Identification and Detection of Traffic Signs based on Faster R-CNN

Ao Qin ； Yujun Zhang

《World Scientific Research Journal》 6卷12期 (2020/12) Pp. 268-274

https://doi.org/10.6911/WSRJ.202012_6(12).0037

摘要

This paper applies the Tensorflow deep learning framework and adopts the target detection algorithm in deep learning to solve the problem of road traffic sign identification. First, 90.8%mAP was obtained by training the deep learning target detection algorithm of Faster R-CNN, and then 93.12%mAP was obtained by training the data set into large target, medium target and small target.

關鍵字

Target detection algorithm ； Deep learning ； Faster R - CNN ； Sub-target training ； mAP

參考文獻

HUANG H,HOU LY.Speed limit sign detection based on Gaussian color model and template matching[c].IEEE International Conference on Vision,Image and Sig-nal Processing,2017: 118-122.

XU Z,REN J Y,BAO C Q. A Traffic signs' detection method of contour approximation based on concave re-moval[c]. 2016 Chinese Control and Decision Confer-ence,2016: 5199-5204.

YANG Y,LUO H L,XU HR,et al. Towards real -time traffic sign detection and classification[J].. IEEE Transactions on Intelligent Transportation Systems,2016,17(7) : 2022-2031.

WANG C Y.Research and application of traffic sign de-tection and recognition based on deep learning[c]. In-ternational Conference on Robots &Intelligent System,2018: 150-152.

Ren S,He K, Girshick R,etal. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2017, 39(6):1137-1149.