With the development of neural networks, detection accuracy and speed constantly improved. However, the detection effect is still insufficient in some special scenarios such as traffic environment. Therefore, we combine neural network with prior knowledge to improve its performance in vehicle detection. In this paper, we propose two effective prior: proportion and area prior to enhance the vehicle detection ability of neural network in traffic environment. The proportion and area prior is the statistical data of the vehicle at different angles and distances from the camera. In the traffic monitoring video, the proportion of most vehicles is mainly divided into several values. The area of all vehicles is also included between the thresholds. Experimental results demonstrate the effect of prior. Detection effect for vehicle in traffic environment of sample network in this paper increase by 6.56%.
Vehicle Detection Based on Area and Proportion Prior with Faster-RCNN
Smart Innovation, Systems and Technologies
2020-07-17
12 pages
Article/Chapter (Book)
Electronic Resource
English
Preprocessed Faster RCNN for Vehicle Detection
IEEE | 2018
|A Faster RCNN-Based Pedestrian Detection System
IEEE | 2016
|Improved Faster RCNN for Traffic Sign Detection*
IEEE | 2020
|Agricultural pest detection algorithm based on improved faster RCNN
British Library Conference Proceedings | 2022
|