have shown that PRNU is a very effective solution for source camera verification, i.e., linking a visual object to its source camera Lukas et al. (2006). However, in order to determine the source camera, there has to be a suspect camera in the possession of a forensics analyst which is often not the case.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Multimedia forensics is a societally important and technically challenging research area that will need significant effort for the foreseeable future. While the research community is growing and work like that in this book demonstrates significant progress, many challenges remain.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Empiricism is the notion that knowledge originates from sensory experience. Implicit in this statement is the idea that we can trust our senses. But in today’s world, much of the human experience is mediated through digital technologies. Our sensory experiences can no longer be trusted a priori. The evidence before us—what we see and hear and read—is, more often than not, manipulated.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
One particular disconcerting form of disinformation are the impersonating audios/videos backed by advanced AI technologies, in particular, deep neural networks (DNNs). These media forgeries are commonly known as the DeepFakes. The AI-based tools are making it easier and faster than ever to create compelling fakes that are challenging to spot. While there are interesting and creative applications of this technology, it can be weaponized to cause negative consequences. In this chapter, we survey the state-of-the-art DeepFake detection methods.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
We describe the threats posed by adversarial examples in an image forensic context, highlighting the differences and similarities with respect to other application domains. Particular attention is paid to study the transferability of adversarial examples from a source to a target network and to the creation of attacks suitable to be applied in the physical domain. We also describe some possible countermeasures against adversarial examples and discuss their effectiveness. All the concepts described in the chapter are exemplified with results obtained in some selected image forensics scenarios.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Photo-response non-uniformity (PRNU) is an intrinsic characteristic of a digital imaging sensor, which manifests as a unique and permanent pattern introduced to all media captured by the sensor. The PRNU of a sensor has been proven to be a viable identifier for source attribution and has been successfully utilized for identification and verification of the source of digital media.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
The literature of multimedia forensics is mainly dedicated to the analysis of single assets (such as sole image or video files), aiming at individually assessing their authenticity. Different from this, image provenance analysis is devoted to the joint examination of multiple assets, intending to ascertain their history of edits, by evaluating pairwise relationships. Each relationship, thus, expresses the probability of one asset giving rise to the other, through either global or local operations, such as data compression, resizing, color-space modifications, content blurring, and content splicing. The principled combination of these relationships unveils the provenance of the assets, also constituting an important forensic tool for authenticity verification. This chapter introduces the problem of provenance analysis, discussing its importance and delving into the state-of-the-art techniques to solve it.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
The rise of deep learning has led to rapid advances in multimedia forensics. Algorithms based on deep neural networks are able to automatically learn forensic traces, detect complex forgeries, and localize falsified content with increasingly greater accuracy. At the same time, deep learning has expanded the capabilities of anti-forensic attackers. New anti-forensic attacks have emerged, including those discussed in Chap. 10.1007/978-981-16-7621-5_14 based on adversarial examples, and those based on generative adversarial networks (GANs). In this chapter, we discuss the emerging threat posed by GAN-based anti-forensic attacks. GANs are a powerful machine learning framework that can be used to create realistic, but completely synthetic data. Researchers have recently shown that anti-forensic attacks can be built by using GANs to create synthetic forensic traces. While only a small number of GAN-based anti-forensic attacks currently exist, results show these early attacks are both effective at fooling forensic algorithms and introduce very little distortion into attacked images.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Physics-based methods anchor the forensic analysis in physical laws of image and video formation. The analysis is typically based on simplifying assumptions to make the forensic analysis tractable. In scenes that satisfy such assumptions, different types of forensic analysis can be performed. The two most widely used applications are the detection of content repurposing and content splicing. Physics-based methods expose such cases with assumptions about the interaction of light and objects, and about the geometric mapping of light and objects onto the image sensor.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Videos can be manipulated in a number of different ways, including object addition or removal, deep fake videos, temporal removal or duplication of parts of the video, etc. In this chapter, we provide an overview of the previous work related to video frame deletion and duplication and dive into the details of two deep-learning-based approaches for detecting and localizing frame deletion (Chengjiang et al. 2017) and duplication (Chengjiang et al. 2019) manipulations.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Images and videos are by now a dominant part of the information flowing on the Internet and the preferred communication means for younger generations. Besides providing information, they elicit emotional responses, much stronger than text does. It is probably for these reasons that the advent of AI-powered deepfakes, realistic and relatively easy to generate, has raised great concern among governments and ordinary people alike.
... Recently, proposed an SVM-based method to classify tampered or non-tampered videos. In this work, we explore the authentication (Valentina et al. 2012;Wang and Farid 2007) of the scene or camera to determine if a video has one or more frame drops without a reference or original video. We expect such authentication is able to explore underlying spatio-temporal relationships across the video so that it is robust to digital-level attacks and conveys a consistency indicator across the frame sequences. ...
... It is very important to develop robust video forensic techniques, to catch videos with increasing sophisticated forgeries. Video forensics techniques Wang and Farid 2007) aim to extract and exploit features from videos that can distinguish forgeries from original, authentic videos. Like other areas in information security, the sophistication of attacks and forgeries continue to increase for images and videos, requiring a continued improvement in forensic techniques. ...
... In recent years, multiple digital video forgery detection approaches have been employed to solve this challenging problem. Wang and Farid (2007) proposed a frame duplication detection algorithm which takes the correlation coefficient as a measure of similarity. However, such an algorithm results in a heavy computational load due to a large number of correlation calculations. ...
Every imaging sensor introduces a certain amount of noise to the images it captures—slight fluctuations in the intensity of individual pixels even when the sensor plane was lit absolutely homogeneously. One of the breakthrough discoveries in multimedia forensics is that photo-response non-uniformity (PRNU), a multiplicative noise component caused by inevitable variations in the manufacturing process of sensor elements, is essentially a sensor fingerprint that can be estimated from and detected in arbitrary images. This chapter reviews the rich body of literature on camera identification from sensor noise fingerprints with an emphasis on still images from digital cameras and the evolving challenges in this domain.
