A Multiscale Analysis Method of Image Texture
影象纹理的多尺度分析

It is important to consider the role of scale for texture analysis. In this paper, a texture detector based on 2D Gabor function and visual textural perception is established first, then based on the texture detector and wavelet theory of multiscaledecomposition, a multiscale texture analysis method is proposed, and techniques formultiscale texture feafure fusion are advanced according to the lateral inhibition andend-inhibition in neurodynamics. The multiscale texture analysis techniques give representations between spatial space and Fourier space, and provide a hierarchical analysisframe for image fexture. They can detect different scale textural features, correspondsto the visual texture perception, and have the ability to recognize textureimage effectively.