No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Novel Skin and Mole Pattern Identification Using Deep Residual Pooling Network (DRPN)
Abstract
The process of identifying a criminal is extremely challenging if no eyewitnesses are available or closedcircuit television (CCTV) footage does not show the criminal's face. With the use of high-definition cameras and soft biometrics, forensic science has made great strides in identifying criminals based on their bodily traits. Due to the use of one type of soft biometric for identification, many researchers have focused on this area and have failed to achieve better matching. Despite the fact that similar soft biometrics share many of the same features, one type of soft biometric matching has been insufficient and resulted in inaccurate results. As a result, it becomes difficult to make distinct classifications. Therefore, this study proposes the implementation of a suspect identification system that uses mole pattern analysis and skin tone matching as soft biometrics. The deep residual pooling network is employed to analyze mole patterns, while the histogram equalization method is employed to analyze skin tones. We have also improved the performance of the deep seg-net and the histogram equalization technique by incorporating an aggregator operator and an adaptive pixel-wise noise cancellation filter. The experimental results show that the proposed method attains higher accuracy when the loss rate is gradually decreased. The proposed method attains 89.21% accuracy which is higher than the existing approaches and the efficiency of the proposed method is compared with various texture encoders © 2023, International Journal of Intelligent Engineering and Systems.All Rights Reserved.
ResearchGate has not been able to resolve any citations for this publication.
Facial skin condition is perceived as a vital indicator of the person’s apparent age, perceived beauty, and degree of health. Machine-learning-based software analytics on facial skin conditions can be a time- and cost-efficient alternative to the conventional approach of visiting facial skin care shops or dermatologist’s offices. However, the conventional CNN-based approach is shown to be limited in the diagnosis performance due to the intrinsic characteristics of facial skin problems. In this paper, the technical challenges in facial skin problem diagnosis are first addressed, and a set of 5 effective tactics are proposed to overcome the technical challenges. A total of 31 segmentation models are trained and applied to the experiments of validating the proposed tactics. Through the experiments, the proposed approach provides 83.38% of the diagnosis performance, which is 32.58% higher than the performance of conventional CNN approach.
Background
The lack of dark skin images in pathologic skin lesions in dermatology resources hinders the accurate diagnosis of skin lesions in people of color. Artificial intelligence applications have further disadvantaged people of color because those applications are mainly trained with light skin color images.
Objective
The aim of this study is to develop a deep learning approach that generates realistic images of darker skin colors to improve dermatology data diversity for various malignant and benign lesions.
Methods
We collected skin clinical images for common malignant and benign skin conditions from DermNet NZ, the International Skin Imaging Collaboration, and Dermatology Atlas. Two deep learning methods, style transfer (ST) and deep blending (DB), were utilized to generate images with darker skin colors using the lighter skin images. The generated images were evaluated quantitively and qualitatively. Furthermore, a convolutional neural network (CNN) was trained using the generated images to assess the latter’s effect on skin lesion classification accuracy.
Results
Image quality assessment showed that the ST method outperformed DB, as the former achieved a lower loss of realism score of 0.23 (95% CI 0.19-0.27) compared to 0.63 (95% CI 0.59-0.67) for the DB method. In addition, ST achieved a higher disease presentation with a similarity score of 0.44 (95% CI 0.40-0.49) compared to 0.17 (95% CI 0.14-0.21) for the DB method. The qualitative assessment completed on masked participants indicated that ST-generated images exhibited high realism, whereby 62.2% (1511/2430) of the votes for the generated images were classified as real. Eight dermatologists correctly diagnosed the lesions in the generated images with an average rate of 0.75 (360 correct diagnoses out of 480) for several malignant and benign lesions. Finally, the classification accuracy and the area under the curve (AUC) of the model when considering the generated images were 0.76 (95% CI 0.72-0.79) and 0.72 (95% CI 0.67-0.77), respectively, compared to the accuracy of 0.56 (95% CI 0.52-0.60) and AUC of 0.63 (95% CI 0.58-0.68) for the model without considering the generated images.
Conclusions
Deep learning approaches can generate realistic skin lesion images that improve the skin color diversity of dermatology atlases. The diversified image bank, utilized herein to train a CNN, demonstrates the potential of developing generalizable artificial intelligence skin cancer diagnosis applications.
