RS-YOLOX: A High-Precision Detector for Object Detection in Satellite Remote Sensing Images

2022-08-30

Applied Sciences

AI Summary

To address critical challenges in satellite remote sensing imagery—namely, low detection accuracy, poor small-object recognition, and severe foreground-background class imbalance—this paper proposes RS-YOLOX. Methodologically, it integrates the ECA (Efficient Channel Attention) mechanism into the YOLOX backbone to enhance channel-wise feature responsiveness; adopts Adaptive Spatial Feature Fusion (ASFF) for adaptive multi-scale feature aggregation; employs Varifocal Loss to mitigate class imbalance and improve hard-example learning; and incorporates Slice-Assisted Hyper-Inference (SAHI) to boost small-object recall. Evaluated on three benchmark remote sensing datasets—DOTA-v1.5, TGRS-HRRSD, and RSOD—RS-YOLOX achieves state-of-the-art (SOTA) performance across all, with particularly notable gains in small-object detection and dense-scene scenarios. These results validate both the effectiveness and generalizability of the proposed framework.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Automatic object detection by satellite remote sensing images is of great significance for resource exploration and natural disaster assessment. To solve existing problems in remote sensing image detection, this article proposes an improved YOLOX model for satellite remote sensing image automatic detection. This model is named RS-YOLOX. To strengthen the feature learning ability of the network, we used Efficient Channel Attention (ECA) in the backbone network of YOLOX and combined the Adaptively Spatial Feature Fusion (ASFF) with the neck network of YOLOX. To balance the numbers of positive and negative samples in training, we used the Varifocal Loss function. Finally, to obtain a high-performance remote sensing object detector, we combined the trained model with an open-source framework called Slicing Aided Hyper Inference (SAHI). This work evaluated models on three aerial remote sensing datasets (DOTA-v1.5, TGRS-HRRSD, and RSOD). Our comparative experiments demonstrate that our model has the highest accuracy in detecting objects in remote sensing image datasets.

Problem

• Improves YOLOX for satellite image detection
• Enhances feature learning with ECA and ASFF
• Uses Varifocal Loss for balanced training samples

Innovation

• Enhanced YOLOX with ECA
• ASFF in YOLOX neck
• Varifocal Loss balancing samples