Defect Classification on Automobile Tire Inner Surfaces with Functional Classifiers
Hirotaro Tada, Akihiko Sugiura
pp. 1-10
DOI:
10.5687/iscie.34.1Abstract
In this paper, we present a method that functionally combines a convolutional neural network (CNN) and a support vector machine (SVM) to classify defects occurring on the inner surface of an automobile tire. Because such defects are usually small, the image requires high resolution to show the shape change of the defect in the image. Dividing one image into multiple images of smaller regions increases the number of images, which limits the applicable machine learning methods. For this reason, CNN is applied to the divided images of the whole tire, while SVM is applied to the divided images within the range delimited by CNN. Experimental results demonstrate that the defect detection rate of the proposed method is 100%, with an area rate of over-detection of 0.040% in the inspection range of a non-defective tire, demonstrating that the method is effective for reducing over-detection errors while maintaining defect detection accuracy.
Readers Who Read This Article Also Read
Transactions of the Institute of Systems, Control and Information Engineers Vol.34(2021), No.1
Transactions of the Institute of Systems, Control and Information Engineers Vol.34(2021), No.1
Transactions of the Institute of Systems, Control and Information Engineers Vol.34(2021), No.1