Fusion of Image Segmentation Algorithms using Consensus Clustering

TL;DR

This paper introduces a novel segmentation fusion method that combines multiple algorithms' outputs into a consensus segmentation using a stochastic optimization approach, improving accuracy and automatically determining the optimal number of clusters.

Contribution

It reformulates the Filtered Stochastic BOEM algorithm as a segmentation fusion problem with a new distance learning approach and integrates automatic cluster number determination.

Findings

Effective fusion of segmentation algorithms on remote sensing images.

Automatic determination of the optimal number of clusters.

Improved segmentation accuracy through consensus clustering.

Abstract

A new segmentation fusion method is proposed that ensembles the output of several segmentation algorithms applied on a remotely sensed image. The candidate segmentation sets are processed to achieve a consensus segmentation using a stochastic optimization algorithm based on the Filtered Stochastic BOEM (Best One Element Move) method. For this purpose, Filtered Stochastic BOEM is reformulated as a segmentation fusion problem by designing a new distance learning approach. The proposed algorithm also embeds the computation of the optimum number of clusters into the segmentation fusion problem.