文献类型：期刊文献

英文题名：Shape Recognition of Tyre Marking Points Based on Support Vector Machine

作者：Wang, Yong[1];Guo, Hui[1]

机构：[1]E China Univ Sci & Technol, Sch Mech & Power Engn, Shanghai 200237, Peoples R China

年份：2015

卷号：32

期号：2

起止页码：123

外文期刊名：IETE TECHNICAL REVIEW

收录：;EI(收录号：20214611180328);WOS:【SCI-EXPANDED(收录号:WOS:000351842700006)】；

语种：英文

外文关键词：Shape recognition; Tyre industry; SVM; Tyre marking points

摘要：With the rapid development of automobile industry, the demand of tyre is increasing greatly. But tyre classification implementation is still not efficient and effective for tyre industry. This paper proposes a method of tyre classification using tyre marking points indicating shapes based on support vector machine (SVM). First, the tyre marking points printed on the profiles of tyre are acquired by a designed image acquisition device. Second, those tyre marking points images are handled by image preprocessing methods of image denoising, segmentation, and edge detection. After image preprocessing, the contour coordinates of tyre marking points are extracted. The Fourier transform method is further applied to the extracted contour coordinates to acquire the Fourier descriptors of tyre marking points as feature vectors. Followed, those extracted Fourier descriptors are inputted as SVM classifiers. Finally, the proposed classification techniques based on SVM is employed to classify tyre as corresponding types according to shape recognition of tyre marking points. The experimental results show that the proposed tyre classification method can meet the requirement of tyre industry by a mean shape recognition accuracy of 97.25%.