International Registered Report Identifier (IRRID)
RR2-10.2196/34896
Identifying criminals in serious crimes from digital images is a challenging forensic task as their faces will be covered in most cases. In addition, the only available information will be hand. A single robust technique to identify the criminals from arm’s hair patterns can be a potential cost-effective and unobtrusive solution in various other areas such as in criminal psychiatric hospitals during rehabilitation to identify and track patients instead of using barcoding, radio frequency identification (RFID), and biometrics. The existing state-of-the-art methods for person identification uses convolutional neural network (CNN) and long short-term memory (LSTM)-based architectures which require the entire data to be trained once again when new data comes. To address these issues, we proposed a novel Siamese network-based architecture which not only reduces this training paradigm but also performs better than several existing methods. Since there were no standard datasets for person identification from arm’s hair patterns, we created a database with several voluntary participants by collecting their hands’ images. Several data augmentation techniques are also used to make the database more robust. The experimental results show that the proposed architecture performs better for the created database with mAP, mINP, and R1 of 94.8, 90.0, and 93.5, respectively. The proposed CTTSN performs well for the closed world person re-identification problem using soft biometric features in real time (52 frames per second).
Skin lesion detection and analysis are very important because skin cancer must be found in its early stages and treated immediately. Once installed in the body, skin cancer can easily spread to other body parts. Early detection would represent a very important aspect since, by ensuring correct treatment, it could be curable. Thus, by taking all these issues into consideration, there is a need for highly accurate computer-aided systems to assist medical staff in the early detection of malignant skin lesions. In this paper, we propose a skin lesion classification system based on deep learning techniques and collective intelligence, which involves multiple convolutional neural networks, trained on the HAM10000 dataset, which is able to predict seven skin lesions including melanoma. The convolutional neural networks experimentally chosen, considering their performances, to implement the collective intelligence-based system for this purpose are: AlexNet, GoogLeNet, GoogLeNet-Places365, MobileNet-V2, Xception, ResNet-50, ResNet-101, InceptionResNet-V2 and DenseNet201. We then analyzed the performances of each of the above-mentioned convolutional neural networks to obtain a weight matrix whose elements are weights associated with neural networks and classes of lesions. Based on this matrix, a new decision matrix was used to build the multi-network ensemble system (Collective Intelligence-based System), combining each of individual neural network decision into a decision fusion module (Collective Decision Block). This module would then have the responsibility to take a final and more accurate decision related to the prediction based on the associated weights of each network output. The validation accuracy of the proposed system is about 3 percent better than that of the best performing individual network.
One of the major health concerns for human society is skin cancer. When the pigments producing skin color turn carcinogenic, this disease gets contracted. A skin cancer diagnosis is a challenging process for dermatologists as many skin cancer pigments may appear similar in appearance. Hence, early detection of lesions (which form the base of skin cancer) is definitely critical and useful to completely cure the patients suffering from skin cancer. Significant progress has been made in developing automated tools for the diagnosis of skin cancer to assist dermatologists. The worldwide acceptance of artificial intelligence-supported tools has permitted usage of the enormous collection of images of lesions and benevolent sores approved by histopathology. This paper performs a comparative analysis of six different transfer learning nets for multi-class skin cancer classification by taking the HAM10000 dataset. We used replication of images of classes with low frequencies to counter the imbalance in the dataset. The transfer learning nets that were used in the analysis were VGG19, InceptionV3, InceptionResNetV2, ResNet50, Xception, and MobileNet. Results demonstrate that replication is suitable for this task, achieving high classification accuracies and F-measures with lower false negatives. It is inferred that Xception Net outperforms the rest of the transfer learning nets used for the study, with an accuracy of 90.48. It also has the highest recall, precision, and F-Measure values.
