A new neural network model based approach to unsupervised image segmentation

This paper proposes a new neural network model UMAN in which the generalized information entropy is used as the quantitative description and measurement of the system stability and asymptotication, and the disadvantage of generalized energy functions is avoided. The improved Kohonen nonlinear mapping structure not only enhances the clustering features, but also reduces the redundant information. In the network, the internal layer and node number are determined dynamically by the system. The unsupervised self-learning function expresses the characteristics of low level visual information processing. The UMAN model could process various types of images and has strong adaptability. Experimental results show that the model and its algorithm are efficient, practical and robust