Assessments Image Segmentation Using Genetic Algorithm

Abstract

 Image segmentation is a crucial technique for processing images. It is a challenging task to
process images, and the quality of the segmentation process affects the following assignments, which
include classification, object recognition, feature extraction, and object detection. It's a significant phase
of a system for computer vision. Image segmentation is the basic problem in many applications for image
processing. Over time, image segmentation has gotten more challenging due to its extensive use in
numerous applications. It is the procedure of segmenting the image into various areas by using a specific
technique. There are many different ways for image segmentation. A new information parameter with a
threshold basis for segmenting images using the genetic algorithm. Due to its ability to calculate the ideal
number of segmentation regions, we employed the Genetic Algorithm. In this work, a novel approach built
upon a genetic algorithm is used to solve the image segmentation problem by utilizing the thresholding
concept. The suggested method uses a genetic algorithm to identify the evolutionary best segmented image
based on a threshold that is based on new information. We presented the results of our experiments using
the suggested method on various grayscale images in the last section. By using parameters used to
evaluate image segmentation quality (PSNR, MSE, SC), we notice the results are good.