Few-shot object detection methods perform well in natural scenes, where meta-learners can effectively extract target features from limited support data. However, PCB defect detection faces unique challenges: scarce defect samples, low-resolution targets, and severe overfitting risks. Additionally, PCBs’ dense circuitry creates low-contrast defects that blend into background noise, while traditional meta-learning struggles to generate realistic synthetic defects under actual manufacturing constraints. To overcome these limitations, we propose SM-FSOD, a second-order meta-learning model featuring a defect-aware prototype network. Unlike conventional approaches, it dynamically emphasizes critical defect features when constructing class prototypes from few-shot samples. Our extensive few-shot experiments on the DeepPCB and DsPCBSD+ datasets demonstrate the performance of the SM-FSOD model. Comparative tests on the DeepPCB dataset show that, compared with the strong Meta-RCNN baseline, our model achieves a maximum performance improvement of 14.0% under the challenging five-shot setting, while still attaining a 7.6% accuracy gain in the more relaxed 50-shot setting. Similarly, evaluation on the DsPCBSD+ dataset reveals that our proposed method maintains an average accuracy improvement of 2.3% to 9.6% compared to the competitive DeFRCN model in complex scenarios, indicating the strong adaptability of SM-FSOD across various application environments. Ablation studies further demonstrate that incorporating the improved MOMP and DGPT modules individually yields average accuracy gains of 3.6% and 4.5%, respectively, under the five-shot setting compared to the baseline, confirming that these enhancements can orthogonally improve the detection precision in few-shot PCB scenarios.

Existing research has demonstrated that meta-learning methods hold considerable promise in addressing the challenges posed by few-shot object detection. However, remote sensing scenarios present two major challenges. The sparse features of small objects provide insufficient support information for query enhancement, and significant morphological variations caused by lighting and viewpoint differences hinder intra-class consistency capture via direct alignment in few-shot learning. To address these challenges, we propose a generative meta-learning detection framework. The framework first introduces a Dynamic Relation Dual-Stream Network to achieve dynamic support-query feature alignment through joint modeling of evolutionary and relational features, thereby enhancing representation in few-shot conditions. Second, an Optimal Transport-based Generative Meta-Learner is developed to mitigate feature distribution bias via generative augmentation in latent space. Additionally, an Orthogonal Frequency Decomposition Head is incorporated to adaptively separate query features into low-frequency contour and high-frequency detail components, effectively suppressing background noise interference. Experiments on multiple remote sensing datasets demonstrate that the proposed method achieves consistent performance gains over leading baseline methods in various few-shot settings. Its effectiveness is further validated across different backbone networks, highlighting strong generalization in few-shot remote sensing object detection.

Few-shot object detection suffers from limited annotations, redundant background interference, insufficient feature interaction, and severe sample imbalance. Existing meta-learning-based methods extract class prototypes from support images but often fail to effectively suppress background noise or align channel-wise features between support and query branches. To address these issues, we propose a few-shot object detection method based on background reconstruction and multi-channel interactive feature fusion. First, a background reconstruction module is designed to suppress redundant background interference by applying random region masking to support set images, thereby generating robust class prototype features that are resistant to background noise. Second, a multi-channel interactive feature fusion module is designed, which leverages depthwise separable convolution to enable effective channel-wise feature interaction and information alignment between support class prototypes and query features, thereby enhancing cross-branch feature interaction and fusion. Finally, to address the uneven sample distribution and the foreground–background imbalance in few-shot scenarios, we proposed a category-aware weighted loss. By appropriately weighting the contributions of different object categories and background samples, the proposed loss encourages balanced optimization, resulting in faster convergence and improved detection performance. Experimental results demonstrate that the proposed method improves detection accuracy and generalization performance under few-shot settings. On the Pascal VOC dataset (Split1), the proposed method achieves 45.1%, 62.9%, and 67.1% under 1-shot, 3-shot, and 10-shot settings, respectively, outperforming the baseline; consistent improvements are also observed on the MS COCO dataset and the DIOR dataset.