Internal Generative Mechanism Based Otsu Multilevel Thresholding Segmentation for Medical Brain Images

Recent brain theories indicate that perceiving an image visually is an active inference procedure of the brain by using the Internal Generative Mechanism (IGM). Inspired by the theory, an IGM based Otsu multilevel thresholding algorithm for medical images is proposed in this paper, in which the Otsu thresholding technique is implemented on both the original image and the predicted version obtained by simulating the IGM on the original image. A regrouping measure is designed to refining the segmentation result. The proposed method takes the predicted visual information generated by the complicated Human Visual System (HVS) into account, as well as the details. Experiments on medical MR-T2 brain images are conducted to demonstrate the effectiveness of the proposed method. The experimental results indicate that the IGM based Otsu multilevel thresholding is superior to the other multilevel thresholdings.