A New Image Segmentation Method Based on Modified Intersecting Cortical Model

The Intersecting Cortical Model (ICM) was derived from several visual cortex models, which can be applied to image segmentation efficiently. However, the performance of the segmentation greatly depends on the appropriate model parameters and the cyclic iteration times. Therefore it is necessary to adjust the ICM parameters with different images and manually select the best result from the iteration output sequences. This paper presents a self-adaptive segmentation method based on a modified ICM (SICM), which can set the parameters adaptively by using the characteristics of the image to be segmented. And the optimal segmentation result is determined by the maximum Mutual Information (MI) between the original and the segmented image. The experimental results show that the SICM has visually better segmentation, and the comprehensive evaluation value of the SICM increases by approximately 15 percent compared with that of the fuzzy C-means algorithm.