New Approach for Texture Classification Based on Concept
Mingxia Liu, Yingkun Hou, Xiangcai Zhu, Deyun Yang, Xiangzeng Meng · 2008
As the rapid development of Content-Based Image Retrieval, the Semantic understanding of image becomes the focus. However, the complexity and diversity of textures present great challenges to the traditional content analysis of images. Nowadays texture classification based on mathematical parameters is becoming very popular, but failed to break through the semantic obstacle between visual features and human understanding of textures. In this paper, a novel approach of texture classification based on conceptual words of Chinese natural language which describe various natural textures has been put forward. Then we make use of SVM classifier to classify natural textures, which transform texture visual features to semantic description. Experimental results show that this approach is useful to negotiating the "semantic gaps" between texture concepts and feature parameters on image understanding and image retrieval based on natural language.