Melanoma skin cancer is the most life-threatening and fatal disease among the family of skin cancer diseases. Modern technological developments and research methodologies made it possible to detect and identify this kind of skin cancer more effectively, however, the automated localization and segmentation of skin lesion at earlier stages is still a challenging task due to the low contrast between melanoma moles and skin portion and a higher level of color similarity between melanoma affected and non-effected areas. In this paper, we present a fully automated method for segmenting the skin melanoma at its earliest stage by employing a deep-learning-based approach named faster region-based convolutional neural networks (RCNN) along with fuzzy k-means clustering (FKM). Several clinical images are utilized to test the presented method, so that it may help the dermatologist in diagnosing this life-threatening disease at its earliest stage. The presented method first preprocesses the dataset images to remove the noise and illumination problems and enhance the visual information before applying the faster-CNN to obtain the feature vector of fixed length. After that FKM has been employed to segment the melanoma-affected portion of skin with variable size and boundaries. The performance of the presented method is evaluated on the three standard datasets namely ISBI-2016, ISIC-2017, and PH2 and results show that the presented method outperforms the state-of-the-art approaches. The presented method attains an average accuracy of 95.40%, 93.1%, and 95.6% on the ISIC-2016, ISIC-2017, and PH2 datasets respectively which is showing its robustness to skin lesion recognition and segmentation.
Melanoma is a lethal skin cancer disease affecting millions of people around the globe and has a high mortality rate. Dermatologists perform the manual inspection through visual analysis of pigmented skin lesions for melanoma identification at the early stage. However, manual inspection for melanoma detection is limited due to variable accuracy and lesser availability of dermatologists. Therefore, there exists an urgent need to develop automated melanoma detection methods that can effectively localize and classify skin lesions. Accurate localization and classification of the melanoma lesions is a challenging task due to the presence of low contrast information between the moles and skin part, the massive color similarity between the infected and non-infected skin portions, presence of noise, hairs, and tiny blood vessels, variations in color, texture, illumination, contrast, blurring, and melanoma size. To address these afore-mentioned challenges, we propose an effective and efficient melanoma detection method. The proposed method consists of three steps: i) image preprocessing, ii) employing Faster Region-based Convolutional Neural Network (Faster-RCNN) for melanoma localization, and iii) application of Support Vector Machine (SVM) for the classification of localized melanoma region into benign and malignant classes. Performance of the proposed method is evaluated on the benchmark ISIC-2016 dataset launched by ISBI challenge-2016 that is diverse in terms of variations in illumination, color, texture, and size of melanoma, and presence of blurring, noise, hairs, and tiny blood vessels, etc. Moreover, we have also performed a cross-dataset validation over the ISIC-2017 dataset to show the efficacy of our method in real-world scenarios. Our experimental results illustrate that the proposed framework is efficient and able to effectively localize and classify the melanoma lesion than state-of-the-art techniques.
Usage of sketches for offender recognition has turned out to be one of the law enforcement agencies and defense systems’ typical practices. Usual practices involve producing a convict’s sketch through the crime observer’s explanations. Nevertheless, researches have effectively proved the failure of customary practices as they carry a maximum level of discrepancies in the process of identification. The advent of computer vision techniques has replaced this traditional procedure with intelligent machines capable of ruling out the possible discrepancies, thus assisting the investigation process and considering the relevant points mentioned earlier. This research paper has investigated an adversarial network toward achieving color photograph images out of sketches, which are then classified using pre-trained transfer learning models to accomplish the identification process. Further, to enhance the adversarial network’s performance factor in terms of photogeneration, we also employed a novel sketch generator based on the gamma adjustment technique. Experimental trials are steered with image datasets open to the research community. The trials’ outcomes evidenced that the proposed system achieved the lowest similarity score of 91% and the average identification accuracy of more than 70% on all the datasets. Comparative analysis portrayed in this work also attests that the proposed technique performs ably better than any other state-of-the-art techniques.
