A Novel Detection Method for Three Date Marks of Industrial Product based on Machine Vision

Abstract

In this paper, a novel identify scheme for Three Date Marks(TDMS) based on delayed residual method is proposed. First of all, two images with delay are acquired by a machine vision detection system, and a residual character image is obtained through delayed residual method. Then, a feature extraction kernel is designed according to the size of characters to strengthen the strokes. In addition, the speckle noise is well removed using designed corrosion and dilation. And we modified the median filter to improve the recognition performance of the proposed method. Finally, we developed a complete visual detection system to implement the proposed method. The extensive experimental results indicate that proposed identify scheme has high recognition rate and detection efficiency, and can fully meet the application requirements of the real-time industrial production lines.