Fast detection and classification of defects on treated metal surfaces using a backpropagation neural network
作者： C. Neubauer
DOI: 10.1109/IJCNN.1991.170551
关键词:
摘要: A fast classifier based on a neural network is described which the central part of an optical inspection system. Defects treated metal surfaces are detected and classified by textural segmentation. The main purpose this work development system for wide range real-time applications. Therefore, preprocessing image data reduced to calculation gray-value histograms 10*10 pixel window. By using only eight classes in histograms, efficient reduction obtained. calculated each window presented three-layered perceptron net defect detection classification. This method applied 100% surface rolling bearing rings. Depending class investigated misclassification rate ranged from 1.5 11.5%. >
sci-hub.se 参考文章(3)
Felicia M. Vilnrotter, Ramakant Nevatia, Keith E. Price, Structural Analysis of Natural Textures IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. PAMI-8, pp. 76- 89 ,(1986) , 10.1109/TPAMI.1986.4767754
D.E. RUMELHART, G.E. HINTON, R.J. WILLIAMS, Learning internal representations by error propagation Neurocomputing: foundations of research. pp. 673- 695 ,(1988) , 10.1016/B978-1-4832-1446-7.50035-2
Daniel I. Barnea, Harvey F. Silverman, A Class of Algorithms for Fast Digital Image Registration IEEE Transactions on Computers. ,vol. C-21, pp. 179- 186 ,(1972) , 10.1109/TC.1972.5008923