基于机器视觉的交通标志识别研究
|
摘要
在对交通标志识别系统分析的基础上,提出了交通标志检测与识别的方案。系统对采集到的图像灰度化、平滑处理后进行归一化操作,并统一图像尺寸大小,方便特征提取。把处理好的图像进行特征分割,利用HOG算法进行特征提取,采用SVM风险算法,训练数据库实现交通标志有效、自动、快速的分类。最后,采用MATLAB进行图像识别仿真,结果表明,该算法能准确识别检测出标志,并且在保证实时性的前提下,准确率较高。
Based on the analyses of traffic sign recognition system, the scheme of traffic sign detection and recognition is put forward in this paper. The system normalizes the collected images after graying and smoothing processing, and unifies the image size to facilitate feature extraction. For the processed image feature segmentation, HOG algorithm is used to extract features, and SVM risk algorithm is used to train the database to realize the effective, automatic and fast classification of traffic signs. Finally, MATLAB is used for image recognition simulation;and the results show that the algorithm can accurately identify and detect signs, and has high accuracy on the premise of ensuring real-time.
Based on the analyses of traffic sign recognition system, the scheme of traffic sign detection and recognition is put forward in this paper. The system normalizes the collected images after graying and smoothing processing, and unifies the image size to facilitate feature extraction. For the processed image feature segmentation, HOG algorithm is used to extract features, and SVM risk algorithm is used to train the database to realize the effective, automatic and fast classification of traffic signs. Finally, MATLAB is used for image recognition simulation;and the results show that the algorithm can accurately identify and detect signs, and has high accuracy on the premise of ensuring real-time.
引文
[1] 张曙.自然环境下交通标志的检测及识别算法研究[D].武汉:武汉理工大学,2014(06):3-5.
[2] 肖若荣.基于机器视觉的交通标志识别关键技术研究[D].长沙:中南大学,2013:2-3.
[3] 肖艳秋,杜江恒,闻萌莎,等.基于颜色特征和改进支持向量机算法的交通标志检测与识别[J].轻工学报,2018,33(03):57-65.
[4] 孙露霞,张尤赛,李永顺,等.基于感兴趣区域和HOG-CTH特征的交通标志检测[J].计算机与数字工程,2018,46(06):1222-1226.
[5] 张传伟,崔万豪.基于Gabor特征提取和SVM交通标志识别方法研究[J].现代电子技术,2018,41(17):136-140.
[2] 肖若荣.基于机器视觉的交通标志识别关键技术研究[D].长沙:中南大学,2013:2-3.
[3] 肖艳秋,杜江恒,闻萌莎,等.基于颜色特征和改进支持向量机算法的交通标志检测与识别[J].轻工学报,2018,33(03):57-65.
[4] 孙露霞,张尤赛,李永顺,等.基于感兴趣区域和HOG-CTH特征的交通标志检测[J].计算机与数字工程,2018,46(06):1222-1226.
[5] 张传伟,崔万豪.基于Gabor特征提取和SVM交通标志识别方法研究[J].现代电子技术,2018,41(17):136-140.