|
|
|
| Space Target Image Categorization Based on the Second Representation |
| Jiang Fei-yun① Sun Rui①② Zhang Xu-dong① Li Chao① |
①(School of Computer anf Information, Hefei University of Technology, Hefei 230009, China)
②(The Chery Car Postdoctoral Workstation, Wuhu 241009, China ) |
|
|
|
|
Abstract According to the characteristics of space target image, an novel method of space target image categorization based on local invariant features is proposed. The method extracts firstly local invariant features of each image and uses Gaussian Mixture Model (GMM) to establish global visual modes. Then co-occurrence matrix of the entire training set is constructed by matching local invariant features and visual models with maximum a posteriori probability and Probability Latent Semantic Analysis (PLSA) model is used to obtain latent class vector of images to achieve sencond representation. Finally, the SVM algorithm is used to implement image categorization. The experimental result demonstrates the effectiveness of the proposed method.
|
|
Received: 10 October 2012
|
|
|
|
Corresponding Authors:
Jiang Fei-yun
E-mail: 317033310@qq.com
|
|
|
|
|
|
|
2013, Vol. 35 