Skin cancer is developed due to abnormal cell growth. These cells are grown rapidly and destroy the normal skin cells. However, it’s curable at an initial stage to reduce the patient’s mortality rate. In this article, the method is proposed for localization, segmentation and classification of the skin lesion at an early stage. The proposed method contains three phases. In phase I, different types of the skin lesion are localized using tinyYOLOv2 model in which open neural network (ONNX) and squeeze Net model are used as a backbone. The features are extracted from depthconcat7 layer of squeeze Net and passed as an input to the tinyYOLOv2. The propose model accurately localize the affected part of the skin. In Phase II, 13-layer 3D-semantic segmentation model (01 input, 04 convolutional, 03 batch-normalization, 03 ReLU, softmax and pixel classification) is used for segmentation. In the proposed segmentation model, pixel classification layer is used for computing the overlap region between the segmented and ground truth images. Later in Phase III, extract deep features using ResNet-18 model and optimized features are selected using ant colony optimization (ACO) method. The optimized features vector is passed to the classifiers such as optimized (O)-SVM and O-NB. The proposed method is evaluated on the top MICCAI ISIC challenging 2017, 2018 and 2019 datasets. The proposed method accurately localized, segmented and classified the skin lesion at an early stage.
Nowadays security became a major global issue. To manage the security issue and its risk, different kinds of biometric authentication are available. Face recognition is one of the most significant processes in this system. Since the face is the most important part of the body so the face recognition system is the most important in the biometric authentication. Sometimes a human face affected due to different kinds of skin problems, such as mole, scars, freckles, etc. Sometimes some parts of the face are missing due to some injuries. In this paper, the main aim is to detect a facial spots present in the face. The total work divided into three parts first, face and facial components are detected. The validation of checking facial parts is detected using the Convolution Neural Network (CNN). The second part is to find out the spot on the face based on Normalized Cross-Correlation and the third part is to check the kind of spot based on CNN. This process can detect a face under different lighting conditions very efficiently. In cosmetology, this work helps to detect the spots on the human face and its type which is very helpful in different surgical processes on the face.
Handwriting and signature biometrics have a long history in the literature, especially in terms of identity recognition and/or verification; nevertheless, it reveals more information therefore provides more opportunities for personal characteristics estimation, particularly, emotional state. However, almost all publicly or commercially available databases do not include contributors’ demographic labels or emotional status in addition to their identification labels, leading these available databases to be only useful for verification and identification based research studies. For this reason, this paper proposes both offline and online handwriting and signature biometric database with a wide range of ground truths (emotional status labels – happy, sad and stress) in addition to the identity labels and tries to predict ones’ emotional state, namely happy, sad and stress, from their online biometric handwriting and signatures. The proposed database comprised total of 134 participants with 804 handwriting and 8040 signature biometric samples. The database presented also includes individuals’ demographic information such as age, gender, handedness, education level and nationality. Subsequently, there are several experiments have been conducted, with different thresholds, which present the usability of the proposed database and preliminary results of emotional state prediction from both signature and handwriting biometrics. The experiments achieved remarkable success especially on stress prediction for handwriting. On the other hand, considering the results from signature biometrics, it is observed that happy and then sad and finally stress class forms the most part of the prediction accuracy.
Combination of multiple information extracted from different biometric modalities in multimodal biometric recognition system aims to solve the different drawbacks encountered in a unimodal biometric system. Fusion of many biometrics has proposed such as face, fingerprint, iris…etc. Recently, electrocardiograms (ECG) have been used as a new biometric technology in unimodal and multimodal biometric recognition system. ECG provides inherent the characteristic of liveness of a person, making it hard to spoof compared to other biometric techniques. Ear biometrics present a rich and stable source of information over an acceptable period of human life. Iris biometrics have been embedded with different biometric modalities such as fingerprint, face and palm print, because of their higher accuracy and reliability. In this paper, a new multimodal biometric system based ECG-ear-iris biometrics at feature level is proposed. Preprocessing techniques including normalization and segmentation are applied to ECG, ear and iris biometrics. Then, Local texture descriptors, namely 1D-LBP (One D-Local Binary Patterns), Shifted-1D-LBP and 1D-MR-LBP (Multi-Resolution) are used to extract the important features from the ECG signal and convert the ear and iris images to a 1D signals. KNN and RBF are used for matching to classify an unknown user into the genuine or impostor. The developed system is validated using the benchmark ID-ECG and USTB1, USTB2 and AMI ear and CASIA v1 iris databases. The experimental results demonstrate that the proposed approach outperforms unimodal biometric system. A Correct Recognition Rate (CRR) of 100% is achieved with an Equal Error Rate (EER) of 0.5%.
