Local Maximum Edge Binary Patterns for Medical Image Segmentation

In this paper, local maximum edge binary patterns (LMEBP) feature extractor is proposed for medical image segmentation. The local region of image is represented by LMEBP, which are evaluated by taking into consideration the magnitude of local difference between the center pixel and its neighbors. First, image split in to sub blocks and LMEBP features are extracted from each sub block. Once the image has been split into blocks of roughly homogeneous texture, we apply an agglomerative procedure to merge similar adjacent regions until one of the two stopping criteria is satis1ed. At each stage we merge the pair of adjacent regions which have the largest merger importance (MI) value. Based on MI the regions are merged and then form the segmented regions for medical image segmentation application. Experimental results are tested on benchmark MRI database for medical image segmentation application. Results after being investigated, proposed method shows a significant improvement for segmentation of images.