Improved Traffic Sign Detection Algorithm in Complex Environment

Qiyuan Xiao

《International Core Journal of Engineering》 7卷1期 (2021/01) Pp. 486-492

https://doi.org/10.6919/ICJE.202101_7(1).0062

摘要

The recognition of traffic signs plays an important role in the safe driving of vehicles, especially in the presence of light changes and occlusions, the automatic recognition of traffic signs with high accuracy and good real-time performance needs to be solved urgently. Among the deep learning algorithms, YOLOV3 and Faster-RCNN have achieved excellent target detection performance. Compared with traditional feature detection and recognition methods, the use of YOLOV3 algorithm helps to solve the problem of small target detection, traffic sign recognition under the conditions of light changes, partial occlusion, etc. It is the main way to improve the performance of automatic driving and unmanned driving in the future.

關鍵字

Traffic Sign Detection ； Traffic Sign Recognition ； YOLOV3 ； Faster-RCNN ； Small Target Detection ； Automatic Driving

參考文獻

A. Mogelmose, M. M. Trivedi and T. B. Moeslund, "Vision-Based Traffic Sign Detection and Analysis for Intelligent Driver Assistance Systems: Perspectives and Survey," in IEEE Transactions on Intelligent Transportation Systems, vol. 13, no. 4, pp. 1484-1497, Dec. 2012.

N. Dwivedi, D. K. Singh and D. S. Kushwaha, "Weapon Classification using Deep Convolutional Neural Network," 2019 IEEE Conference on Information and Communication Technology, Allahabad, India, 2019, pp. 1-5.

LECUN Y,BINGO Y,HINTON G.Deep learning[J].Nature,2015,521:436-444.

R. Girshick, J. Donahue, T. Darrell and J. Malik, "Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation," 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, 2014, pp. 580-587.

R. Girshick, "Fast R-CNN," 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, 2015, pp. 1440-1448.