One of the most common types of human malignancies is skin cancer, which is chiefly diagnosed visually, initiating with a clinical screening followed by dermoscopic analysis, histopathological assessment, and a biopsy. Due to the fine-grained differences in the appearance of skin lesions, automated classification is quite challenging through images. To attain highly segregated and potentially general tasks against the finely grained object categorized, deep convolutional neural networks (CNNs) are used. In this paper, we propose a new prediction model that classifies skin lesions into benign or malignant lesions based on a novel regularizer technique. Hence, this is a binary classifier that discriminates between benign or malignant lesions. The proposed model achieved an average accuracy of 97.49%, which in turns showed its superiority over other state-of-the-art methods. The performance of the CNN in terms of AUC-ROC with an embedded novel regularizer is tested on multiple use cases. The area under the curve (AUC) achieved for nevus against melanoma lesion, seborrheic keratosis vs. basal cell carcinoma lesion, seborrheic keratosis vs. melanoma lesion, solar lentigo vs. melanoma lesion is 0.77, 0.93, 0.85, and 0.86, respectively. Our results showed that the proposed learning model outperformed the existing algorithm and can be used to assist medical practitioners in classifying various skin lesions.
Training of neural networks for automated diagnosis of pigmented skin lesions is hampered by the small size and lack of diversity of available datasets of dermatoscopic images. We tackle this problem by releasing the HAM10000 ("Human Against Machine with 10000 training images") dataset. We collected dermatoscopic images from different populations acquired and stored by different modalities. Given this diversity we had to apply different acquisition and cleaning methods and developed semi-automatic workflows utilizing specifically trained neural networks. The final dataset consists of 11788 dermatoscopic images, of which 10010 will be released as a training set for academic machine learning purposes and will be publicly available through the ISIC archive. This benchmark dataset can be used for machine learning and for comparisons with human experts. Cases include a representative collection of all important diagnostic categories in the realm of pigmented lesions. More than 50% of lesions have been confirmed by pathology, the ground truth for the rest of the cases was either follow-up, expert consensus, or confirmation by in-vivo confocal microscopy.
The human hand possesses distinctive features which can reveal gender information. In addition, the hand is considered one of the primary biometric traits used to identify a person. In this work, we propose a large dataset of human hand images with detailed ground-truth information for gender recognition and biometric identification. The proposed dataset comprises of 11,076 hand images (dorsal and palmar sides), from 190 subjects of different ages under the same lighting conditions. Using this dataset, a convolutional neural network (CNN) can be trained effectively for the gender recognition task. Based on this, we design a two-stream CNN to tackle the gender recognition problem. This trained model is then used as a feature extractor to feed a set of support vector machine classifiers for the biometric identification task. To the best of our knowledge, this is the first dataset containing the image of the dorsal side of the hand, captured by a regular digital camera and subsequently considered the first study attempting to use the features extracted from the dorsal side of the hand for gender recognition and biometric identification.
In this paper, we propose a novel M-SegNet architecture with global attention for the segmentation of brain magnetic resonance imaging (MRI). The proposed architecture consists of a multiscale deep network at the encoder side, deep supervision at the decoder side, a global attention mechanism, different sizes of convolutional kernels, and combined-connections with skip connections and pooling indices. The multiscale side input layers were used to support deep layers for extracting the discriminative information and the upsampling layer at the decoder side provided deep supervision, which reduced the gradient problem. The global attention mechanism is utilized to capture rich contextual information in the decoder stage by integrating local features with their respective global dependencies. In addition, multiscale convolutional kernels of different sizes were used to extract abundant semantic features from brain MRI scans in the encoder and decoder modules. Moreover, combined-connections were used to pass features from the encoder to the decoder path to recover the spatial information lost during downsampling and makes the model converge faster. Furthermore, we adopted uniform non-overlapping input patches to focus on fine details for the segmentation of brain MRI. We verified the proposed architecture on publicly accessible datasets for the task of segmentation of brain MRI. The experimental results show that the proposed model outperforms conventional methods by achieving an average Dice similarity coefficient score of 0.96.
