Determining Distance Measure in Fast Scanning Algorithm for Image Segmentation

Abstract

Segmentation is an essential and important process that separates an image into regions that have similar characteristics or features. Various algorithms have been proposed for image segmentation and this includes the Fast Scanning algorithm which has been employed on food, sport and medical images. The clustering process in Fast Scanning algorithm is performed by merging pixels with similar neighbor based on Euclidean Distance. Such an approach leads to a weak reliability and shape matching of the produced segments. This study investigates the alternatives distance measure to be employed in Fast Scanning algorithm. Distance between pixels is identified for four measures; Euclidean, City Block, Dice and Sorensen. Results show that the Sorensen is a better measure to be used in Fast Scanning algorithm for image segmentation.