article · Multimedia Tools and Applications
High-resolution remote sensing image classification often faces difficulties because conventional convolutional neural networks struggle with multi-scale semantic relationships, long-distance feature connections, and high visual variation within identical categories. To tackle these issues, a novel model called the Residual Channel-attention network has been developed. The architecture uses a lightweight residual structure to capture spatial details across multiple scales, alongside a channel-attention mechanism that highlights useful feature channels and suppresses unhelpful ones. It also incorporates a squeeze-and-excitation component to focus on informative image characteristics while filtering out background noise. When tested on three standard public benchmark datasets, named RSSCN7, PatternNet, and EuroSAT, the model reached classification accuracies of 97 percent, 99 percent, and 96 percent, respectively. Visual assessments confirm that the approach successfully isolates critical features to deliver robust image classification.
Remote sensing imagery contains complex spatial relationships and background noise that make automated interpretation difficult. By using targeted attention mechanisms, this research improves the ability of computer vision systems to accurately identify ground scenes across diverse scales and categories, offering higher precision for systems analysing satellite and aerial data.
The model could enable automated scene recognition in software platforms that process high-resolution satellite and aerial imagery, potentially serving developers of geospatial and earth observation tools. Because the system has been tested on standard public benchmark datasets rather than deployed in operational pipelines, it represents early-stage, applied computer vision research that requires integration and validation in practical workflow environments before commercial use.
AI-generated from the published abstract. Always read the original work before citing.
Abstract High-resolution remote sensing (HRRS) image scene classification has gained increasing importance in recent years, with convolutional neural networks (CNNs) showing particular promise due to their proficiency in extracting spatial features. However, traditional CNNs face significant limitations. Specifically, they struggle to capture complex semantic relationships between objects at varying scales, and they lack the ability to effectively capture long-distance dependencies between features. This limitation is especially problematic in HRRS images, where spatial relationships and semantic content are deeply intertwined. Additionally, traditional CNNs are limited in handling substantial intra-class variation and inter-class similarity, which are common in remote sensing images. To overcome these challenges, we introduce a novel Residual Channel-attention (RCA) network for scene classification. The RCA network introduces a lightweight residual structure to better capture multi-scale spatial features and incorporates a channel attention mechanism that selectively emphasizes relevant feature channels while suppressing irrelevant ones. To further refine the focus on critical image features, we integrate a squeeze-and-excitation (SE) mechanism as a self-attention component, which helps the network prioritize the most informative features and ignore background noise. We evaluated the RCA network on three public datasets: RSSCN7, PatternNet, and EuroSAT, achieving classification accuracies of 97%, 99%, and 96%, respectively. The results demonstrate that superior of the RCA network compared to state-of-the-art strategies in remote sensing image classification. Furthermore, visualization using the Grad-CAM++ algorithm highlights the effectiveness of our channel attention mechanism and underscores the RCA network’s robust feature representation capabilities.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1007/s11042-024-20546-8
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
