An Improved DETR Based on Angle Denoising and Oriented Boxes Refinement for Remote Sensing Object Detection
Open Access
- 26 November 2024
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 16 (23) , 4420
Abstract
Remote sensing image object detection presents significant challenges, due to the difficulty in accurately predicting the rotational angles of ground-oriented objects, coupled with issues such as the false and missed detection caused by insufficient object information. Moreover, traditional convolutional neural networks are inherently limited in their capacity to capture global contextual information. To address these challenges, a DETR-based remote sensing image object detection model is designed for oriented objects. Except for the backbone, transformer encoders and decoders, scenario query guiding modules, oriented boxes refinement modules, auxiliary multiple detectors, and oriented boxes denoising modules are also designed and included in our network. The scenario query guiding module is proposed that implicitly guides the decoder to focus more on object classification information specific to that scene when inferring. The multiple deformable attention mechanism is improved to the oriented one and utilized into the oriented boxes refinement module which repeatedly corrects the oriented boxes, enhancing the network’s ability to predict the oriented boxes precisely. At the same time, the improved auxiliary multiple detectors and oriented boxes denoising module are applied only for the training process to enhance the learning ability of the encoder and decoder for oriented objects. The ablation experiments proved the effectiveness of the designed modules. The detection accuracy of our network on DOTAv1.0 (76.77%) and HRCS2016 (97.01%) is improved compared with the state-of-the-art methods, which are especially significantly higher than DETR detection algorithms.Keywords
Funding Information
- National Natural Science Foundation of China (62376224, 2023-YBGY-232)
- Key Research and Development Program of Shaanxi Province (62376224, 2023-YBGY-232)
This publication has 38 references indexed in Scilit:
- S2ANet: Combining local spectral and spatial point grouping for point cloud processingVirtual Reality & Intelligent Hardware, 2024
- BiF-DETR:Remote sensing object detection based on Bidirectional information fusionDisplays, 2024
- Multistage Enhancement Network for Tiny Object Detection in Remote Sensing ImagesIEEE Transactions on Geoscience and Remote Sensing, 2024
- FFCA-YOLO for Small Object Detection in Remote Sensing ImagesIEEE Transactions on Geoscience and Remote Sensing, 2024
- ASNet: Adaptive Semantic Network Based on Transformer–CNN for Salient Object Detection in Optical Remote Sensing ImagesIEEE Transactions on Geoscience and Remote Sensing, 2024
- RoMP-transformer: Rotational bounding box with multi-level feature pyramid transformer for object detectionPattern Recognition, 2023
- MashFormer: A Novel Multiscale Aware Hybrid Detector for Remote Sensing Object DetectionIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023
- Improved Swin Transformer-Based Semantic Segmentation of Postearthquake Dense Buildings in Urban Areas Using Remote Sensing ImagesIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022
- Anchor-Free Oriented Proposal Generator for Object DetectionIEEE Transactions on Geoscience and Remote Sensing, 2022
- Align Deep Features for Oriented Object DetectionIEEE Transactions on Geoscience and Remote Sensing, 2021