Traffic sign recognition improves speed and accuracy with branched yolov2

Traffic Sign Recognition for Autonomous Driving Using Branched YOLOv2 and Geometric Features

Computer Vision and Pattern Recognition

Summary

Recognizing traffic signs quickly and correctly is important for self-driving cars. The authors improve an existing method called YOLOv2 by adding branches that can stop early when the task is easy, saving time. They also use geometric shapes inside traffic signs to tell similar signs apart more accurately. Their tests with 10 types of traffic signs show better speed or accuracy using these ideas.

What this means in practice

Tested on one dataset.

Authors

Arefeh Rezaei

Abstract

Traffic sign recognition (TSR) is an important perception task for autonomous driving and advanced driver-assistance systems, where a system must both localize traffic signs and determine their semantic classes efficiently. This work presents a TSR system based on YOLOv2 for simultaneous detection and classification. Two complementary modifications are studied. First, YOLOv2 is extended with intermediate prediction layers, forming a branched architecture that can terminate inference early for easy cases and reduce computation time. Both whole-image and cell-wise branching strategies are investigated. Second, geometric information is introduced to reduce classification errors between visually similar signs. An unsupervised Bayesian image-segmentation method produces binary representations that are compared with class-specific geometric templates inside YOLOv2 bounding boxes. This information is used either during inference or as an additional signal during training. A dedicated dataset is constructed by combining GTSDB and GTSRB samples using seamless cloning and controlled image transformations. Experiments cover ten traffic-sign classes, with 3,000 training and 300 test samples. The selected branched architecture reports 0.647 s runtime and 0.680 mAP, compared with 0.6607 s and 0.680 mAP for baseline YOLOv2. Geometric verification during inference increases mAP to 0.713, while the geometric-feature training variant achieves 0.697 mAP with a reported runtime of 0.6608 s.