Real‐time vehicle detection and counting in complex traffic scenes using background subtraction model with low‐rank decomposition

Yang, Honghong; Qu, Shaojian

doi:10.1049/iet-its.2017.0047

Cited by 91 publications

(50 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, vehicle tracking was not used for vehicle counting in the three methods mentioned in [2,27,31]. Moreover, Yang et al [32] adopted background subtraction method for vehicle detection and Kalman filter algorithm for vehicle tracking to achieve vehicle counting. However, vehicle counting employed Deep Neural Networks (DNN) for vehicle detection and KLT tracker for vehicle tracking were proposed by Abdelwahab [16].…”

Section: Results and Discussion Of Vehicle Countingmentioning

confidence: 99%

“…Compared with the background subtraction method in image processing to detect vehicles, the accuracy of our deep learning-based detection method has been greatly improved. On the other hand, compared with KLT [16] and Kalman filter tracking algorithms [32], our proposed tracking algorithm does not need to predict the vehicle position. Instead, it considers the occlusion, deformation, and other problems caused by the vehicle movement, in which constraints have greatly improved tracking efficiency and accuracy.…”

Section: Results and Discussion Of Vehicle Countingmentioning

confidence: 99%

See 1 more Smart Citation

Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Meng

Song

Zhang

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Vehicle counting plays a significant role in vehicle behavior analysis and traffic incident detection for established video surveillance systems on expressway. Since the existing sensor method and the traditional image processing method have the problems of difficulty in installation, high cost, and low precision, a novel vehicle counting method is proposed, which realizes efficient counting based on multivehicle detection and multivehicle tracking. For multivehicle detection tasks, a construction of the new expressway dataset consists of a large number of sample images with a high resolution (1920 × 1080) captured from real-world expressway scenes (including the diversity climatic conditions and visual angles) by Pan-Tilt-Zoom (PTZ) cameras, in which vehicle categories and annotation rules are defined. Moreover, a correlation-matched algorithm for multivehicle tracking is proposed, which solves the problem of occlusion and vehicle scale change in the tracking process. Due to the discontinuity and unsmooth of the trajectories that occurred during the tracking process, we designed a trajectory optimization algorithm based on least square method. Finally, a new vehicle counting method is designed based on the tracking results, in which the driving direction information of the vehicle is added in the counting process. The experimental results show that the proposed counting method in this research can achieve more than 93% accuracy and an average speed of 25 frames per second in expressway video sequence.

show abstract

Section: Results and Discussion Of Vehicle Countingmentioning

confidence: 99%

Section: Results and Discussion Of Vehicle Countingmentioning

confidence: 99%

Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Meng

Song

Zhang

et al. 2020

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…Based on this, many researchers devote to the research on this hot topic. For decades of years, many algorithms and frameworks are proposed [4][5][6][7][8][9][10][11]. Due to the development of computer vision technology and the popularity of the camera, the vision-based detection and tracking have became the most popular research branch.…”

Section: Introductionmentioning

confidence: 99%

“…Regard to the detection, there are two technical frameworks: appearance-feature and motion-based. For the former, the appearance, such as the size, shape and color etc is extracted by some typical descriptors including HOG, SURF, gabor, Harris and optical flow etc [4][5][6][7][8]. Considering the dynamic of vehicle, motion-based algorithms are given in [9][10][11].…”

Section: Introductionmentioning

confidence: 99%

Vehicle Detection and Tracking Based on GMM and Enhanced Camshift Algorithm

Zhong¹,

Zhang²,

Zhao³

2018

JEEE

View full text Add to dashboard Cite

Vehicle detection and tracking is an important part of the intelligent transportation system. With the rapid development of computer vision, video based vehicle detection and tracking technology has become a hot topic. In this paper, on the foundation of the present work, an enhanced detection tracking algorithm is proposed based on the popular Gauss mixture model(GMM) and Camshift. First, GMM is used to extract the foreground, and then the morphological operations is carried out to enhance the image, so that to remove the random noises. Finally, enhanced Camshift is designed to track the vehicle which is discussed in detail below. The experimental results demonstrate that the tracking accuracy can be improved.

show abstract

“…Several approaches have been proposed to achieve vehicle counting in unconstrained environments by relying on portable roadside sensors [5] or IoT using Raspberry Pi [6]. Promising results have also been demonstrated in recent studies that utilises conventional computer vision techniques [3,8,9,10] and convolutional neural network (CNN) based method [7]. Vehicle counting is generally accomplished by first inferring motion flow of objects in a video sequence, followed by using the detected salience or statistical features to count number of objects [8,9,10].…”

mentioning

confidence: 99%

Contour Based Tracking for Driveway Entrance Counting System

Abdullah

Zaini

Yilmaz

et al. 2019

IJIE

View full text Add to dashboard Cite

Urbanization has been one of the core socioeconomic trends that lead to the ever-expanding transportation supply and growing number of vehicles [1]. Generally, motorized vehicles, including cars, is the primary mode of transportation in urban areas. However, on the contrary to counting the number of people in a video, vehicle counting is largely under-studied but essential for intelligent transport management and forensic analysis [2]. For instance, real-time vehicle counting is useful for traffic control and traffic management [3]. In some cases, vehicle counting is important for parking management and it can be used to ensure safety of drivers and passengers on the road [4].

show abstract

Real‐time vehicle detection and counting in complex traffic scenes using background subtraction model with low‐rank decomposition

Cited by 91 publications

References 33 publications

Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Video-Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation-Matched Tracking Using Image Data from PTZ Cameras

Vehicle Detection and Tracking Based on GMM and Enhanced Camshift Algorithm

Contour Based Tracking for Driveway Entrance Counting System

Contact Info

Product

Resources

About