纹理图像统计模型与纹理图像分割

STATISTICAL TEXTURE MODEL AND TEXTURE SEGMENTATION

本文详细介绍了一种新的纹理图像统计模型──Ｇｉｂｂｓ模型，给出了两种常用的Ｇｉｂｂｓ分布。在此基础上作者详细推导出了两种纹理图像的分割算法，并用一幅遥感图像进行了实验，获得了较好的分割结果。

In this paper,a new statistical model GRF (Gibbs Random Fields)is presented to represent textures in image. Two widely used distribution functions are given. Based on RGF models, two kinds of algorithms for texture segmentation have been developed. An experiment has been carried out using remotely sensed image a fairly good results were achieved.