摘要
图像分割对于图像分析和图像理解有着极为重要的意义。提出一个基于遗传算法的图像分割算法 ,在充分利用遗传算法优异的全局搜索能力的基础上 ,该算法根据图像的灰度分布按照模糊聚类的分割策略实现图像的分割 ,并且根据预定的隶属度阈值自动地确定聚类数目。理论分析和实验表明 ,所提出的算法能产生很好的分割效果 。
Image segmentation is very important for image analysis and image understanding.A genetic algorithm based approach has been proposed for image segmentation.Taking full advantage of the excellent global searching capability of genetic algorithm and basing on gray distribution of image and segmentation strategy of fuzzy clustering,the algorithm proposed implements the process of segmentation,while the class number is ascertained automatically based on a predefined threshold of membership.The analysis in theory and the experiments indicate that the algorithm proposed in the paper not only can get very good effect of segmentation but also can ascertain the count of the classes automatically.
出处
《测控技术》
CSCD
2002年第5期16-18,共3页
Measurement & Control Technology
基金
国家自然科学基金资助项目 (6993 10 10 )
