Image Segmentation using Improved Genetic Algorithm
Roopa Kumari1, Neena Gupta2, Narender Kumar3
1Roopa Kumari, Research scholar of Gurukul Kangri Vishvidhalaya Haridwar, Uttrakhand, India.
2Neena Gupta*, Assistant Professor, Gurukul Kangri Vishvidhalaya Haridwar, Uttrakhand, India.
3Narender Kumar, Assistant Professor, Hemvati Nandan Bahuguna Garhwal University, Srinagar Garhwal Uttarakhand, India.
Manuscript received on September 22, 2019. | Revised Manuscript received on October 20, 2019. | Manuscript published on October 30, 2019. | PP: 1784-1792 | Volume-9 Issue-1, October 2019 | Retrieval Number: F9063088619/2019©BEIESP | DOI: 10.35940/ijeat.F9063.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Segmentation of image is a complex task. To recognize an image, segmentation is essential parts. During image segmentation, subsets of images on the basis of some features like gray levels values of pixels or position of pixels find out. This is an NP-complete problem, to find the solution to such problems is a computationally hard task. Some heuristic algorithm can be used to find out the solution to such a hard task. These algorithms find approximate solutions. Exact solution of such problems is not possible. Genetic algorithm can be considered a metaheuristic algorithm used the evolution of the population of solutions. This paper propose dan improved Genetic Algorithm that used to find multi-level thresholding segmentation of the image. The threshold value can be calculated by cumulative histogram and satisfactory result have been given by the experiments done on test images that are taken from Mnist datasets.
Keywords: Approximate Solutions, Evolution, Evolutionary Algorithm, Cumulative Histogram, Segmentation, Soft Computing, Multi-level Thresholding.