Texture feature based on local Fourier transform

Texture analysis plays an important role in image analysis and pattern recognition. Texture-based modelling has become a powerful tool for content-based image retrieval. A new texture feature named LFH is proposed based on the local-similarity and 8-neighbour gray-tone spatial dependencies by using local Fourier transform. We extracted a histogram of local Fourier coefficients maps of the texture image for texture classification and texture retrieval. Experiments on 13 samples of Brodatz textures demonstrate the efficiency and effect of texture classification. The performance of texture retrieval is illustrated.