Image Edge-Segmentation Techniques : A Review

Abstract

Image segmentation is commonly applied technique in different domains such as automatic pattern recognition, image retrieval based content, machine vision, face detection, medical imaging, and object detection. Image segmentation involves classifying or identifying sub patterns in a given image. Many of algorithms and techniques for image segmentation have been proposed to optimize segmentation problems in a specific application area. In this work, different image segmentation techniques had been applied (threshold based, region based segmentation and edge based preserving methods. This Experiment have been done using MATLAB R2018b. Different edge detection methods such as Sobel, Prewitt, Roberts, Laplacian, Kiresh and Canny methods are performed on the benchmark image and the performance is analyzed with respect to the standard measure peak signal-to-noise ratio (PSNR), and mean square error. The results present that the Laplacian method is more effective than the other methods.