Multi-Feature Fusion Method Applied in Texture Image Segmentation
Hui Du, Zhihe Wang, Dan Wang, Xiaoli Wang · 2018
Texture patterns are complex and varied, and their applications are diverse. In many cases, the effect of image segmentation by a single texture feature extraction method is not ideal. In response to this problem, this paper proposes a multi-feature fusion method to process the texture feature extraction. The proposed method combines the gray level co-occurrence matrix (GLCM), Gabor wavelet transform and local binary pattern (LBP). It has the advantages of the above three texture feature extraction methods. Then, we use the algorithm K-means to implement the image segmentation by clustering the extracted texture features. As a result, the proposed algorithm can precisely realize the clustering for texture image segmentation. The experimental results show that the proposed algorithm is more efficient than the single texture feature extraction methods.