Statistical texture analysis is used to find the digital features that describe the texture, and these features or other non texture features are used to classify the regions in the image (rather than a single pixel). The autocorrelation function, gray level co-occurrence matrix, gray level run and various statistics of gray level distribution in local area of image are commonly used digital texture features. For example, the gray level co-occurrence matrix uses the spatial distribution of gray level to represent the texture. Because the gray distribution of coarse texture changes more slowly with distance than that of fine texture, they have totally different gray level co-occurrence matrix.<br>
正在翻译中..