Research on the recognition of surface defects in copper strip based on fuzzy neural network

Autor:	Zhen-tao Zhou, Xuewu Zhang, Yan-qiong Ding, Yan-yun Lv
Rok vydání:	2008
Předmět:	Production line Artificial neural network Computer science business.industry Feature vector Feature extraction Wavelet transform Pattern recognition STRIPS Machine learning computer.software_genre Backpropagation law.invention Statistical classification law Artificial intelligence business computer
Zdroj:	2008 IEEE Conference on Cybernetics and Intelligent Systems.
DOI:	10.1109/iccis.2008.4670927
Popis:	The quality of copper strips directly affects the performance and quality of copper and its products. So there is great significance to detect and recognize the surface defects in copper strips. The testing results from traditional manual inspection methods are unsatisfactory. So, this paper presents a novel recognition method of surface defects in copper strip based on fuzzy neural network. In this paper, the feature vectors of typical defects picked by the moment invariants form the neural network training samples and fuzzy wavelet neural network based on learning rate dynamically regulated BP algorithm identifies defects. Experiments show that this method can effectively detect surface defects in copper strips in the production line. Besides, it has a high recognition accuracy and speed.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::8e0609b6907b59c1f77744f4b1e8d3a5 https://doi.org/10.1109/iccis.2008.4670927 Zobrazit plný text záznamu