A Geometry-Conditioned Symmetry-Aware Domain-Robust Observation Correction Front-End for Anti-UAV Visual Perception
Open Access
Abstract
Although the bounding boxes produced by an object detector provide real-time target localization cues for anti-UAV visual perception, they remain susceptible to geometric deviations under long-range small-target conditions, complex backgrounds, motion blur, and cross-domain environmental variations. Consequently, these detector outputs cannot always serve directly as stable observations for state estimation, trajectory prediction, and interception control. To address this issue, this paper proposes CDBR-Net, a causally conditioned domain-robust observation correction network for post-detection UAV bounding-box refinement. CDBR-Net employs a shared encoder, disentangled multi-branch representations, and a quality-aware gating mechanism to jointly produce a corrected observation box, a robust representation, and an observation uncertainty estimate. To preserve this conditional environment-transformation symmetry without suppressing geometry-induced symmetry breaking, CDBR-Net constructs geometry-conditioned cross-domain sample pairs and imposes cross-domain consistency and geometry-sensitivity preservation constraints. After training on 8749 post-detection observations, CDBR-Net is evaluated on 997 aligned validation observations from three simulated scene domains. It reduces the YOLO bounding-box mean absolute error (MAE) by 6.8%, from 0.002924 to 0.002725, and increases the intersection over union (IoU) by 0.007979, from 0.852167 to 0.860146. It further reduces MAE by 1.7% and increases IoU by 0.002031 relative to the YOLO + MLP Residual baseline. Ablation studies demonstrate the complementary roles of geometry-conditioned cross-domain consistency and geometry-sensitivity preservation. These results indicate that CDBR-Net provides a more stable and geometrically consistent post-detection observation interface for anti-UAV visual perception.Keywords
Funding Information
- National Natural Science Foundation of China (62473216)
- Natural Science Foundation of Nantong Municipality (JC2023006)
- Natural Science Foundation of Nantong Municipality (JC2025059)
This publication has 25 references indexed in Scilit:
- Learning adaptive detection and tracking collaborations with augmented UAV synthesis for accurate anti-UAV systemExpert Systems with Applications, 2025
- Anti-UAV detection and identification technology: Fundamentals, methods and challengesPhysical Communication, 2025
- MMFW-UAV dataset: multi-sensor and multi-view fixed-wing UAV dataset for air-to-air vision tasksScientific Data, 2025
- Vision-Based Drone Detection in Complex Environments: A SurveyDrones, 2024
- A Survey on Vision-Based Anti Unmanned Aerial Vehicles MethodsDrones, 2024
- Towards better small object detection in UAV scenes: Aggregating more object-oriented informationPattern Recognition Letters, 2024
- Uncertainty Regularized Evidential RegressionProceedings of the AAAI Conference on Artificial Intelligence, 2024
- Multi-Step Temporal Modeling for UAV TrackingIEEE Transactions on Circuits and Systems for Video Technology, 2024
- Toward Causal Representation LearningProceedings of the IEEE, 2021
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016