... Traditional watermark-based approaches require dedicated modules on security cameras for video integrity preservation, while not all camera manufacturers support such modules [6]. Alternatively, many video forensics approaches that exploit video statistic characteristics [7], [8], [9], [10] are developed to detect tampered frames and further localize forgery traces. However, these approaches rely on spatial-temporal analysis on relatively long video clips, which are ill-suited for live video streams in time-critical surveillance systems. ...
... Video Forgery Detection. Passive video forensics approaches leverage video statistic characteristics to discover forgery traces [6], [7], [8], [9], [10]. However, such approaches generally rely on relatively long video clips and are illsuited for live video feeds. ...
... K m = k m j,n ∈ R 2 : for j ∈ {1, · · · , J} , n ∈ 1, · · · , N j , (10) where N j denotes the number of candidates of j-th keypoint and k m j,n is the location of n-th candidate of j-th keypoint. The second step is to associate the suspicious keypoints K m to form abnormal objects using PAFs L m I and L m R . ...
The cybersecurity breaches expose surveillance video streams to forgery attacks, under which authentic streams are falsified to hide unauthorized activities. Traditional video forensics approaches can localize forgery traces using spatial-temporal analysis on relatively long video clips, while falling short in real-time forgery detection. The recent work correlates time-series camera and wireless signals to detect looped videos but cannot realize fine-grained forgery localization. To overcome these limitations, we propose Secure-Pose, which exploits the pervasive coexistence of surveillance and Wi-Fi infrastructures to defend against video forgery attacks in a real-time and fine-grained manner. We observe that coexisting camera and Wi-Fi signals convey common human semantic information and forgery attacks on video streams will decouple such information correspondence. Particularly, retrievable human pose features are first extracted from concurrent video and Wi-Fi channel state information (CSI) streams. Then, a lightweight detection network is developed to accurately discover forgery attacks and an efficient localization algorithm is devised to seamlessly track forgery traces in video streams. We implement Secure-Pose using one Logitech camera and two Intel 5300 NICs and evaluate it in different environments. Secure-Pose achieves a high detection accuracy of 98.7% and localizes abnormal objects under playback and tampering attacks.
... Various video tampering detection techniques have been proposed in the literature to detect inter-frame tampering; these techniques are based on extracting manual features, such as statistical features [10][11][12], pixel and texture characteristics [13][14][15], motion residual, and optical flow [16][17][18], and a few are based on deep learning [8,[19][20][21]. The manual features are sensitive to post-processing operations like blurring, brightness, noise, and compression. ...
... All these similarity detection techniques access the stored surveillance footage from the stored database; resulting in significant computation time to process each video frame. Features such as correlation [6,10,13], optical flow [17,42,43], prediction residual [22,44,45], bag-of-words (BoW) model [23], standard deviation of residual frames [40], motion vector and motion residual [36,46], and noise residue [32] have been utilized in the literature to identify inconsistencies introduced by inserted, deleted, or duplicated frames in videos. Some methods, like those of Wang et al. [11] and Huang et al. [47], use statistical features like the consistency of correlation coefficients of gray values (CCCoGV) and triangular polarity feature classification (TPFC) to detect inter-frame tampering. ...
Inter-frame tampering in surveillance videos undermines the integrity of video evidence, potentially influencing law enforcement investigations and court decisions. This type of tampering is the most common tampering method, often imperceptible to the human eye. Until now, various algorithms have been proposed to identify such tampering, based on handcrafted features. Automatic detection, localization, and determine the tampering type, while maintaining accuracy and processing speed, is still a challenge. We propose a novel method for detecting inter-frame tampering by exploiting a 2D convolution neural network (2D-CNN) of spatiotemporal information and fusion for deep automatic feature extraction, employing an autoencoder to significantly reduce the computational overhead by reducing the dimensionality of the feature’s space; analyzing long-range dependencies within video frames using long short-term memory (LSTM) and gated recurrent units (GRU), which helps to detect tampering traces; and finally, adding a fully connected layer (FC), with softmax activation for classification. The structural similarity index measure (SSIM) is utilized to localize tampering. We perform extensive experiments on datasets, comprised of challenging videos with different complexity levels. The results demonstrate that the proposed method can identify and pinpoint tampering regions with more than 90% accuracy, irrespective of video frame rates, video formats, number of tampering frames, and the compression quality factor.
... Several passive (blind) stateof-the-art methods are listed in survey papers [15][16][17][18][19][20][21][22] that exploit either hand-crafted [23][24][25] or deep-learned footprints [26,27] to identify video attacks. In the literature, some approaches investigate frame-level video forgeries such as frame mirroring [3], upscale-crop [4], region-duplication [5], video splicing [6], replayed video [7], video face spoofing [8], video re-capture [9], video copy [10], video phylogeny [11], green screening [5] and deepfake [12] etc. Some of the techniques examine the traces of interframe level video tampering such as frame deletion [13], frame insertion, frame shuffling, frame Cloning [14], interpolation [15] etc. ...
... Several passive (blind) stateof-the-art methods are listed in survey papers [15][16][17][18][19][20][21][22] that exploit either hand-crafted [23][24][25] or deep-learned footprints [26,27] to identify video attacks. In the literature, some approaches investigate frame-level video forgeries such as frame mirroring [3], upscale-crop [4], region-duplication [5], video splicing [6], replayed video [7], video face spoofing [8], video re-capture [9], video copy [10], video phylogeny [11], green screening [5] and deepfake [12] etc. Some of the techniques examine the traces of interframe level video tampering such as frame deletion [13], frame insertion, frame shuffling, frame Cloning [14], interpolation [15] etc. ...
