Font Size: a A A

Design And Implementation Of The Prompt System For Special Traffic Signs Based On DSP

Posted on:2022-05-24Degree:MasterType:Thesis
Country:ChinaCandidate:Z Z JiaoFull Text:PDF
GTID:2492306509456204Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of artificial intelligence,traffic sign prompt system has important research significance in the fields of assisted driving,autonomous driving and traffic sign maintenance.Based on the analysis and summary of the latest research progress at home and abroad,the paper focuses on the design and realization of the special traffic sign prompt system,focusing on the detection and identification of traffic signs,as well as solutions for special traffic signs such as poor lighting conditions,tilt,breakage and blocking.The main research work and results of the thesis are as follows:(1)Traffic Sign Detection.Based on the analysis of traffic sign a priori information,the detection method of multi-color spatial fusion segmentation combined with shape features can effectively improve the accuracy and recall rate of traffic sign detection based on the correspondence between traffic sign color and shape.(2)Special traffic sign processing.In real road traffic scenarios,traffic signs have problems such as poor lighting conditions,tilted,broken and obscured,which increase the difficulty of accurate traffic sign detection.The paper proposes corresponding solutions for various scenarios to improve the accuracy and robustness of the system.(3)Traffic Sign Recognition(TSR).The HOG-max LBP features are fused with HOG features and LBP features.Combined with the trained SVM classifier to achieve the classification of traffic signs.The algorithm is experimentally verified to have good recognition accuracy,real-time performance and generalization capability.(4)The implementation of the traffic signs prompt system.The paper selects TMS320DM642 as the hardware development platform of the system,makes full use of the internal resources of DSP to transplant and optimize the algorithm,and improves the running efficiency of the program,so that it can be used as an in-vehicle device to detect and identify road traffic signs in real time,and prompt the driver by voice.In summary,the paper mainly studies and explores the related problems involved in the traffic sign prompt system,aiming at improving the real-time and accuracy of the system,and verifying the reliability of the method used in the paper through experiments.
Keywords/Search Tags:traffic signs detection and recognition, special traffic sign processing, feature fusion, SVM, DSP
PDF Full Text Request
Related items