期刊名称：International Journal on Smart Sensing and Intelligent Systems
印刷版ISSN：1178-5608
出版年度：2015
卷号：8
期号：2
页码：992-1010
出版社：Massey University
摘要：In order to solve the low discrimination of image representations in complicated duplicateimage detection, this paper presents a complicated duplicate image representation approach based ondescriptor learning. This approach firstly formulates objective function as minimizing empirical erroron the labeled data. Then the tag matrix and the classification matrix of training dataset are broughtinto the objective function to ensure semantic similarity. Finally, by relaxing the constraints, we can getthe learning hashes. The learning hashes are used to quantify local descriptors of images into binarycodes and the frequency histograms of binary codes are as image representations. Experimental resultsdemonstrate that compared with the state-of-the-art algorithms, this approach can effectively improvethe discrimination of image presentations by introducing semantic information