Sanctity and integrity of digital videos are crucial for the diverse real-world applications. It has significant social and legal implications. The technological advancements are posing new challenges as the video processing software that are typically designed to enhance the visual content, can adversely spawn unauthentic and malicious data that can be potentially hazardous. Robust algorithms are therefore needed to counter the deleterious effects. In this paper, we propose a passive-blind approach to detect and localize multiple kinds of inter-frame forgeries in digital videos like frame insertion, deletion and duplication. The forensic artefacts are designed based on correlation inconsistencies between the histogram-similarity patterns of the adjacent texture-feature encoded video frames. For the empirical evaluation, the algorithm uses texture features such as Histogram of Oriented Gradients (HoG), uniform and rotation invariant Local Binary Pattern (LBP). A customized dataset of 1370 tampered videos is created using the benchmark SULFA dataset due to lack of standard video dataset with inter-frame forgeries. A supervised SVM classifier is trained to detect video tampering where extensive analysis based on different histogram-similarity metrics is carried out with the proposed approach that exhibits an overall accuracy 99%. Further, the proposed method localizes the position of tampered frames in the video. It highlights forged frames using Chebyshev's inequality in case of frame insertion and deletion attacks. A comparative analysis with state-of-the-art methods is also presented that exhibits good efficacy of the proposed approach.
... The remaining signal is compressed employing spatial redundancy Fig. 9 Group of pictures (GOPs) representing I-frame, P-frame and B-frame reduction (DCT), termed prediction error. The study by Wang et al. [90] for the MPEG compressed videos states that an approximation of motion between the frames has been obtained as a measure of error, which could be a possible way of inspecting the forgery. Further understanding by Wang et al. [90] says that motion errors have a significant value between each MPEG file frame, which could help predict motion error detection. ...
... The study by Wang et al. [90] for the MPEG compressed videos states that an approximation of motion between the frames has been obtained as a measure of error, which could be a possible way of inspecting the forgery. Further understanding by Wang et al. [90] says that motion errors have a significant value between each MPEG file frame, which could help predict motion error detection. The addition or deletion of frames within this MPEG video sequence produces a difference in the measured motion error. ...
The massive dependence on digital interaction via multimedia tools like audio, images, and videos is leading to an era that will be digitized in almost every aspect. This credible development in multimedia technology with the easy availability of advanced cheap editing or tampering tools/techniques attracts the researcher’s community to ensure the multimedia content’s authenticity. This paper reviews video forensics, the statistical ways manipulations have been deployed on the data, and detection approaches for tracing the tampering attacks and challenges. The study gives better clarity for understanding the demand of direct relation for video content to be shared on social media platforms as the legal evidence towards the growth of tampering detection tools/technology. This paper gives a detailed review of the existing forgery attacks and passive forgery detection techniques. The video forensics algorithm is studied extensively, along with its limitations for varying datasets in an ordered manner. This discussion explores current active challenges in video forensics, raising scientific investigation and developing reliable multimedia forensics approaches.
... Frame insertion, deletion and duplication are temporal forgeries in which frames are; removed from the video to conceal (deletion), inserted into the videos to convey or delay (insertion), and/or copied and pasted into the video to delay (duplication) by forgers (Anas et al., 2022). To detect frame forgeries, numerous techniques have been employed by researchers Wang & Farid (2006, 2007and 2009 ;Bestagini et al., (2011);Feng et al., (2011); Lin and Chan (2012); Yang et al., (2016); Zhao et al., (2018); and Anas et al., (2022). ...
... Researchers had proposed several schemes to detect inter-frame forgeries. Wang and Farid (2006, 2007and 2009), Bestagini et al., (2011, Lin andChan (2012), Lou et al., (2014), Yang et al., (2016), Zhao et al., (2018), Anas et al., (2022) among others. The proposed schemes can be broadly categorized in to single aspect, dual aspects and multiple aspects (Sitara, & Mehtre 2016). ...
Surveillance videos provide security and increases work efficiency in places of work and homes. as the most acceptable form of evidence, surveillance videos are now tampered to hide actions or convey wrong information. Researchers have proposed ways to mitigate the effect of activities of the attackers through checking the authenticity of the video. The proposed schemes suffer performance degradation in the presence of scene changes. Recently a scheme that addresses the effects of scene change on inter-frame forgery detection was developed where it detects scene changes and divides multiple scenes in to shots. The scheme improves the overall performance of the inter-frame forgery detection at the expense of high average computational time. In this research, a video scene change aware forgery detection scheme is proposed to mitigate the effect of scene change on inter-frame forgery detection with low average computational time. The proposed scheme utilizes the luminance level within frame region which is a more efficient feature to detect scene change. The experimental results show that the scheme has 57% decreases in computational average time and increased in accuracy to 99.03%.
... We have described some methods that fall under this category and a summary of these methods is depicted in Table 8. We can find such attempts to detect forged video in the works reported in [150,151]. Wang and Farid [150] have shown larger motion errors due to double compression by taking double MPEG compression while they have proved it using double compression in their work [151]. In both works, the distribution of DCT coefficients of each macro-block in I-frames is calculated and compared with the actual scenario to measure the amount of contrast deficiency. ...
... We can find such attempts to detect forged video in the works reported in [150,151]. Wang and Farid [150] have shown larger motion errors due to double compression by taking double MPEG compression while they have proved it using double compression in their work [151]. In both works, the distribution of DCT coefficients of each macro-block in I-frames is calculated and compared with the actual scenario to measure the amount of contrast deficiency. ...
Video plays a key role in carrying authenticity, especially in the surveillance system, medical field, court evidence, journalism, and social media among others. However, nowadays the trust in videos is decreasing day by day due to the forgery of the videos made by easily accessible video editing tools. Hence, a thrust for finding a robust solution to the problem of video forgery detection arises. As a result, researchers around the world are indulging themselves to come up with various methods for the said problem. In this article, we have comprehensively discussed many such initiatives made by researchers across the globe, keeping the focus on recent trends. In addition to this, we have also covered a wide range of forgery detection techniques that follow either an active or a passive approach, while the state-of-the-art surveys made so far on this research topic include only a few specific cases. In this article, we have described some recent technologies that are used in video forging, made a summary of the performances (provided categorically) of all the techniques discussed here, and briefed the available datasets. Finally, we have concluded this survey by clearly mentioning some future directions of the video forgery detection research based on a thorough review of existing techniques.
