Towards the steel plate defect detection: Multidimensional feature information extraction and fusion

摘要

Product surface quality inspection based on machine vision has been paid more and more attention in modern industrial production. Multidimensional feature information fusion also plays an important role in the detection rate and accuracy of steel plate defects. Based on the machine vision based defect detection of steel plate surface, this article mainly studies the collection, extraction and fusion of multidimensional feature information. The visual imaging system, the surface defect detection technology based on vision, the depth camera and the light source type are briefly reviewed. At the same time, the roadmap and related methods of multidimensional feature information acquisition system are studied. Finally, the multidimensional fusion defect detection technology is prospected and summarized.