Integrating Metaheuristics in Computer Vision for Real-World Optimization Problems

11A Detailed Description on Various Techniques of Edge Detection Algorithms

Pritha A.¹* and G. Fathima^2†

¹Science and Humanities, Anna University, Chennai, India

²Department of CSE, Adhiyamaan College of Engineering, Hosur, India

Abstract

Edge detection plays a dominant part in image processing, image recognition and computer vision. This paper confers various edge detection algorithms such as Sobel, Robert, Canny and Prewitt detectors that exposes the complete performance of image information. Architecture, block diagram, accuracy parameters of edge detection prove in reducing the noise in the image and filter out unwanted information. Edge detection exactly lies on valid and definite number of contingent data. Pixels are individually defined so as to improve the quality of edges and betterment of efficiency. Less exaggeration and noise reduction maintained completely. Finally, the output of the aforementioned edge detection techniques is shown in detail and compared with the experimental results using MATLAB.

Keywords: Noise reduction, filtering, accuracy, PSNR, MSE, MATLAB

11.1 Introduction

Images consist of various figures about the entity, such as size, shape, color, and its orientation. The output of changes in light, color, shade and texture called edge. The edge detection methods is one of the major techniques in image processing. Its applications are broadly used in pattern recognition, medical image processing etc. There are two types that are Gaussian-based and ...