论文信息 - Based on the Research of Two-Way Dynamic Matching to the Central Area in CBIR

Based on the Research of Two-Way Dynamic Matching to the Central Area in CBIR

As the differences of the backgrounds and the distribution of the key areas in the images, if carring out similarity fully by comparing the corresponding regional areas of the images in the retrieval,it is easily to interfere the search results. To this problem, this paper presents a new idea that when doing the similarity compare of the two images based on ensuring information of the image edge, get the central areas respectively, and take two-way image matching, and then receive the similarity of two images by getting the sum of the two similarity. By the experimental test, this method can reduce interference and increase accuracy under the premise that retrieve time is close.

Dazhong Wu | Aibing Zhang

[1] Jitendra Malik,et al. Color- and texture-based image segmentation using EM and its application to content-based image retrieval , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[2] Xiaojun Qi,et al. A novel fusion approach to content-based image retrieval , 2005, Pattern Recognit..

[3] M. Tahoun,et al. A robust content-based image retrieval system using multiple features representations , 2005, Proceedings. 2005 IEEE Networking, Sensing and Control, 2005..