Content-based lace fabric image retrieval system using texture and shape features

A content-based lace fabric image retrieval system using texture and shape features is presented in this article. The retrieval system consists of two steps: registration and identification. During the registration procedure, a feature vector corresponding to each lace fabric sample image is extracted and then stored in a feature database. During the identification procedure, a feature vector of a lace fabric query image is extracted and then the similarities between the query image and all the sample images are calculated. Finally, the retrieval results can be sorted based on their similarities to the query image. Experimental results demonstrate the effectiveness of retrieval performance of the proposed algorithm and possible practical application of the retrieval system in lace fabric industry to improve management efficiency.