A self-organizing neural network for image segmentation
He Kong, Ling Guan · 2002
A new method is proposed for multiscale image segmentation. The method is based on pixel classification by means of a self organizing neural network. The core concept of this processing method is to explicitly treat segmentation as a classification problem. An unsupervised learning algorithm is utilized in the processing. Compared with other segmentation methods, the proposed one has a number of desirable features. It is self adaptive, efficient, and easy to control. The effectiveness of the proposed method is verified through several experiments.>