Land use/land cover classification of fused Sentinel-1 and Sentinel-2 imageries using ensembles of Random Forests

TL;DR

This paper investigates the fusion of Sentinel-1 SAR and Sentinel-2 VNIR-SWIR imagery for land use/land cover classification, employing ensemble random forests with novel rotation techniques to improve accuracy and speed.

Contribution

It introduces ensemble random forests with random rotations (PCA, SRP, CRP) for fused SAR and VNIR-SWIR data, enhancing classification performance and computational efficiency.

Findings

SRP-RFE outperforms others on initial datasets with kappa ~62-68%.

CRP-RFE achieves highest accuracy on fused datasets with kappa ~96%.

Texture addition improves classification accuracy by up to 10%.

Abstract

The study explores the synergistic combination of Synthetic Aperture Radar (SAR) and Visible-Near Infrared-Short Wave Infrared (VNIR-SWIR) imageries for land use/land cover (LULC) classification. Image fusion, employing Bayesian fusion, merges SAR texture bands with VNIR-SWIR imageries. The research aims to investigate the impact of this fusion on LULC classification. Despite the popularity of random forests for supervised classification, their limitations, such as suboptimal performance with fewer features and accuracy stagnation, are addressed. To overcome these issues, ensembles of random forests (RFE) are created, introducing random rotations using the Forest-RC algorithm. Three rotation approaches: principal component analysis (PCA), sparse random rotation (SRP) matrix, and complete random rotation (CRP) matrix are employed. Sentinel-1 SAR data and Sentinel-2 VNIR-SWIR data from the IIT-Kanpur region constitute the training datasets, including SAR, SAR with texture, VNIR-SWIR, VNIR-SWIR with texture, and fused VNIR-SWIR with texture. The study evaluates classifier efficacy, explores the impact of SAR and VNIR-SWIR fusion on classification, and significantly enhances the execution speed of Bayesian fusion code. The SRP-based RFE outperforms other ensembles for the first two datasets, yielding average overall kappa values of 61.80% and 68.18%, while the CRP-based RFE excels for the last three datasets with average overall kappa values of 95.99%, 96.93%, and 96.30%. The fourth dataset achieves the highest overall kappa of 96.93%. Furthermore, incorporating texture with SAR bands results in a maximum overall kappa increment of 10.00%, while adding texture to VNIR-SWIR bands yields a maximum increment of approximately 3.45%.