Systematic review of image segmentation using complex networks

TL;DR

The paper addresses the challenge of segmenting complex images by leveraging complex networks and community-detection principles. It surveys traditional modularity- and graph-based methods as well as modern spectral, deep-learning–augmented, and hybrid approaches, highlighting their applicability to irregular shapes and textured regions. The key contributions lie in synthesizing a broad range of network-based segmentation frameworks, outlining their clinical and visual-scene applications, and discussing evaluation metrics and limitations. The findings point to significant potential for improved segmentation accuracy and robustness through integration of contextual information and multi-layer networks.

Abstract

This review presents various image segmentation methods using complex networks. Image segmentation is one of the important steps in image analysis as it helps analyze and understand complex images. At first, it has been tried to classify complex networks based on how it being used in image segmentation. In computer vision and image processing applications, image segmentation is essential for analyzing complex images with irregular shapes, textures, or overlapping boundaries. Advanced algorithms make use of machine learning, clustering, edge detection, and region-growing techniques. Graph theory principles combined with community detection-based methods allow for more precise analysis and interpretation of complex images. Hybrid approaches combine multiple techniques for comprehensive, robust segmentation, improving results in computer vision and image processing tasks.