Teledermatology is one of the most illustrious applications of telemedicine and e-health. In this area, telecommunication technologies are utilized to transfer medical information to the experts. Due to the skin's visual nature, teledermatology is an effective tool for the diagnosis of skin lesions, especially, in rural areas. Further, it can also be useful to limit gratuitous clinical referrals and triage dermatology cases. The objective of this research is to classify the skin lesion image samples, received from different servers. The proposed framework comprises two modules including the skin lesion localization/segmentation and classification. In the localization module, we propose a hybrid strategy that fuses the binary images generated from the designed 16-layered convolutional neural network model and improved high dimension contrast transform (HDCT) based saliency segmentation. To utilize maximum information extracted from the binary images, a maximal mutual information method is proposed, which returns the segmented RGB lesion image. In the classification module, a pre-trained DenseNet201 model is re-trained on the segmented lesion images using transfer learning. Afterward, the extracted features from two fully connected layers are down-sampled using the t-distribution stochastic neighbor embedding (t-SNE) method. These resultant features are finally fused using a multi canonical correlation (MCCA) approach and are passed to a multi-class ELM classifier. Four datasets (i.e., ISBI2016, ISIC2017, PH2, and ISBI2018) are employed for the evaluation of segmentation task, while HAM10000, the most challenging dataset, is used for the classification task. Experimental results in comparison with the state-of-the-art methods affirm the strength of our proposed framework.
Current deep learning-based texture recognition methods extract spatial orderless features from pre-trained deep learning models that are trained on large-scale image datasets. These methods either produce high dimensional features or have multiple steps like dictionary learning, feature encoding and dimension reduction. In this paper, we propose a novel end-to-end learning framework that not only overcomes these limitations, but also demonstrates faster learning. The proposed framework incorporates a residual pooling layer consisting of a residual encoding module and an aggregation module. The residual encoder preserves the spatial information for improved feature learning and the aggregation module generates orderless feature for classification through a simple averaging. The feature has the lowest dimension among previous deep texture recognition approaches, yet it achieves state-of-the-art performance on benchmark texture recognition datasets such as FMD, DTD, 4D Light and one industry dataset used for metal surface anomaly detection. Additionally, the proposed method obtains comparable results on the MIT-Indoor scene recognition dataset. Our codes are available at https://github.com/maoshangbo/DRP-Texture-Recognition.
Introduction:
In a case of child pornography, only the dorsum of the offender's hand was clearly visible. After identification of a suspect, the question arose of whether and how it is possible to identify or exclude the suspect as perpetrator according to the morphology of the hand vein pattern.
Material and methods:
A simple approach to use the hand vein pattern in crime suspects as a tool for identification was tested. In this study, the hand vein patterns of 30 study participants were analysed from conventional frames on videography. A standardised grid system consisting of six lines and four sectors was applied on the dorsum of the hands. Vein branchings within the sectors and line crossings of the veins were counted, leading to a total of 11 variables for each hand.
Results:
A positive identification of each of the 30 test participants was possible for each hand when taking only the first five variables into account. A random overlapping prediction was obtained by statistically simulating hand vein patterns of different numbers of persons using this sample. Considering the hand vein frequencies in this sample, the results indicate that the chance for two persons having the same pattern is smaller than 1:1000.
Conclusions:
It can be concluded that the introduced grid system approach can be an appropriate simple and non-costly tool for the analysis of the pattern of hand veins for identification purposes.
We present DeepPrint, a deep network, which learns to extract fixed-length fingerprint representations of only 200 bytes. DeepPrint incorporates fingerprint domain knowledge, including alignment and minutiae detection, into the deep network architecture to maximize the discriminative power of its representation. The compact, DeepPrint representation has several advantages over the prevailing variable length minutiae representation which (i) requires computationally expensive graph matching techniques, (ii) is difficult to secure using strong encryption schemes (e.g., homomorphic encryption), and (iii) has low discriminative power in poor quality fingerprints where minutiae extraction is unreliable. We benchmark DeepPrint against two top performing COTS SDKs (Verifinger and Innovatrics) from the NIST and FVC evaluations. Coupled with a re-ranking scheme, the DeepPrint rank-1 search accuracy on the NIST SD4 dataset against a gallery of 1.1 million fingerprints is comparable to the top COTS matcher, but it is significantly faster (
DeepPrint:
98.80% in 0.3 seconds vs.