... Video recordings used for temporal correlation of the live events are primarily targeted using frame shuffling or duplication attacks [30]. The perception of live events is affected, which disables the effectiveness of live monitoring [31]. ...
... Analyzing the compression algorithms used by H.264 coding, the presence of any recompression artifacts is used to identify frame manipulations [51]. The spatial and temporal correlation is used to create motion vector features [30,52]. The de-synchronization caused by removing a group of frames introduces spikes in the Fourier transform of the motion vectors. ...
Intelligent surveillance systems play an essential role in modern smart cities to enable situational awareness. As part of the critical infrastructure, surveillance systems are often targeted by attackers aiming to compromise the security and safety of smart cities. Manipulating the audio or video channels could create a false perception of captured events and bypass detection. This chapter presents an overview of the attack vectors designed to compromise intelligent surveillance systems and discusses existing detection techniques. With advanced machine learning (ML) models and computing resources, both attack vectors and detection techniques have evolved to use ML-based techniques more effectively, resulting in non-equilibrium dynamics. The current detection techniques vary from training a neural network to detect forgery artifacts to use the intrinsic and extrinsic environmental fingerprints for any manipulations. Therefore, studying the effectiveness of different detection techniques and their reliability against the defined attack vectors is a priority to secure the system and create a plan of action against potential threats.
... Secondly, message integrity means that the sensed data of the terminal device is not changed during transmission. For message integrity, Forgery Attack can tamper the sensing data of IoT devices, so that the cloud server or users can obtain false physical world information [36]. Therefore, in order to protect the security of IoT terminal devices, it is very important to authenticate the identity authenticity and message integrity of IoT devices. ...
As an emerging wireless communication technology, narrowband Internet of Things is widely used in various scenarios, but it also brings new security problems, especially the group authentication between terminal devices and servers. In narrowband Internet of Things, terminal devices are often deployed in unattended environments, which are vulnerable to various attacks and controlled by adversaries, resulting in identity information leakage. Secondly, a large number of terminal devices send authentication request information to the server at the same time, which is easy to cause network signaling congestion, and the centralized calculation mode of the server will cause a large delay in the authentication process. To solve these problems, we propose an edge-assisted group authentication scheme for narrowband Internet of Things. Firstly, a lightweight encryption algorithm LCHAOSAES was used to encrypt the terminal device identity to prevent the terminal device identity information leakage. Then, the edge computing method was used to add an edge network layer to the NB-IoT system architecture to assist in completing the group authentication of massive terminal devices, and the edge assistance could realize decentralization and reduce the load of centralized processing of the server. The Burrows Abadi Needham logic verification, informal security analysis and performance analysis show that the proposed scheme is robust under malicious attacks such as replay attack, man-in-the-middle attack and insider attack, and has less signaling overhead and delay.
... The main goals of passive forensics are given below. A Farid research group is utilized as a sequence of detecting techniques, like discovering forgeries of video based on MPEG double compression [32], revealing digital forgeries [33], proposing double quantization (Wang and Farid 2009), and determining video frame imitation based on the gray level vector [34]. In [9], a technique is devised to discover forged regions from video using noise features and inconsistencies that occurred because of forged regions from various videos. ...
Nowadays, a surveillance camera is used extensively to provide security. With easier accessibility of tools like video editing, it became simple to destroy evidence. Different detection techniques are in practice howsoever due to many reasons it is confined. Hence, the competitive swarm sunflower optimization algorithm (CSSFOA)-based random multimodal deep learning (RMDL) is proposed for discovering forgeries. CSSFOA is the integration of competitive swarm optimizer and sunflower optimization. Here, extraction of the keyframe is carried out utilizing discrete cosine transform and Tanimoto distance. Also, by using the Viola-Jones algorithm, face detection is performed considering the light coefficients and face image coefficients by extracting the local optimal oriented pattern. The deep composite images are obtained utilizing RMDL. RMDL is trained utilizing developed CSSFOA. The proposed CSSFOA-based RMDL shows superior performance with maximum accuracy of 96.6%, true positive rate of 95.0%, and true negative rate of 95.5%.
... In case of crop approach, a portion of a video gets cropped in order to conceal the fact. Wang and Farid (2009) came up with a double quantization-based approach that would detect the deepfaked video when different videos of varying resolutions are conjoined. The methodology lags in accuracy and robustness. ...
With the help of advancements in connected technologies, social media and networking have made a wide open platform to share information via audio, video, text, etc. Due to the invention of smartphones, video contents are being manipulated day-by-day. Videos contain sensitive or personal information which are forged for one’s own self pleasures or threatening for money. Video falsification identification plays a most prominent role in case of digital forensics. This paper aims to provide a comprehensive survey on various problems in video falsification, deep learning models utilised for detecting the forgery. This survey provides a deep understanding of various algorithms implemented by various authors and their advantages, limitations thereby providing an insight for future researchers.
... After applying the Fourier transformation to videos' encoding motion vectors, lots of spikes appeared in the frequency histograms. Furthermore, the authors extended the work [5] to explore the abnormal change of the quantized coefficients before and after double compression. It proved that the quantized coefficients would be sparser after frame deletion. ...
Frame deletion detection has played an essential role in digital forensics. The existing literature suggests that detection work is accomplished by appropriately revealing continuity-attenuation traces of video contents caused by frame deletion in the temporal direction. In this work, we propose a new network architecture, one module of which is exploited as a detector to capture the spatiotemporal features with continuity-attenuation in the forgery videos. First, through a study on the statistical characteristics of the motion trajectory of moving objects, we reveal a new continuity-attenuation trace, based on which the inter-frame residual feature is selected as the basis for continuity-attenuation tracking. Second, to capture the continuity-attenuation phenomenon, we design a network framework consisting of three components: a detector module, a reference module, and a decision module. Three modules work cooperatively under the contrast learning strategy to make the detector more sensitive to capture the forensic trace. The experiment results show that the detection rate can reach 93.85%, indicating the effectiveness of our proposed deep learning-based detection strategy.
