Color Image Retrieval Using Fuzzy Measure Hamming and S-Tree

Abstract

Abstract. This chapter approaches the image retrieval system on the base of the colors of image. It creates fuzzy signature to describe the color of image on color space HSV and builds fuzzy Hamming distance (FHD) to evaluate the similarity between the images. In order to reduce the storage space and speed up the search of similar images, it aims to create S-tree to store fuzzy signature relies on FHD and builds image retrieval algorithm on S-tree. Then, it provides the content-based image retrieval (CBIR) and an image retrieval method on FHD and S-tree. Last but not least, based on this theory, it also presents an application and experimental assessment of the process of querying similar image on the database system over 10,000 images.

Content-based image retrieval using a signature graph and a self-organizing map

“…A number of works on image retrieval based on meta-data have been published recently, including image retrieval using a binary index and a signature tree (Manolopoulos et al, 2003), a similar image retrieval method based on the binary signature (Chappell and Geva, 2013), color image retrieval based on the binary signature , a color-based image retrieval technique that uses a spiral 425 bit-string representation for the color content of an image (Abdesselam et al, 2010), image retrieval based on the similarity measure EMD and the S-tree (Le and Van, 2013), color image retrieval using the Hamming distance and the S-tree ( Van and Le, 2014a), image retrieval based on the signature graph ( Van and Le, 2014c;2014b), multimedia content-based visual retrieval based on the binary signature of the SIFT (Zhou et al, 2014), image matching using a variable-length signature of the image (Liu et al, 2015), video matching using the binary signature (Li et al, 2005), content-based video using a geometric relation based on the binary signature (Ozkan et al, 2014), etc.…”

“…Moreover, we compare the performance of the proposed method with the state-of-the-art methods such as the S-tree, the S-kGraph, the points-of-interest method, the color histogram, the fuzzy color histogram, or local feature regions Le and Van, 2013;Van and Le, 2014a;2014c;2014b). The results of comparisons are shown in Tables 1-5.…”

“…Additionally, we compare the results with other co-techniques such as linear methods, the S-tree, or the S-kGraph (Le and Van, 2013;Van and Le, 2014a;2014c;2014b). The results are described in Figs.…”