COTS A:
98.85% in 27 seconds). To the best of our knowledge, the DeepPrint representation is the most compact and discriminative fixed-length fingerprint representation reported in the academic literature.
Person re-identification (re-ID) has been a popular topic in computer vision and pattern recognition communities for a decade. Several important milestones such as metric-based and deeply-learned re-ID in recent years have promoted this topic. However, most of existing re-ID works are designed for closed-world scenarios rather than realistic open-world settings, which limits the practical application of re-ID technique. On one hand, the performance of the latest re-ID methods have surpassed the human-level performance on several commonly used benchmarks (e.g. Market1501 and CUHK03), which are collected from closed-world scenarios. On the other hand, open-world tasks that are less developed and more challenging have received increasing attention in the re-ID community. Therefore, this study starts the first attempt to analyze the trends of open-world re-ID and summarizes them from both narrow and generalized perspectives. In the narrow perspective, open-world re-ID is regarded as person verification (i.e., Open-set re-ID) instead of person identification, that is, the query person may not occur in the gallery set. In the generalized perspective, application-driven methods that are designed for specific applications are defined as generalized open-world re-ID. Their settings are usually close to realistic application requirements. Specifically, this survey mainly includes the following four points for open-world re-ID: (1) analyzing the discrepancies between closed-and open-world scenarios; (2) describing the developments of existing Open-set re-ID works and their limitations; (3) introducing specific application-driven works from three aspects, namely, raw data, practical procedure, and efficiency; (4) summarizing state-of-the-art methods and future directions for open-world re-ID. This survey on open-world re-ID provides a guidance for improving the usability of re-ID technique in practical applications.
An important aim of research on the blind image quality assessment (IQA) problem is to devise perceptual models that can predict the quality of distorted images with as little prior knowledge of the images or their distortions as possible. Current state-of-the-art “general purpose” no reference (NR) IQA algorithms require knowledge about anticipated distortions in the form of training examples and corresponding human opinion scores. However we have recently derived a blind IQA model that only makes use of measurable deviations from statistical regularities observed in natural images, without training on human-rated distorted images, and, indeed without any exposure to distorted images. Thus, it is “completely blind.” The new IQA model, which we call the Natural Image Quality Evaluator (NIQE) is based on the construction of a “quality aware” collection of statistical features based on a simple and successful space domain natural scene statistic (NSS) model. These features are derived from a corpus of natural, undistorted images. Experimental results show that the new index delivers performance comparable to top performing NR IQA models that require training on large databases of human opinions of distorted images. A software release is available at http://live.ece.utexas.edu/research/quality/niqe_release.zip.
This paper presents a new fuzzy switching median (FSM) filter employing fuzzy techniques in image processing. The proposed filter is able to remove salt-and-pepper noise in digital images while preserving image details and textures very well. By incorporating fuzzy reasoning in correcting the detected noisy pixel, the low complexity FSM filter is able to outperform some well known existing salt-and-pepper noise fuzzy and classical filters.
Modified-BRISQUE as no reference image quality
- L S Chow
L. S. Chow and H. Rajagopal, "Modified-BRISQUE as no reference image quality
International Journal of Intelligent Engineering and Systems, Vol.16, No.3, 2023
DOI: 10.22266/ijies2023.0630.53
Evaluation of Perception based Image Quality Evaluator (PIQE) no-reference image quality score for 99mTc-MDP bone scan images
- A Pandey
- D Yadav
- A Sharma
- D Sonker
- C Patel
- C Bal
- R Kumar
A. Pandey, D. Yadav, A. Sharma, D. Sonker, C.
Patel, C. Bal, and R. Kumar. "Evaluation of
Perception based Image Quality Evaluator
(PIQE) no-reference image quality score for
99mTc-MDP bone scan images", Journal of
Nuclear Medicine, Vol. 61, 2020.
Comparative analysis of Histo-gram Equalization techniques
- S Patel
- M Goswami
S. Patel and M. Goswami, "Comparative
analysis of Histo-gram Equalization techniques",
In: Proc. of 2014 International Conference on
Contemporary Computing and Informatics
(IC3I), IEEE, pp. 167-168, 2014.