As Pulse Coupled Neural Network(PCNN) in image segmentation has to adjust the network parameters by manual operation for several times, this paper proposed a automatic image segmentation method based on the PCNN. The effect factors on setting PCNN parameters in an image were analyzed and an adaptive segment protocol was invited to divide the images into several sub-pieces with the similar inside complexities. By which, the weakness that the same parameter could not segment exactly the region with quite different complexities in an image at the same time was overcome. Furthermore, the index of Activity Degree of Local Area(ADLA) proposed by the paper was used to determine the PCNN model parameters for different sub-pieces adaptively and to avoid the manual operation for important parameter selection in the traditional PCNN image segmentation. Finally, the best result was chosen from the binary-result sequences with the max two-dimension Tallis entropy protocol. The experiments proved that segmented figure is clear, complete and has excellent performance, even in conditions of a low contrast or a changeable background. Compared with the traditional PCNN segmentation methods, the proposed method not only can determine the PCNN model parameters automatically, and its indexes on quantitative evaluation of the segmentation result, such as Uniformity Measure(UM), Regional Contrast (CR), Shape Measure (SM) and Comprehensive Index (CI) are all 12% better than those of the traditional PCNN method.