一种基于GMM和BYY后向结构的非监督图像分割

Unsupervised image segmentation based on GMM and BYY learning system having backward structure

贝叶斯阴阳和谐学习系统(BYY和谐学习系统)被广泛地应用于聚类算法中.BYY和谐学习系统有两种基本结构:具有后向结构的BYY学习系统和具有双向结构的BYY学习系统.本文将先对具有后向结构的BYY学习系统做简单的介绍,然后进行数值试验验证其聚类的有效性,最后将其应用于图像分割中.

Bayesian Yin and Yang harmony learning system （BYY harmonious learning system） is widely used inclustering algorithm. BYY harmonious learning system has two kinds of basic structures： BYY learning system having backward structure and BYY learning system having two -way structure. This review will have after a brief introduction to the BYY learning system having backward structure, and then proceed to numerical experiments to verify the effectiveness of their clustering, and finally apply it into image segmentation.