Assessing CNN and Semantic Segmentation Models for Coarse Resolution Satellite Image Classification in Subcontinental Scale Land Cover Mapping
Open Access
- 27 September 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract
Based on studies using high-medium resolution images, convolutional neural networks (CNNs) and semantic segmentation have shown superiority over classical machine learning (ML), particularly in small-scale mapping. However, few/no studies have assessed the techniques on coarse resolution image classification for extensive area land cover mapping. In this study, we evaluated the performance and feasibility of three CNN models (1D CNN, 2D CNN, and 3D CNN), and U-net for coarse-resolution satellite image classification and compared them to a random forest (RF) classifier. We utilized time-series, coarse resolution (1 km) composite imageries acquired by FY-3C VIRR. Labeled datasets were collected as shapefiles and split into three independent datasets: training, validation, and test datasets, and preprocessed to meet each model's input format requirements. We conducted several experiments to optimize models and select the best models. Then, the best models were evaluated on an unseen dataset. Among the DL models, 1D CNN achieved the highest overall accuracy (OA) 0. 87 and kappa (k) 0.84, 2% higher than the best results attained by 2D CNN, 3D CNN, and U-net models. However, 1D CNN is outperformed by RF which achieved 0.89 (OA) and 0.87 (k). Achieving the best and the second-best results using RF and 1D CNN models, respectively, indicates the superiority of the pixel-based method and the insignificance of spatial information in coarse-resolution image classification. Furthermore, although the DL models can yield high accuracy, especially 1D CNN, they are less feasible than RF classifiers for coarse-resolution satellite image classification in extensive area land cover mapping.Keywords
Funding Information
- National Key R&D Program (2021YFE0117400)
This publication has 65 references indexed in Scilit:
- Optimised U-Net for Land Use–Land Cover Classification Using Aerial PhotographyPFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 2023
- Fine hyperspectral classification of rice varieties based on attention module 3D-2DCNNComputers and Electronics in Agriculture, 2022
- Convolutional Neural Networks-An Extensive arena of Deep Learning. A Comprehensive StudyArchives of Computational Methods in Engineering, 2021
- On the Evaluation of CNN Models in Remote-Sensing Scene Classification DomainPFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 2020
- Deep learning in remote sensing applications: A meta-analysis and reviewISPRS Journal of Photogrammetry and Remote Sensing, 2019
- Algorithms for semantic segmentation of multispectral remote sensing imagery using deep learningISPRS Journal of Photogrammetry and Remote Sensing, 2018
- Deep Learning with Python2017
- Random forest in remote sensing: A review of applications and future directionsISPRS Journal of Photogrammetry and Remote Sensing, 2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Kernel Methods for Remote Sensing Data Analysis2009