Techniques for Generation of Random Fields for Image Segmentation

The book consists of image processing in particular related to image segmentation. The book depicts four issues. These issues related to segmentation of low contrast images, removal of shadow in the images, reduction of high dimensional images, and computational complexity of segmentation techniques. To address these issues, the author proposed OT-FO, GRF-B, GORF-LLE-ML, and H-ML technique. The results demonstrated that the H-ML approach obtained an accurate segmentation outcome on low contrast, shadow, high dimensional images, and different kinds of noises with reduced computational time. In the study of the performance of the proposed methods, the author has conducted experiments using benchmark databases like Berkely dataset (BSD500), Semantic dataset and author own dataset, etc., and has made a study of mean, standard deviation, Jaccard index, correlation, dice coefficient, segmentation accuracy, and time complexity