Figure - available from: Journal of Electrical and Computer Engineering
This content is subject to copyright. Terms and conditions apply.
Source publication
Nowadays, real-time vehicle detection is one of the biggest challenges in driver-assistance systems due to the complex environment and the diverse types of vehicles. Vehicle detection can be exploited to accomplish several tasks such as computing the distances to other vehicles, which can help the driver by warning to slow down the vehicle to avoid...
Citations
... Most vehicle detection system developed in vision-based because of it easy to understand more than sensor-based. The vision-based consist of two steps [5] that are hypothesis generation (HG) and hypothesis verification (HV). The HG has three basic categories; (1) knowledge, (2) stereo-visionbased, and (3) motion-based methods. ...
HOG has been developed successfully in many intelligent vehicle detection systems. HOG has still interesting problems that consist of (i) redundant features and (ii) ambiguous features (similarities between vehicles and non-vehicles), which these problems effect to time computation and misclassification. The vertical direction of HOG method (V-HOG) and adding the position of orientation bins and intensity features (πHOG) improve the problems of HOG but they produced redundant and ambiguous features in various regions of vehicles. This paper proposed a new method for improving the performance of HOG that is flexibility in various regions of vehicles. The proposed method use combine different size cell of HOG that is called CDC-HOG. The CDC-HOG were conducted on GTI dataset, which consists of four regions (far, front, left, and right regions). The CDC-HOG compared with HOG, V-HOG, πHOG, and PHOG and used kernel extreme learning machine (KELM) and support vector machine (SVM) for evaluating features. The CDC-HOG with KELM produced the highest performance in terms of accuracy, true positive rate, and false positive rate for all regions .
... In the process of hypothesis generation, Abdelmoghit Zaarane et al. [133] have firstly determined position of potential vehicles using the cross-correlation technique of template matching. In the second stage, the two-dimensional discrete wavelet transformation (2D-DWT) is used in the extraction and classification of features from the generated hypotheses. ...
Smart traffic and information systems require the collection of traffic data from respective sensors for regulation of traffic. In this regard, surveillance cameras have been installed in monitoring and control of traffic in the last few years. Several studies are carried out in video surveillance technologies using image processing techniques for traffic management. Video processing of a traffic data obtained through surveillance cameras is an instance of applications for advance cautioning or data extraction for real-time analysis of vehicles. This paper presents a detailed review of literature in vehicle detection and classification techniques and also discusses about the open challenges to be addressed in this area of research. It also reviews on various vehicle datasets used for evaluating the proposed techniques in various studies.
... Different methods, such as basic filtering, Gabor filtering, SIFT features, harr and histogram of oriented gradients, adaptive filtering, wavelet denoising, homomorphic enhancement, and other techniques have been used to circumvent this area by a number of researchers. [20][21][22][23][24][25][26] These techniques focus on emphasizing the details of the picture being improved. Deep features-based systems have been presented as a solution to the difficulties associated with handmade features-based classifiers. ...
For controlling and managing the traffic and to help traffic surveillance, the vehicles classification is a matter of great importance. In the last few decades, vehicle classification systems based on pattern recognition have been utilized to enhance the efficiency for traffic monitoring systems. In the literature many deep learning networks are suggested for vehicle classification. Even though deep learning algorithms are fascinating and growing research area. However, there are several barriers that slow down its progress. The greatest factor that reduces the progress of deep learning systems is the quality of the image. The available vehicle image datasets are affected by noise, weather, and illumination variations. To overcome these issues, we suggest a robust deep learning system by combining bilateral filter individually with three different networks for the improvement of the robustness of vehicle classification in real‐time application. For validation the suggested networks are assessed on CompCars dataset. The study of literature has reported many Suggested CNN models, some of are discussed here. Lee et al. have employed SqueezeNet model and achieved accuracy of 0.963. Wang et al. have achieved accuracy of 0.989 through H‐squeezeNet model. By employing Faster‐RCNN Inception and SSD MobileNet‐v2 Giron et al. have achieved accuracy of 0.864. Zhang et al. have used Inception‐v3 and attained an accuracy of 0.974. In this article, the proposed CNN bilateral integrated models have accomplished the accuracies as 0.999, 0.997, and 0.988 for Inception‐v3, MobileNet‐v2, SqueezeNet networks, respectively. The results demonstrate the recommended techniques outperform the traditional deep learning networks.
... The range of problems investigated by cross-correlation analysis is very broad, starting from sociology, economy, econophysics [20,23,33,34], transport [35,36], genome analysis, biology, food network, biochemistry network, science collaboration network [37], up to sport [38], and many others. ...
Within the paper, the problem of globalisation during financial crises is analysed. The research is based on the Forex exchange rates. In the analysis, the power law classification scheme (PLCS) is used. The study shows that during crises cross-correlations increase resulting in significant growth of cliques, and also the ranks of nodes on the converging time series network are growing. This suggests that the crises expose the globalisation processes, which can be verified by the proposed analysis.
... Traffic congestion, traffic violations, stealing cars, and fugitive criminals impose big challenges on traffic management systems. Several systems are developed to solve these problems such as self-driving systems [1], Traffic surveillance systems [2,3], Tracking vehicle systems and Vehicle speed detection systems [4,5]. License plate detection and character recognition (LPDR) is one of the most important topics in intelligent infrastructure systems, like electronic payment systems (for tolls, parking, and public transportation). ...
