Title :
Content-based image retrieval algorithm oriented by users´ experience
Author :
Jiang, Linying ; Ren, Jiefu ; Li, Dancheng
Author_Institution :
Software Coll., Northeastern Univ. Shenyang, Shenyang, China
Abstract :
In order to not only improve the storage efficiency, retrieval speed and accuracy of the existing content-based image retrieval algorithm, but also improve the quality of user experience, this paper proposes a novel improved algorithm about content-based image retrieval oriented by users´ experience. According to the three elements of human visual color characteristics, this algorithm firstly quantifies the image; then it calculates the histogram in HSV color space. With the feature vectors of the obtained histogram, the index will be set in Lucene indexing library. At the same time, users´ relevance feedback information is made use of to modify the searching results, which makes results closer to users´ needs. Experimental results show that the proposed algorithm has perfect efficiency and accuracy, which makes users´ operating experience better.
Keywords :
content-based retrieval; image colour analysis; image retrieval; HSV color space; Lucene indexing library; accuracy improvement; content-based image retrieval algorithm; efficiency improvement; equality improvement; feature vectors; histogram calculation; human visual color characteristics; image quantification; relevance feedback information; retrieval speed improvement; storage improvement; user experience quality improvement; Accuracy; Algorithm design and analysis; Engines; Feature extraction; Image color analysis; Image retrieval; Vectors; Content-based image retrieval; Lucene; Relevance feedback; Users´ experience;
Conference_Titel :
Computer Science & Education (ICCSE), 2012 7th International Conference on
Conference_Location :
Melbourne, VIC
Print_ISBN :
978-1-4673-0241-8
DOI :
10.1109/ICCSE.2012.6295116