... No Method in [24] 92.60 Mirror Method in [7] 70.00 No Method in [38] 90.80 No ...
... Patch similarity is determined by comparing the spatial correlation of the frames in the fake and real clips using a blockbased technique. Wang et al. [216] explored this similarity perception. ...
Thousands of videos are posted on websites and social media every day, including Twitter, Facebook, WhatsApp, Instagram, and YouTube. Newspapers, law enforcement publications, criminal investigations, surveillance systems, Banking, the museum, the military, imaging in medicine, insurance claims, and consumer photography are just a few examples of places where important visual data may be obtained. Thus, the emergence of powerful processing tools that can be easily made available online poses a huge threat to the authenticity of videos. Therefore, it’s vital to distinguish between true and fake data. Digital video forgery detection techniques are used to validate and check the realness of digital video content. Deep learning algorithms lately sparked a lot of interest in the field of digital forensics, such as Recurrent Neural Networks (RNN), Deep Convolutional Neural Networks (DCNN), and Adaptive Neural Networks (ANN). In this paper, we give a soft taxonomy as well as a thorough overview of recent research on multimedia falsification detection systems. First, the basic knowledge needed to comprehend video forgery is provided. Then, a summary of active and passive video manipulation detection approaches is provided. Anti-forensics, compression video methods, datasets required for video forensics, and challenges of video detection approaches are also addressed. Following that, we presented an overview of deepfake, and the datasets required for detection were also provided. Also, helpful software packages and forensics tools for video detection are covered. In addition, this paper provides an overview of video analysis tools that are used in video forensic applications. Finally, we highlight research difficulties as well as interesting research avenues. In short, this survey provides detailed information and a broader investigation to extract data and detect fraud video contents under one umbrella.
... Thanks to the use of DCT as the basis of the MPEG-4 compression format, the effect of the normal distribution of the DCT coefficients of the frames in the extracted feature-based techniques based on DCT has been studied [34,48,49]. Wang and Farid calculated the Euclidean distance according to the Gaussian distribution by applying the DCT to each macroblock obtained from the initial (Intra) frames [44]. The disadvantage of the method is that the threshold value is used to compare and process only on I frames. ...
Videos are one of the most substantial evidence that can be used to detect incidents. However, videos can be altered easily using current technologies. Alterations can be made for malicious purposes. Therefore, it is essential to determine the integrity of the videos that will be used as evidence or to give people the right idea. Alterations on the videos that have not previously added control data are within the scope of passive fraud detection. This study proposes an effective solution for detecting frame duplication attacks, which is one of the passive forgery types. The study is based on the visualization of feature vectors. A binary image is created with the feature matrix using feature vectors. Thus, a representative approach to the problem is presented. The forged frame-group template is obtained by processing the binary image, and then a search is done using this template. The proposed method provides solutions for both uncompressed and compressed videos. The algorithm’s durability against compression has been tested by evaluating MPEG4 and H264 coded videos. A blurring attack can also be applied to the altered videos to hide the forgery. The results show that it is resistant to blurring attacks. Another factor that complicates fraud detection is the location of the forgery. The algorithm can detect forgery at the beginning or end of the video. The source of the forged frames can also be detected in the study. Experimental results show that the algorithm is resistant to compression, fast, and has a high accuracy rate.
... To detect a double-compressed video, Wang [17] proposed a method using specific static and temporal statistical perturbations of a double-compressed Moving Picture Experts Group (MPEG) video. The double quantization effect is used in [18] to detect double MPEG compression. Markov-based features are proposed to detect double compression artifacts in [19]. ...
In digital forensics, video becomes important evidence in an accident or a crime. However, video editing programs are easily available in the market, and even non-experts can delete or modify a section of an evidence video that contains adverse evidence. The tampered video is compressed again and stored. Therefore, detecting a double-compressed video is one of the important methods in the field of digital video tampering detection. In this paper, we present a new approach to detecting a double-compressed video using the proposed descriptors of video encoders. The implementation of real-time video encoders is so complex that manufacturers should develop hardware video encoders considering a trade-off between complexity and performance. According to our observation, hardware video encoders practically do not use all possible encoding modes defined in the video coding standard but only a subset of the encoding modes. The proposed method defines this subset of encoding modes as the descriptor of the video encoder. If a video is double-compressed, the descriptor of the double-compressed video is changed to the descriptor of the video encoder used for double-compression. Therefore, the proposed method detects the double-compressed video by checking whether the descriptor of the test video is changed or not. In our experiments, we show descriptors of various H.264 and High-Efficiency Video Coding (HEVC) video encoders and demonstrate that our proposed method successfully detects double-compressed videos in most cases.
... The very first method [192] to detect frame insertion/removal forgery in a video sequence analyzed double compression in MPEG video sequences by detecting periodicity in DCT coefficient distribution of I-frames, and motion error generated by P-frames. Although the author did not formulate any quantitative results in the paper but claimed that the technique's performance boosts if number of frames deleted/inserted are multiples of 3. Same group of researchers, then, extended the technique in [193] where they also detected composite videos. Double compression was detected by analyzing Gaussian distribution of DCT coefficients. ...
Digital content, particularly the digital videos recorded at specific angle, though, provides a truthful picture of reality but the widespread proliferation of easy-to-use content editing softwares doubt about its authenticity. Recently, Artificial Intelligence (AI) based content altering mechanism, known as deepfake, became popular on social media platforms, wherein any person can be able to purport the behaviour of another person in a video who is actually not there. Depending on the type of manipulation performed, different types of deepfakes are described in this paper. Moreover, rely on digital content for trustworthy evidence as well as to avoid spread of misinformation, integrity and authenticity of digital content has-been of utmost concerns. This paper aims to present a survey of the state-of-art video integrity verification techniques with special emphasis on emerging deepfake video detection approaches. Seeing the advancement in creation of more realistic deepfake videos, this review facilitates the development of more generalized methods with a thorough discussion on different research trends in the wake of deepfake detection.
