Research on Defect Detection Method for Steel Metal Surface based on Deep Learning

Xiaoyang Gai, Peiran Ye, Jinglin Wang, Bingquan Wang

20202020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC)23 citationsDOI

Abstract

In the process of modern industry, the surface defects of industrial products seriously affect the quality, safety, usability and aesthetics of products. Based on the method of deep learning, this paper takes steel surface defects in industrial parts as the breakthrough point, and mainly USES the convolutional neural network algorithm in deep learning to classify and detect steel surface defects. Firstly, industrial cameras were used to collect and pre-process the steel defect images to obtain relevant data sets. Secondly, VGG model was used to improve the network features to improve the recognition of defects and realize the classification and recognition of defects. Compared with traditional methods, this method has higher accuracy and efficiency.

Topics & Concepts

Convolutional neural networkUsabilityDeep learningArtificial intelligenceComputer scienceProcess (computing)Artificial neural networkSurface (topology)Quality (philosophy)Point (geometry)Pattern recognition (psychology)Computer visionHuman–computer interactionMathematicsOperating systemGeometryEpistemologyPhilosophyIndustrial Vision Systems and Defect DetectionSurface Roughness and Optical MeasurementsImage and Object Detection Techniques