Journal of Mathematical Sciences: Advances and ApplicationsAuthor's: Qinge Wu, Anping Zheng, Yanfeng Wang and Guangzhao Cui
Pages: [161] - [183]
Received Date: November 14, 2008
Submitted by: Jianxin Dai
In order to be able to better deal with uncertain information, this paper presents an uncertain reasoning approach based on rough set theory and other uncertainty theories. This paper studies mainly the application of the reasoning approach on image recognition. The simulation results show, the recognition precision based on the new reasoning approach is improved from previous 75.95 percent to now 83.33 percent at average, and the new reasoning approach has also some advantages, such as, it has the faster recognition speed, the lower storage capacity, and does not need any prior information in addition to data processing, these results indicate that the reasoning approach is more effective and feasible than the old reasoning approaches. Moreover, this paper also makes a comprehensive comparison to the new reasoning approach and the old reasoning approaches. Finally, some prospects for future research are given. In this paper, these researches on the reasoning approach for the image recognition not only are of important theoretical value to image processing, but also promote the applications of navigation systems and target recognition.
rough sets, basic degree of belief, threshold value, image recognition.
