Segmentation of large images based on super-pixels and community detection in graphs

TL;DR

This paper introduces a graph-based image segmentation framework utilizing super-pixels and community detection algorithms, which improves accuracy and speed over traditional methods, especially for large images.

Contribution

The paper presents a novel segmentation approach combining super-pixels with community detection, enabling efficient processing of large images with enhanced accuracy.

Findings

More precise segmentation than traditional methods

Faster processing of large images

Outperforms Felzenszwalb and Huttenlocher, and Arbelaez algorithms

Abstract

Image segmentation has many applications which range from machine learning to medical diagnosis. In this paper, we propose a framework for the segmentation of images based on super-pixels and algorithms for community identification in graphs. The super-pixel pre-segmentation step reduces the number of nodes in the graph, rendering the method the ability to process large images. Moreover, community detection algorithms provide more accurate segmentation than traditional approaches, such as those based on spectral graph partition. We also compare our method with two algorithms: a) the graph-based approach by Felzenszwalb and Huttenlocher and b) the contour-based method by Arbelaez. Results have shown that our method provides more precise segmentation and is faster than both of them.