Fast Localization and High Accuracy Recognition of Tire Surface Embossed Characters Based on CNN

• Published in: Applied sciences Vol. 13; no. 11; p. 6560
• Main Authors: Guo, Zhongfeng, Yang, Junlin, Qu, Xinghua, Li, Yuanxin
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 28.05.2023
• Subjects: Accuracy; Algorithms; Cameras; character recognition; Classification; Deep learning; Equipment and supplies; Image processing; LeNet-5 network; Light; Metal forming; Neural networks; SSR algorithm; template matching; Tires; Vision systems; China
• ISSN: 2076-3417; 2076-3417
• DOI: 10.3390/app13116560

To solve the problem of recognizing artificial tire-side pressure printing characters with low efficiency and high labor intensity, we propose a CNN-based method for tire surface character recognition. In the image pre-processing, the SSR algorithm is improved to enhance the contrast of characters, and the Normalized Cross Correlation template matching algorithm based on pyramid acceleration is proposed to quickly locate the “DOT” characters and segment them. The improved LeNet-5 network structure is used to recognize characters, and a self-built digital sample library is randomly divided according to the ratio of 8:2 to conduct digital recognition experiments. The experimental results show that the recognition accuracy of the training set can reach 95.9%, and the accuracy of the validation set is 99.5%. The accuracy of the testing set is 95.6%, which meets the practical application requirements. Moreover, the whole algorithm only needs to be implemented on a commonly configured CPU, reducing equipment costs.