Highly-Performed Fuzzily-logicized Edge Detecting Algorithm for Noisy Handwritings

  • Sami Hasan College of Information Engineering, Al-Nahrain University Baghdad, Iraq
  • Shereen S. Jumaa Al-Farahidi University, Baghdad, Iraq
Keywords: Classical edge detection, Edge detection, Fuzzy logic, Image noise, Image processing

Abstract

The main targets for using the edge detection techniques in image processing are to reduce the number of features and find the edge of image based-contents. In this paper, comparisons have been demonstrated between classical methods (Canny, Sobel, Roberts, and Prewitt) and Fuzzy Logic Technique to detect the edges of different samples of image's contents and patterns. These methods are tested to detect edges of images that are corrupted with different types of noise such as (Gaussian, and Salt and pepper). The performance indices are mean square error and peak signal to noise ratio (MSE and PSNR). Finally, experimental results show that the proposed Fuzzy rules and membership function provide better results for both noisy and noise-free images.

Published
2021-01-14
How to Cite
Sami Hasan, & Shereen S. Jumaa. (2021). Highly-Performed Fuzzily-logicized Edge Detecting Algorithm for Noisy Handwritings. Iraqi Journal of Science, 198-206. https://doi.org/10.24996/ijs.2021.SI.1.28