Gaussian Pyramid Based Laws' Mask Descriptor for Texture Classification

An image pyramid is a special type of subband transformation. It is constructed with a recursive method that represents multi-resolution and multi-scale format. Pyramid analysis is useful for texture mapping at different resolutions. Laws has developed to detect various types of textures by using local masks named as Laws' mask descriptor. Laws' mask descriptor does not consider about the multiple scales of an image while extracting the texture features. In this paper, we have suggested a novel method of integrating Gaussian pyramid with Laws' mask descriptor for feature extraction. Experimental results show that the proposed method of feature extraction technique provides better classification accuracy the traditional Laws' mask descriptor.