Title :
Research and Design of Drawings Retrieval System Based on Spatial Relationships and Shape
Author :
Song, Haiyu ; Wang, Pengjie ; Li, Xiongfei ; Li, Houjie
Abstract :
CAD drawings reuse is of great importance in promoting design efficiency and productivity. The ability to retrieve CAD drawings is the core of drawings reuse task. Most of the existing drawings retrieval systems rely only on text descriptions. Due to complexity of visual feature, a little system is based on content. A simplified topological relationship is proposed to describe visual features for dimension reduction. An effective topological relationship vector is constructed by graph spectra. A novel retrieval system based on visual contents, including topological relationships, shape, and directional relationships, is proposed. The experiments show that the proposed approach and system are superior to the text-based and traditional content-based retrieval systems.
Keywords :
CAD; content-based retrieval; graph theory; vectors; CAD drawings reuse; content-based retrieval systems; dimension reduction; drawings retrieval system; graph spectra; text-based retrieval systems; topological relationship vector; Computer aided manufacturing; Content based retrieval; Data mining; Design automation; Engineering drawings; Group technology; Indexing; Information retrieval; Shape control; Spatial databases; Dimension reduction; Drawings retrieval; Graph spectra; Topological relationship;
Conference_Titel :
Computing, Communication, Control, and Management, 2008. CCCM '08. ISECS International Colloquium on
Conference_Location :
Guangzhou
Print_ISBN :
978-0-7695-3290-5
DOI :
10.1109/CCCM.2008.43