中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Quick image stitching algorithm based on template matching for Mask defect detection

文献类型:会议论文

作者Wei, Hongbin; Hu, Song
出版日期2020-06-29
会议日期January 18, 2020 - January 19, 2020
会议地点Chongqing, China
卷号1549
期号5
DOI10.1088/1742-6596/1549/5/052023
英文摘要With the computer technology and image processing technology gradually applied to the mask detection system, computer vision detection method has become the mainstream way of automatic mask inspection. Masking is usually a large area, and in order to ensure detection accuracy, the camera's field of view is small, and the panorama stitching of mask images has become an essential image processing link in defect detection. In the mask image acquisition system, the precision workbench is in the working mode of horizontal vertical translation, and the adjacent two images have horizontal and vertical offsets. In view of the characteristics of big data volume of mask image, cell repetition, small variation of rotation scale, this paper proposes a fast multi-mask image stitching algorithm based on template matching. The template matching algorithm is optimized, compared with the traditional image stitching algorithm, which greatly improves the stitching speed, avoids the problem of feature matching repeated graphics mismatch, and the stitching effect is good. © Published under licence by IOP Publishing Ltd.
会议录Journal of Physics: Conference Series
文献子类会议论文
语种英语
ISSN号1742-6588
源URL[http://ir.ioe.ac.cn/handle/181551/9892]  
专题光电技术研究所_微电子装备总体研究室(四室)
作者单位Institute of Optics and Electronics Chinese Academy of Sciences, ChengDu; 610000, China
推荐引用方式
GB/T 7714
Wei, Hongbin,Hu, Song. Quick image stitching algorithm based on template matching for Mask defect detection[C]. 见:. Chongqing, China. January 18, 2020 - January 19, 2020.

入库方式: OAI收割

来源:光电技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。