摘要
提出了一种基于同步自回归（ＳＡＲ） 模型和模糊信息原理进行纹理分割的方法。利用二阶ＳＡＲ 模型对图像纹理进行描述，用最小平方误差方法对模型参数进行估计，在对模型参数分析的基础上，将估计的参数进行改进后作为纹理的特征向量用于纹理图像的分类与分割。由于实际图像带有许多的模糊信息，组成纹理的基元和基元之间的关系也具有很大的模糊性，文中根据模糊信息原理，分析了纹理图像的模糊特性，给出了一种基于模糊贴近度的纹理分割方法。实验结果表明，与常规的距离方法相比，用文中的方法进行图像纹理分割能取得更好的效果。
In this paper, a texture segmentation approach which is based on Simultaneous Autoregressive (SAR) model and the theory of fuzzy set is presented. Image texture is expressed by the second order basic SAR model and the least square error estimation is employed for the solution of model parameters. To perform texture segmentation, the model parameters of the textures are improved, the theory of fuzzy set is introduced, and the characteristic of fuzzy information of texture image is analyzed, and an approach for segmentation of texture image based on the theory of fuzzy information is presented. Experiment shows that the better segmentation results can be obtained by the method in this paper than the traditional Euclidean distance method.
出处
《红外与激光工程》
EI
CSCD
1999年第6期21-24,共4页
Infrared and Laser Engineering
基金
国家自然科学基金资助项目!(69775022)
湖北师范学院重点基金
关键词
纹理分析
同步自回归模型
模糊信息
图像分割
texture analysis simutaneous autoregressive model segmentation fuzzy information similarity
