Camera localization aims to estimate the six-degree-of-freedom camera pose from a single RGB image and is essential for robotics, autonomous navigation, and augmented reality. Despite recent progress in learning-based localization, robust pose estimation remains challenging under large viewpoint changes, occlusions, repetitive textures, and complex scene layouts. Existing methods mainly rely on local feature matching or independent feature regression, while the structural dependencies among multi-scale visual representations are not fully explored. To address this issue, we propose Uncertainty-aware Bidirectional Graph Learning (UBGL) for RGB-based camera localization. UBGL extracts hierarchical visual features and converts them into graph-structured representations, where nodes describe scene information at different semantic levels. A bidirectional graph learning module is introduced to exchange information between low-level geometric cues and high-level contextual features, enabling joint modeling of local correspondences and global scene structure. In addition, an uncertainty-aware relation modeling strategy estimates the reliability of graph connections and helps reduce the influence of unstable feature interactions during graph reasoning. The refined graph representations are then projected into dense scene-coordinate maps, and the final camera pose is recovered using a differentiable pose solver. Experiments on the 7Scenes and Cambridge Landmarks datasets show that UBGL achieves competitive localization accuracy, especially improving rotation estimation on the 7Scenes benchmark while maintaining comparable performance on Cambridge Landmarks. Ablation studies further demonstrate the effectiveness of bidirectional graph interaction and uncertainty-aware relation modeling.

Ultraviolet (UV) insect traps enable continuous field pest monitoring, but their images contain severe scale variation, dense and overlapping targets, incomplete specimens, background stains, insect fragments, non-target insects, inter-class similarity, and long-tailed distributions, leading to missed detections, false positives, and class confusion. To address these challenges, we constructed a real-world UV insect-trap pest dataset comprising 12,426 high-resolution images and 38,003 annotated instances from 36 pest categories, and propose FCD-DETR, a foreground-aware and context-enhanced Detection Transformer based on RT-DETR-R18. The main distinction of FCD-DETR lies in a progressive enhancement design tailored to UV insect-trap images, where FADM first reduces foreground–background confusion by decoupling pest foreground cues from multi-scale semantic information, C2f_CFBlock strengthens local-detail and contextual representations in the lightweight backbone, and DHSA further alleviates high-level semantic ambiguity through dynamic-range histogram self-attention in the Transformer encoder. On the proposed dataset, FCD-DETR achieved 64.42% mAP@0.5 and 41.33% mAP@50:95, improving the baseline by 5.25 and 5.11 percentage points, respectively, and outperforming representative CNN- and Transformer-based detectors. Comprehensive experimental analyses confirm that FCD-DETR improves foreground discrimination and detection robustness in complex UV insect-trap scenarios.