The License Plate detection and recognition (LPDR) is a challenging task that plays a significant role in intelligent transportation systems (ITS). Where it could be used as a core in various applications, such as security, traffic control, and electronic payment systems (e.g. freeway toll payment and parking fee payment). A variety of algorithms are developed for this work and each one has advantages and disadvantages for extracting plates in images under different circumstances. However, the complexity of some methods requires a high calculation cost and this could be time-consuming. In the current paper, a simple and efficient method is proposed to tackle the issue of license plate detection and character recognition. The license plate is detected first based on the two-dimensional wavelet transform to extract the vertical edges of the input image. The high density of vertical edges is computed first to detect the potential areas of the license plate. Then these potential areas are verified by using a plate/non-plate CNN classifier. After the license plate is detected, the characters are segmented by using a simple method that is based on the empty distance between the characters. Finally, these character candidates are classified by training another CNN classifier. The experiments were done on vehicles that carry Moroccan license plates and showed high accuracy, where the results obtained go up to 99.43% in term of localization and 98.9% in term of recognition. Besides, the efficiency and the high accuracy of the proposed method were proved by performing a comparison with other works from the literature on different datasets. All processes of the proposed method were implemented on a Hardware Processor System (HPS) located in a VEEK-MT2S provided by TERASIC.
... In the process of hypothesis generation, Abdelmoghit Zaarane et al. [133] have firstly determined position of potential vehicles using the cross-correlation technique of template matching. In the second stage, the two-dimensional discrete wavelet transformation (2D-DWT) is used in the extraction and classification of features from the generated hypotheses. ...
Smart traffic and information systems require the collection of traffic data from respective sensors for regulation of traffic. In this regard, surveillance cameras have been installed in monitoring and control of traffic in the last few years. Several studies are carried out in video surveillance technologies using image processing techniques for traffic management. Video processing of a traffic data obtained through surveillance cameras is an instance of applications for advance cautioning or data extraction for real-time analysis of vehicles. This paper presents a detailed review of literature in vehicle detection and classification techniques and also discusses about the open challenges to be addressed in this area of research. It also reviews on various vehicle datasets used for evaluating the proposed techniques in various studies.
... The researchers face a lot of difficulties in self-driving field due to the dynamic and complex environment and the complex movement in fast way. The automated vehicles need to detect the other vehicles whatever their shape and type [1][2][3]16]. Thus, several algorithms should be performed such as vehicle detection, license plate detection [15] and speed and distance estimation. ...
... Detecting objects is an important task in distance measurement systems where the performance of vehicle detection algorithm acts in proportion to the distance measurement performance. Therefore, before measuring the vehicle distance, an efficient vehicle detection algorithm is applied [1]. This algorithm is composed of two steps: hypothesis generation step and hypothesis verification step. ...
... Then, they apply some stereo matching algorithms to match the detected objects in both cameras, which consume time. However, the main idea in this paper is to detect vehicles by applying the vehicle detection method [1] to the images captured by single camera. Then, match them with the same vehicles in the images captured by the other camera. ...
The focus of this paper is inter-vehicles distance measurement which is a very important and challenging task in image processing domain. Where it is used in several systems such as Driving Safety Support Systems (DSSS), autonomous driving and traffic mobility. In the current paper, we propose an inter-vehicle distance measurement system for self-driving based on image processing. The proposed system uses two cameras mounted as one stereo camera, in the hosting vehicle behind the rear-view mirror. The detection of vehicles is performed first in a single camera using a recent powerful work from the literature. Then, the same vehicle is detected in the image captured by the second camera using template matching technique. Thus, the inter-vehicle distance is calculated using a simple method based on the position of the vehicle in both cameras, geometric derivations and additional technical data such as distance between the cameras and some other specific angles (e.g. the cameras view field angle). The results of the extensive experiments showed the high accuracy of the proposed method compared to the previous works from literature and it allows to measure efficiently the distances between the vehicles and the hosting vehicle. In addition, this method could be used in several systems of various domains in real time regardless of the object types. The experiments results were done on a Hardware Processor System (HPS) located in a VEEK-MT2S provided by TERASIC.
... The vehicle detection task has become a very challenging task and gained a very important place in research field, where it has played the main role in several systems such as DAS and Traffic Surveillance System. There are lot of algorithms dedicated for properly performing this important task [2] [4]. However, these works, identical to those of the most researchers, are proposed to detect vehicles at daytime and good lighting conditions, which leaves us a great lack of works dedicated to the night-time vehicle detection. ...
... The first experiment was done on a Moroccan License Plate using three videos sequences collected from various situations in different lighting conditions. The vehicles are detected first from the sequences using the background subtraction method [12] for two sequences taken by a fixed camera and using vehicle detection algorithm proposed in [13] for the third sequence taken by a camera fixed in moving car, where the vehicles images are the input of the proposed system. The second experiment was done using the Caltech cars dataset (Rear) [14] and AOLP dataset which is categorized into three subsets: access control (AC), traffic law enforcement (LE), and road patrol (RP) that are available in [15]. ...
... The accuracy of distance measurement between vehicles is implicitly related to the performance of the vehicle detection. Hence, a high performance vehicle detection algorithm [7] is used first to detect the exact location of the vehicle. the used algorithm is divided into two principal parts: hypothesis generation and hypothesis verification. ...
... In the previous works, the vehicles are detected in the images taking by both cameras then they are matched using a matching method, which consumes time. the principle idea proposed in our system to overcome this issue is to detected vehicles by performing a high performance vehicle detection method [7] on images taken by only one camera, then to look for the corresponding vehicle in the images taken by the second camera by crossing the images in the same horizontal location of the vehicle using the crosscorrelation technique, as shown in figure 2. ...
