A new image retrieval model based on monogenic signal representation |
| |
| Affiliation: | 1. Graduate Institute of Mathematics and Science Education, National Chiayi University, Chiayi 621, Taiwan, ROC;2. Department of Computer Science and Information Engineering, National Chiayi University, Chiayi 600, Taiwan, ROC;1. Faculty of Information Science and Engineering, Ningbo University, Ningbo 315211, China;2. School of Information and Mechatronics, Gwangju Institute of Science and Technology (GIST), Gwangju 500-712, Republic of Korea;1. School of Computer and Electronic Information, Guangxi University, Nanning, Guangxi 530004, China;2. Department of Computer Science and Engineering, Arizona State University, Tempe, AZ 85287, USA;1. School of Computer Science and Technology, Harbin Institute of Technology, Weihai 264209, China;2. School of Information and Electrical Engineering, Harbin Institute of Technology, Weihai 264209, China;3. Science and Technology on Information Assurance Laboratory, Beijing 100072, China |
| |
| Abstract: | This paper proposes a novel image retrieval model based on monogenic signal representation. An original image is decomposed into three complementary components: amplitude, orientation and phase by monogenic signal representation. The monogenic variation in each local region and monogenic feature in each pixel are encoded, and then the statistical features of the local features encoded are calculated. In order to overcome the problem of high feature dimensionality, the local statistical features extracted from the complementary monogenic components are projected by block-based fisher discriminant analysis, which not only reduces the dimensionality of the features extracted, but also enhances its discriminative power. Finally, these features reduced are fused for effective image retrieval. Experimental results show that our scheme can effectively describe an image, and obviously improve the average retrieval precision. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