... Baghel [2] has defined a method to distinguish correlated frames in a video. Wang [26] explains digital forgeries in a video by detecting duplication, which includes insertion forgery. Subramanyam [25] finds video forgery detection using HOG features. ...
It has become easy to alter and tamper the video content flawlessly by using easily available editing software in today's era. Thus, the authenticity of media resources is at high risk. Therefore, one must verify the video to detect the originality of that video content. If the originality and authenticity of the video are compromised, it can change the viewers' perception. This work presents an automatic forgery detection tool that can help to identify the frame insertion type forgery and its location in a video. The deep features are a significant feature in recognizing the forgery and abnormal variations in the video in this work. Based on a parallel CNN model, the proposed method extracts deep features. It also calculates the distance of the correlation coefficient from the deep features, which helps to find the disassociation between the adjacent frames to identify video forgery. The VIFFD and SULFA standard datasets are used to validate the proposed method. It shows that the proposed method is beneficial in differentiating original & insertion type forged videos. The total accuracy of 99.96% achieved in frame-level forgery detection. On video-level forgery detection, 86.5% & 92% accuracy has been achieved in VIFFD & SULFA dataset, respectively. This work also helps to find the inserted frame's location, which benefits in regenerating the original video from the forged video. The proposed method is non-invasive, efficient, robust, and has low time complexity making it suitable for real applications.
... It does not address the insertion and deletion type of forgery. To detect the frame and region duplications in videos, Farid et al. [29] presented a method, which extracts both subsequences from full video and temporal and spatial correlations. This method used, the correlation coefficient as a similarity measure. ...
Video forgery can be defined as the modification of the video contents. The alteration of the video by deletion and modification in the sequence of frames is a trivial task, which has made the authentication and originality detection more important. Frame insertion and deletion are the most common type of video forgery. The proposed method can identify these types of forgery along with its forged location, which makes this unique method. It defines the relationship between the adjacent frames using the correlation coefficient, finds the inter-frame correlation distance between the frames, calculates the minimum distance score, statistical features, and computes upper-bound, lower-bound threshold and sigma coefficient for the identification of forgery location. The proposed method defines insertion and deletion type forgery by using threshold controlled parameters and it is validated on the VIFFD dataset. The proposed method has also identified forgery with 97% accuracy at the frame level and 83% accuracy at the video level. The result analysis shows the superiority of the proposed method over the existing methods. This method is very effective in identifying the forgery type with its frame location.
The extensive availability of software tools and technological advancements have made it possible to alter or modify digital data in today's digital environment. It was fairly simple to properly manipulate video content thanks to the accessibility of inexpensive software products like Adobe Premiere, Magix Vegas, Mokey by Imagineer Systems, and Microsoft Movie Maker. Thanks to these sophisticated video editing programmes, changing the contents of digital videos has become very simple. In order to hide the truth, video editing software is frequently used to copy and paste a section of the frame from one location to another region in the same frame or another frame. One of the most popular techniques for manipulating with videos is region duplication. By examining the spatial and temporal correlations between the frame's pixels, numerous approaches to identify such tampering in digital video sequences have been presented by various researchers. Nevertheless, majority of the algorithms have significant computational cost and low accuracy. In this research, we present a novel region duplication tampering detection method that uses wavelet transform and Euclidean distance to identify intra frame region duplication in a video sequence. The proposed method is assessed using a variety of video sequences where region duplication in frames introduced. Performance of the suggested technique is evaluated based on experimental findings in terms of Precision, Recall, F1, Accuracy, and calculation time. Our experimental results proved that this method is highly suitable for detecting the duplicated regions in the video which is very much useful in forensic applications.
A video becomes forged, if it is altered by changing the information contained within a frame or by changing the original sequencing of frames by deleting some frames or adding some frames in between, referred to as intra-frame forgery and inter-frame forgery respectively. This paper proposes an effective method for inter-frame video forgery detection which is capable of detecting duplication of frames, deletion of frames and also insertion of frames in the video. The method proposed is also capable of locating the forgery. There are many other existing methods which detect video forgery using features such as correlation coefficient between adjacent frames, optical flow, Zernike moment and so on. The proposed method detects forgery in a simple method compared to the existing ones. It consumes less computational power and time. The fact that manipulation done on the video alters the original sequencing of frames, which can be detected by examining the difference in pixel intensities of adjacent frames is made use of by this method. This method separates the frames of the video and uses the difference in pixel intensities of adjacent frames in two different ways to detect forgery. The original sequence of frames in the video follows a smooth pattern of adjacent frame differences, but any change occurring to the sequencing causes spikes. By checking the presence of these spikes, forgery along with the location of forgery can be detected. This method is found to have better accuracy compared to state-of-the-art methods and experimentation is done using the publicly available datasets.
Surveillance cameras are widely used to provide protection and security through online tracking or investigation of stored videos of an incident. Furthermore, footage of recorded videos may be used as strong evidence in the courts of law or insurance companies, but their authenticity cannot be taken for granted. Two common video inter-frame forgery types are frame duplication (FD) and frame insertion (FI). Several techniques exist in the literature to deal with them by analyzing the abnormalities caused by these operations. However, they have limited applicability, poor generalization, and high computational complexity. To tackle these issues, we propose a robust hybrid forensic system based on the idea that FD or FI causes motion inconsistency at the start and end of duplicated/inserted frames. These inconsistencies, when analyzed in an appropriate manner, help reveal the evidence of forgery. The system encompasses two forensic techniques. The first is a novel method based on the texture of motion residual component where a motion residual-based local binary pattern histogram (MR-LBPH) and an SVM classifier with the linear kernel are employed to detect suspected tampered positions. The second component is the sum consistency of optical flow (OF) and standard deviation of MR-LBPH of consecutive frames to remove false positives and precise localization of forgery. By taking the intersection of the frames detected by the two methods, we remove the false positives and get the frames bounding the duplicated/inserted region. The methods were trained and tested on our developed large Video Tampering Evaluation Dataset (VTED) and cross-validated on publicly available datasets. Cross-dataset evaluation yielded detection accuracy above 99.5%, ensuring the proposed method’s generalization; it also precisely locates the locations of tampering. As the public datasets used for cross-dataset validation include videos of different formats and frame rates, it ensures the wide applicability of the method. Moreover, the method is computationally efficient and can be run in a matter of microseconds.
