Abstract
Intelligent transportation systems and safety driver-assistance systems are important research topics in the field of transportation and traffic management. This study investigates the key problems in front vehicle detection and tracking based on computer vision. A video of a driven vehicle on an urban structured road is used to predict the subsequent motion of the front vehicle. This study provides the following contributions. (1) A new adaptive threshold segmentation algorithm is presented in the image preprocessing phase. This algorithm is resistant to interference from complex environments. (2) Symmetric computation based on a traditional histogram of gradient (HOG) feature vector is added in the vehicle detection phase. Symmetric HOG feature with AdaBoost classification improves the detection rate of the target vehicle. (3) A motion model based on adaptive Kalman filter is established. Experiments show that the prediction of Kalman filter model provides a reliable region for eliminating the interference of shadows and sharply decreasing the missed rate.
Highlights
Vehicle detection technology is an important topic in computer vision, image processing, and pattern recognition of vehicle safety driver-assistance systems
(2) Symmetric computation based on a traditional histogram of gradient (HOG) feature vector is added in the vehicle detection phase
Shadows on the road cast by trees, buildings, and roadside signs led to a high false detection rate
Summary
Vehicle detection technology is an important topic in computer vision, image processing, and pattern recognition of vehicle safety driver-assistance systems. Srinivasa used edge filter for image segmentation and adopted clustering algorithm to reduce computation time after detection.[8] Bertozzi et al investigated large-scale parallel architecture on vehicle detection systems to improve real-time performance.[9] Vehicle detection and target tracking systems have several applications in different scenes and environments. Improving the real-time efficiency of detection and tracking requires denoising the original image and extraction of relevant information to improve the accuracy of operation and reduce the computation of subsequent modules. Image segmentation algorithm is used for vehicle detection and tracking applications, such as the optical flow and frame difference methods. To satisfy the requirement of target detection and tracking in dynamic scenes and improve robustness to environmental changes, we adopted a new method of image segmentation, namely, adaptive threshold segmentation.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
