Recognition of flaw on pre-sensitized plates using computer vision

	Recognition of flaw on pre-sensitized plates using computer vision
	Lang B.; Shen L.; Han T.
	2011
发表期刊	Journal of Computational Information Systems
ISSN	15539105
卷号	7 期号:1 页码:168-175
摘要	This paper presents a machine vision recognition scheme for flaw on pre-sensitized plates (PS Plates) using hybrid segmentation for detecting flaws and the back propagation neural network for classifying flaws. Firstly, median filtering, wavelet denosing and mathematical morphology are applied to enhance flaw information. Secondly, segmentation based on an improved Canny algorithm is used to detect the flaws. In the meantime, features of candidate area of flaw are extracted. Thirdly, classification of flaws is accomplished by the well-trained Back Propagation Neural Network (BPNN) based on flaws database that consist of flaws judged by experienced human inspector. Experiment results show that hybrid segmentation and the back propagation neural network have better performance in detection and recognition of flaw with low contrast and blurry contour. 2011 Binary Information Press.
收录类别	EI
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/34602
专题	中科院长春光机所知识产出
推荐引用方式 GB/T 7714	Lang B.,Shen L.,Han T.. Recognition of flaw on pre-sensitized plates using computer vision[J]. Journal of Computational Information Systems,2011,7(1):168-175.
APA	Lang B.,Shen L.,&Han T..(2011).Recognition of flaw on pre-sensitized plates using computer vision.Journal of Computational Information Systems,7(1),168-175.
MLA	Lang B.,et al."Recognition of flaw on pre-sensitized plates using computer vision".Journal of Computational Information Systems 7.1(2011):168-175.