Date of Publication :15th May 2024

Abstract:Normalised Cut measures both the total dissimilarity between the different groups as well as the total similarity within the groups. But it suffers from the excessive normalization problem and weakens the small object and twig segmentation. So, we propose a novel approach for solving the perceptual grouping problem in vision. It describes an advanced approach to image segmentation using a technique is known as Explored Normalized Cut(ENCut). ENCut model establishes a balance graph model by adopting a meaningful-loop & K-step random walk, which enhances the small object segmentation. ENCut model combined with an enhancement is called extended Random Walk Refining. This approach aim to achieve more accurate and effective image segmentation result. By incorporating the Refined Random Walk term, the method explores new possibilities for optimizing the segmentation process, leading to improved outcomes in image analysis and understanding. To approach this process we are using MATLAB software version R2017b as a tool.

Reference :