A Novel Approach Towards Clustering Based Image Segmentation

TL;DR

This paper proposes a new clustering-based image segmentation method that uses LAB color space, cosine distance with K-means, Sobel filtering, and watershed algorithm, showing promising performance metrics.

Contribution

It introduces a novel combination of clustering, color space selection, and filtering techniques for improved image segmentation.

Findings

Uses LAB color space for better segmentation.

Employs cosine distance in K-means for improved clustering.

Evaluates performance with MSE and PSNR metrics.

Abstract

In computer vision, image segmentation is always selected as a major research topic by researchers. Due to its vital rule in image processing, there always arises the need of a better image segmentation method. Clustering is an unsupervised study with its application in almost every field of science and engineering. Many researchers used clustering in image segmentation process. But still there requires improvement of such approaches. In this paper, a novel approach for clustering based image segmentation is proposed. Here, we give importance on color space and choose lab for this task. The famous hard clustering algorithm K-means is used, but as its performance is dependent on choosing a proper distance measure, so, we go for cosine distance measure. Then the segmented image is filtered with sobel filter. The filtered image is analyzed with marker watershed algorithm to have the final segmented result of our original image. The MSE and PSNR values are evaluated to observe the performance.