Speed limit traffic sign recognition in night mode based on deformable spatial pyramid

This thesis report is submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science and Engineering, 2016.Cataloged from PDF version of thesis report.Includes bibliographical references (page 50-51).Traffic signs play a vital role in transportation system. Many road accidents occur due to over speed. Detecting and classifying different group of Speed limit traffic signs can save our lives as well as resources. In this research we propose a novel approach towards the detection of these signs. In our proposed system with the help of color and non-color information of traffic signs we first detect the presence of a sign and the classify it. For detection we used circle Hough transformation along with segmentation and labeling. After extracting sign from a scene we match the sign against a classified dataset. We used deformable spatial pyramid matching for recognition of the sign. Once we find a match, our system returns the class or speed limit. Our experiment shows that the recognition rate is very high and we compare our result with another approach at the end.Subroto Nag PinkuB. Computer Science and Engineerin