In today’s digital world all kind of enhancement is becoming possible and at the same time the usage of images and videos have been growing day by day in our lives, the enthusiasm to make manipulation of images also increases concurrently. In this study, the most recent technical analysis and observations of various copy-move image and video forgery techniques were carried out. Image splicing, copy-move forgery, and image resampling are the three basic types of image counterfeiting. Copy-Move forgery is commonly used for making tampered photographs. As the forgery of images and video is increasing, it is much essential to develop tools for the detection of such forgeries. This study examines several forms of digital image and video forgeries as well as detection techniques.KeywordsCopy-move attackImage forgeryImage splicingActivePassive
This paper contributes in detecting temporal tampering in the Moving Picture Experts Group-4 (MPEG-4) digital video using passive blind forensic scheme. In temporal tampering attack, one or more frames are copied from video and pasted at some other location in the same video to hide or highlight particular activity. Sometimes the original frames are modified and then pasted at some other location. This tampering activity is carried out so smartly that it is difficult to identify with naked eyes. The proposed forensic scheme uses two-step algorithm to detect the tampering. First, the suspicious frame is detected and then this frame is compared with others frames in the sequence to find the duplicate frame. Comparison of the frames is carried out with the help of Scale Invariant Feature Transform (SIFT) key-points. Random Sample Consensus (RANSAC) is used to take the final decision on the matching of the frames. The simulation results show that the proposed scheme is able to detect and locate the tampered frame with 100% accuracy regardless of how many number of frames of the video sequence are duplicated.
Video forgery detection and localization is one of the most important issue due to the advanced editing software that provides strengthen to tools for manipulating the videos. Object based video tampering destroys the originality of the video. The main aim of the video forensic is to eradicate the forgeries from the original video that are useful in various applications. However, the research on detecting and localizing the object based video forgery with advanced techniques still remains the open and challenging issue. Many of the existing techniques have focused only on detecting the forged video under static background that cannot be applicable for detecting the forgery in tampered video. In addition to this, conventional techniques fail to extract the essential features in order to investigate the depth of the video forgery. Hence, this paper brings a novel technique for detecting and localizing the forged video with multiple features. The steps involved in this research are keyframe extraction, pre-processing, feature extraction and finally detection and localization of forged video. Initially, keyframe extraction uses the Gaussian mixture model (GMM) to extract frames from the forged videos. Then, the pre-processing stage is manipulated to convert the RGB frame into a grayscale image. Multi-features need to be extracted from the pre-processed frames to study the nature of the forged videos. In our proposed study, speeded up robust features (SURF), principal compound analysis histogram oriented gradients (PCA-HOG), model based fast digit feature (MBFDF), correlation of adjacent frames (CAF), the prediction residual gradient (PRG) and optical flow gradient (OFG) features are extracted. The dataset used for the proposed approach is collected from REWIND of about 40 forged and 40 authenticated videos. With the help of the DL approach, video forgery can be detected and localized. Thus, this research mainly focuses on detecting and localization of forged video based on the ResNet152V2 model hybrid with the bidirectional gated recurrent unit (Bi-GRU) to attain maximum accuracy and efficiency. The performance of this approach is finally compared with existing approaches in terms of accuracy, precision, F-measure, sensitivity, specificity, false-negative rate (FNR), false discovery rate (FDR), false-positive rate (FPR), Mathew’s correlation coefficient (MCC) and negative predictive value (NPV). The proposed approach assures the performance of 96.17% accuracy, 96% precision, 96.14% F-measure, 96.58% sensitivity, 96.5% specificity, 0.034 FNR, 0.04 FDR, 0.034 FPR, 0.92 MCC and 96% NPV, respectively. Along with is, the mean square error (MSE) and peak-to-signal-noise ratio (PSNR) for the GMM model attained about 104 and 27.95, respectively.
We propose a W-Net architecture-based approach for detecting and localizing regions of video forged using copy-move forgery technique. The proposed methodology can be utilized for the detection of forged videos with a high degree of efficiency. The model is capable of detecting forgery even when manipulations are done in complex settings, such as those with a dynamic background or complex movement of the object in the video files. The portion of the video frame that has been tampered with using temporal copy and paste (TCP) video in painting techniques can also be localized by proposed model. With lossless video clips, we were able to achieve the best video results. Finally, we were able to develop an algorithm capable of performing a task that simply required a video clip as input.
