PROJECT TITLE :

A New Iterative Triclass Thresholding Technique in Image Segmentation - 2014

ABSTRACT:

We gift a replacement methodology in image segmentation that's based on Otsu's methodology but iteratively searches for subregions of the image for segmentation, rather than treating the full image as an entire region for processing. The iterative methodology starts with Otsu's threshold and computes the mean values of the 2 categories as separated by the edge. Based mostly on the Otsu's threshold and therefore the two mean values, the strategy separates the image into three classes instead of two as the quality Otsu's methodology does. The first two categories are determined as the foreground and background and they can not be processed any. The third class is denoted as a to-be-determined (TBD) region that's processed at next iteration. At the succeeding iteration, Otsu's technique is applied on the TBD region to calculate a new threshold and 2 category suggests that and also the TBD region is once more separated into three categories, specifically, foreground, background, and a new TBD region, that by definition is smaller than the previous TBD regions. Then, the new TBD region is processed within the similar manner. The method stops when the Otsu's thresholds calculated between two iterations is but a preset threshold. Then, all the intermediate foreground and background regions are, respectively, combined to form the ultimate segmentation result. Tests on synthetic and real images showed that the new iterative methodology can achieve higher performance than the quality Otsu's methodology in several challenging cases, like identifying weak objects and revealing fine structures of complicated objects whereas the added computational value is minimal.