Title :
Computer-aided design bitmap retrieval methods based on shape features
Author :
Yuan Yin; Shengxiang Gao; Zhengtao Yu; Yunshan Zhang
Author_Institution :
School of Information Engineering and Automation, Kunming University of Science and Technology, China, 650500
Abstract :
This paper presents a computer-aided design (CAD) bitmap retrieval method based the on shape characteristics of CAD bitmaps. First, using the Canny edge detection algorithm to extract edge shapes of the CAD bitmaps, the edge direction histogram of CAD bitmaps are constructed according to the edge shapes. Next, using Euclidean distance and cosine value to calculate the distances between edge histograms, the similarities between CAD bitmaps are obtained according to the distances between histograms. Finally, sorting series of CAD bitmaps are obtained according to similarity values. The experimental results show that using shape features results in better effects than using color or texture features for CAD bitmap retrieval.
Keywords :
"Design automation","Image edge detection","Shape","Histograms","Feature extraction","Euclidean distance","Color"
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2015 12th International Conference on
DOI :
10.1109/FSKD.2015.7382126