Specular reflection Surface Defects Detection by using Deep Learning

Abstract

As you know that defects inspection of specular surface is very difficult because its specular reflection is very strong and defects' reflection is weaker. And the existing computer vision-based industrial parts surface defect detection methods are limited by environmental factors, and the image preprocessing process is complex. On the other hand, with the rapid development of Convolutional Neural Networks (CNN) that is one type of deep learning and has excellent performance for image processing, has led to the rapid development of computer vision research based on deep learning. In this paper, we proposed an ensemble CNN in which integrated two convolutional neural network models for surface defect detection, and obtained better results.