Abstract
We propose a localization refinement approach for candidate traffic signs. Previous traffic sign localization approaches, which place a bounding rectangle around the sign, do not always give a compact bounding box, making the subsequent classification task more difficult. We formulate localization as a segmentation problem, and incorporate prior knowledge concerning color and shape of traffic signs. To evaluate the effectiveness of our approach, we use it as an intermediate step between a standard traffic sign localizer and a classifier. Our experiments use the well-known German Traffic Sign Detection Benchmark (GTSDB) as well as our new Chinese Traffic Sign Detection Benchmark. This newly created benchmark is publicly available, 1 and goes beyond previous benchmark data sets: it has over 5000 high-resolution images containing more than 14 000 traffic signs taken in realistic driving conditions. Experimental results show that our localization approach significantly improves bounding boxes when compared with a standard localizer, thereby allowing a standard traffic sign classifier to generate more accurate classification results. 1 http://cg.cs.tsinghua.edu.cn/ctsdb/
Highlights
T RAFFIC signs are specially designed graphics which give instructions and information to drivers
Color and shape priors are utilized in an iterative optimization approach to accurately segment the traffic signs as foreground objects
We have shown the effectiveness of our approach by comparing the localization quality of a cascade detector using HoG feature or Haar features, as well as the advantages of our approach when using convolutional neural network (CNN): results using the German Traffic Sign Detection Benchmark (GTSDB) and CTSDB benchmarks show that our approach can improve localization quality
Summary
T RAFFIC signs are specially designed graphics which give instructions and information to drivers. Different countries’ traffic signs vary somewhat in appearance, they share some common design principles. Traffic signs are divided according to function into different categories, in which each particular sign has the same generic appearance but differs in detail. This allows traffic sign recognition to be carried out as a two-phase task: detection and classification. The detection step focuses on localizing candidates for a certain traffic sign category, typically by placing a bounding box around regions believed to contain such a traffic sign. Manuscript received September 8, 2015; revised July 13, 2016 and November 18, 2016; accepted January 27, 2017.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: IEEE Transactions on Intelligent Transportation Systems
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.