Utilizing a Novel Deep Learning Method for Scene Categorization in Remote Sensing Data

2025-06-28

AI Summary

Remote sensing image scene classification faces challenges including high noise levels, low-quality annotations, and insufficient feature diversity, which collectively limit model performance. To address these issues, this paper proposes a Cuttlefish Optimization–enhanced Bidirectional Recurrent Neural Network (CO-BRNN), which intelligently optimizes network parameters and gating mechanisms to significantly improve joint modeling of temporal and spatial features in complex remote sensing scenes. By integrating bio-inspired metaheuristic optimization with bidirectional sequence modeling, CO-BRNN achieves robust feature learning under limited supervision. Evaluated on standard remote sensing benchmarks, CO-BRNN attains a 97% classification accuracy—outperforming mainstream approaches such as MLP-CNN, CNN-LSTM, and LSTM-CRF. This work establishes a new paradigm for remote sensing scene understanding that is highly robust and less dependent on large-scale, high-fidelity annotated data.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Scene categorization (SC) in remotely acquired images is an important subject with broad consequences in different fields, including catastrophe control, ecological observation, architecture for cities, and more. Nevertheless, its several apps, reaching a high degree of accuracy in SC from distant observation data has demonstrated to be difficult. This is because traditional conventional deep learning models require large databases with high variety and high levels of noise to capture important visual features. To address these problems, this investigation file introduces an innovative technique referred to as the Cuttlefish Optimized Bidirectional Recurrent Neural Network (CO- BRNN) for type of scenes in remote sensing data. The investigation compares the execution of CO-BRNN with current techniques, including Multilayer Perceptron- Convolutional Neural Network (MLP-CNN), Convolutional Neural Network-Long Short Term Memory (CNN-LSTM), and Long Short Term Memory-Conditional Random Field (LSTM-CRF), Graph-Based (GB), Multilabel Image Retrieval Model (MIRM-CF), Convolutional Neural Networks Data Augmentation (CNN-DA). The results demonstrate that CO-BRNN attained the maximum accuracy of 97%, followed by LSTM-CRF with 90%, MLP-CNN with 85%, and CNN-LSTM with 80%. The study highlights the significance of physical confirmation to ensure the efficiency of satellite data.

Problem

• Improving accuracy in remote sensing scene categorization
• Addressing challenges of noise and data variety in deep learning
• Introducing CO-BRNN for superior performance over existing methods

Innovation

• Novel deep learning for remote sensing scene categorization
• Cuttlefish Optimized Bidirectional Recurrent Neural Network
• Achieves 97% accuracy surpassing existing methods