Remote Sensing Image Classification Using Deep Ensemble Learning

2026-03-06

AI Summary

This work addresses the challenge in remote sensing image classification where single models struggle to simultaneously capture local details and global contextual information, and naive stacking of CNNs and Vision Transformers (ViTs) often leads to feature redundancy and performance bottlenecks. To overcome this, the authors propose a multi-model ensemble architecture that integrates CNN and ViT through four complementary fusion models, followed by decision-level ensembling to effectively mitigate redundancy and enhance discriminative capability. The proposed method achieves state-of-the-art accuracy rates of 98.10%, 94.46%, and 95.45% on the UC Merced, RSSCN7, and MSRSI datasets, respectively, significantly outperforming existing mainstream approaches while maintaining computational efficiency and delivering higher classification accuracy.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Remote sensing imagery plays a crucial role in many applications and requires accurate computerized classification techniques. Reliable classification is essential for transforming raw imagery into structured and usable information. While Convolutional Neural Networks (CNNs) are mostly used for image classification, they excel at local feature extraction, but struggle to capture global contextual information. Vision Transformers (ViTs) address this limitation through self attention mechanisms that model long-range dependencies. Integrating CNNs and ViTs, therefore, leads to better performance than standalone architectures. However, the use of additional CNN and ViT components does not lead to further performance improvement and instead introduces a bottleneck caused by redundant feature representations. In this research, we propose a fusion model that combines the strengths of CNNs and ViTs for remote sensing image classification. To overcome the performance bottleneck, the proposed approach trains four independent fusion models that integrate CNN and ViT backbones and combine their outputs at the final prediction stage through ensembling. The proposed method achieves accuracy rates of 98.10 percent, 94.46 percent, and 95.45 percent on the UC Merced, RSSCN7, and MSRSI datasets, respectively. These results outperform competing architectures and highlight the effectiveness of the proposed solution, particularly due to its efficient use of computational resources during training.

Problem

• Remote Sensing Image Classification
• Convolutional Neural Networks
• Vision Transformers
• Feature Redundancy
• Performance Bottleneck

Innovation

• Deep Ensemble Learning
• CNN-ViT Fusion
• Remote Sensing Image Classification
• Feature Redundancy Mitigation
• Vision Transformers