As humanity cruises ahead in its artificial intelligence race to reach new horizons in the field, especially in the image processing arena, there have also emerged some “flies” in the ointment along the way. One such fly is the recent application known as deepfake. Deepfake signifies the media which is morphed using deep learning and artificial intelligence tools. These types of media cannot be distinguished easily and can fool one on the first instance. To counter deepfakes, deep learning is used in this paper. Here, convolutional neural networks (CNNs) are used to classify images from our dataset used (deepfake dataset). Deep learning networks to learn features of the deepfake images and the predict if an image is real or deepfake. CNNs are used as these have proved to be robust in the deep learning applications which deal with images.KeywordsDeepfakeDeep learningArtificial intelligenceConvolutional neural networks (CNNs)
In a court of law, surveillance videos and recordings are the major sources of evidence for any incident or crime. With a simple video editor, the reality could be easily manipulated. This introduces the challenge of verifying the authenticity of contents before they may be used in any critical application domains. In this paper, a video forensic technique is proposed to detect frame duplication forgery in surveillance videos. The proposed technique utilizes convolutional neural network for integrity embedding generation and matching, for video frames duplication detection and localization. Through this work, we compare the efficacy of various deep learning models in generating true embeddings of surveillance footages. To perform experiments a dataset of over 100 authentic and 300 forged, high-resolution HEVC-coded video clips is prepared from surveillance clips. Experimental results indicate that the proposed technique is a good replacement of traditional hand-crafted and compression domain feature-based approaches.KeywordsVideo forensicsFrame duplicationCNNPCAAgglomerative clustering
In today’s digital world because of the widespread availability of software tools and advancement in technology created a way for editing or modifying digital data. Availability of low cost software tools such as Adobe Premiere, Magix Vegas, Mokey by Imagineer Systems and Microsoft Movie Maker made it very easy to manipulate video content effectively. Altering the contents of digital videos became very easy because of these powerful video editing software. Video editing software is often used to copy and paste a portion of the frame from one region to another region in the same frame or another frame, in order to conceal the truth. Region duplication is one of the most common methods of video tampering. Many methodologies have been discussed by different researches to detect such tampering in digital video sequences by analyzing the spatial and temporal correlations between the pixels of the frame. However, most of the algorithms exhibit low accuracy and high computational complexity. In this paper, we are introducing a novel region duplication tampering detection technique to detect intra frame region duplication from the video sequence by means of wavelet transform and Euclidean distance. The proposed technique is evaluated using various videos where region duplication in frames introduced. From the experimental results performance of the proposed technique is evaluated in terms Precision, Recall, F1, Accuracy and computation time. Our experimental results proved that this method is highly suitable for detecting the duplicated regions in the video which is very much useful in forensic applications.
Computed Tomography signifies an imaging process that constructs
cross sectional images, which reveal the internal structure of a
scanned object in a nondestructive manner. Cone-beam CT is one of
the most advanced developments in CT imaging, which presents an
indispensable imaging technique that constructs three-dimensional
images from the scanned object, Using cone-beam X-ray projections
rather than the more often used fan-beam planar projections ensures
quicker but safer three-dimensional imaging with less radiation than
traditional CT imaging methodologies. However, these advancements
come at the expense of a challenging 3D reconstruction issue that still
requires changes in the speed and accuracy of the reconstructed image.
In this article, we present a parallel implementation of an efficient
parallel Maximum Likelihood Expectation Maximization (MLEM)
algorithm for CBCT that provides a more reliable and faster
reconstruction even with a small number of projections. We
reconstructed a test volume image with this parallel version of the
MLEM algorithm and used various image quality indices to test the
effects of our proposed approach as well as the effect of the number of
iterations on the reconstruction time and image quality. According to
the data, the reconstruction of volume images using the proposed
parallel algorithm was quicker than the standard reconstruction
method using the non-parallel MLEM algorithm.
With the advent of Internet, images and videos are the most vulnerable media that can be exploited by criminals to manipulate for hiding the evidence of the crime. This is now easier with the advent of powerful and easily available manipulation tools over the Internet and thus poses a huge threat to the authenticity of images and videos. There is no guarantee that the evidences in the form of images and videos are from an authentic source and also without manipulation and hence cannot be considered as strong evidence in the court of law. Also, it is difficult to detect such forgeries with the conventional forgery detection tools. Although many researchers have proposed advance forensic tools, to detect forgeries done using various manipulation tools, there has always been a race between researchers to develop more efficient forgery detection tools and the forgers to come up with more powerful manipulation techniques. Thus, it is a challenging task for researchers to develop h a generic tool to detect different types of forgeries efficiently. This paper provides the detailed, comprehensive and systematic survey of current trends in the field of image and video forensics, the applications of image/video forensics and the existing datasets. With an in-depth literature review and comparative study, the survey also provides the future directions for researchers, pointing out the challenges in the field of image and video forensics, which are the focus of attention in the future, thus providing ideas for researchers to conduct future research.
We describe an efficient technique that automatically detects duplicated regions in a digital image. This technique works by first applying a principal component analysis to small fixed-size image blocks to yield a reduced dimension representation. This representation is robust to minor variations in the image due to additive noise or lossy compression. Duplicated regions are then detected by lexicographically sorting all of the image blocks. We show the efficacy of this technique on credible forgeries, and quantify its robustness and sensitivity to additive noise and lossy JPEG compression. the Office for Domestic Preparedness, U.S. Department of Homeland Security (points of view in this document are those of the authors and do not necessarily represent the official position of the U.S. Department of Homeland Security).
A well-known method for image registration is based on a conventional correlation between phase-only, or whitened, versions of the two images to be realigned. The method, covering rigid translational movements, is characterized by an outstanding robustness against correlated noise and disturbances, such as those encountered with nonuniform, time varying illumination. This correspondence discusses an extension of the method to cover both translational and rotational movements. Copyright © 1987 by The Institute of Electrical and Electronics Engineers. Inc.
With the advent of sophisticated and low-cost video editing software, it is becoming increasingly easier to tamper with digital video. In addition, an ever-growing number of video surveillance cameras is giving rise to an enormous amount of video data. The ability to ensure the integrity and au- thenticity of this data poses considerable challenges. Here we begin to explore techniques for detecting traces of tam- pering in digital video. Specifically, we show how a doubly- compressed MPEG video sequence introduces specific static and temporal statistical perturbations whose presence can be used as evidence of tampering.
With the advent of high-quality digital video cameras and sophisticated video editing software, it is becoming increasingly easier to tamper with digital video. A growing number of video surveillance cameras are also giving rise to an enormous amount of video data. The ability to ensure the integrity and authenticity of these data poses considerable challenges. We describe two techniques for detecting traces of tampering in deinterlaced and interlaced video. For deinterlaced video, we quantify the correlations introduced by the camera or software deinterlacing algorithms and show how tampering can disturb these correlations. For interlaced video, we show that the motion between fields of a single frame and across fields of neighboring frames should be equal. We propose an efficient way to measure these motions and show how tampering can disturb this relationship.