论文部分内容阅读
利用我们研制成的基于知识的区域分析系统,我们对自然景物图象做了分析、解释。本文介绍:(1)图象初始分割;(2)图象特征提取;(3)规则集构造。这里,图象初始分割采用的是模糊域方法,它基于Fuzzy C-means 算法,并在此基础上修改了收敛准则,增加了迭代分割功能。图象特征分为主持征及从特征,它们建立在层次化的区域数据结构上。适合于区域分析的规则集已包括三类规则,它们不仅具有较强的知识表示能力,而且易于控制及利用。本文介绍针对包括天空、道路、树木、建筑物等物体的简单景物图象所做的解释实验,并给出了实验结果.
Using our knowledge-based regional analysis system developed by us, we analyze and explain natural landscape images. This article describes: (1) initial segmentation of images; (2) image feature extraction; (3) rule set construction. Here, the image segmentation is based on fuzzy domain method, which is based on Fuzzy C-means algorithm, and on the basis of which the convergence criterion is modified and the iterative segmentation function is added. Image features are divided into host sign and feature, they are based on the hierarchical regional data structure. Rule sets suitable for regional analysis have included three types of rules, they not only have a strong ability to represent knowledge, but also easy to control and use. This paper presents an explanation experiment for simple scene images including sky, roads, trees, buildings and other objects, and gives the experimental results.