The pollination of kiwiberry flowers is closely related to fruit growth, development, and yield. Rapid and precise identification of flowers under natural field conditions plays a key role in enhancing pollination efficiency and improving overall fruit quality. Flowers and buds are densely distributed, varying in size, and exhibiting similar colors. Complex backgrounds, lighting variations, and occlusion further challenge detection. To address these issues, the YOLO11s-RFBS model was proposed. The P5 detection head was replaced with P2 to improve the detection of densely distributed small flowers and buds. RFAConv was incorporated into the backbone to strengthen feature discrimination across multiple receptive field scales and to mitigate issues caused by parameter sharing. The C3k2-Faster module was designed to reduce redundant computation and improve feature extraction efficiency. A weighted bidirectional feature pyramid slim neck network was constructed with a compact architecture to achieve superior multi-scale feature fusion with minimal parameter usage. Experimental evaluations indicated that YOLO11s-RFBS reached a mAP@0.5 of 91.7%, outperforming YOLO11s by 2.7%, while simultaneously reducing the parameter count and model footprint by 33.3% and 31.8%, respectively. Compared with other mainstream models, it demonstrated superior comprehensive performance. Its detection speed exceeded 21 FPS in deployment, satisfying real-time requirements. In conclusion, YOLO11s-RFBS enables accurate and efficient detection of kiwiberry flowers and buds, supporting intelligent pollination robots.

Accurate identification of tool wear states is crucial for ensuring machining quality and reliability. However, non-stationary signal characteristics, feature coupling, and limited use of multimodal information remain major challenges. This study proposes a hybrid framework that integrates a Sparrow Search Algorithm–optimized Continuous Wavelet Transform (SSA-CWT) with a Cross-Modal Time–Frequency Fusion Network (TFF-Net). The SSA-CWT adaptively adjusts Morlet wavelet parameters to enhance energy concentration and suppress noise, generating more discriminative time–frequency representations. TFF-Net further fuses cutting force and vibration signals through a sliding-window multi-head cross-modal attention mechanism, enabling effective multi-scale feature alignment. Experiments on the PHM2010 dataset show that the proposed model achieves classification accuracies of 100%, 98.7%, and 98.7% for initial, normal, and severe wear stages, with F1-score, recall, and precision all exceeding 98%. Ablation results confirm the contributions of SSA optimization and cross-modal fusion. External validation on the HMoTP dataset demonstrates strong generalization across different machining conditions. Overall, the proposed approach provides a reliable and robust solution for intelligent tool condition monitoring.

Accurate and efficient detection of hot forging die surface defects is essential for die condition assessment and stable forging production. However, automated visual inspection remains challenging because of large variations in defect scale and morphology, complex metallic textures and reflections, and weak fine-grained features. To address these challenges, we propose MCDE-DETR, a lightweight real-time detector based on RT-DETR-r18. Specifically, a Multi-Scale Edge Enhancement Network (MSENet) is designed as the backbone to strengthen multi-scale and edge-detail representation while reducing model complexity. Then, CGAFusion is introduced to enhance defect-related features and suppress background interference. DPCF is employed to preserve fine-grained information during cross-scale feature fusion. Finally, EDFFN is integrated into AIFI to improve frequency-aware feature representation with limited computational overhead. Experiments on a factory-collected dataset show that MCDE-DETR achieves 72.5% mAP50 and 46.6% mAP50:95, improving the baseline by 3.0 and 1.8 percentage points, respectively, while reducing parameters and GFLOPs by 25.1% and 10.4%. The model reaches 80.69 FPS, and Jetson Orin Nano deployment further demonstrates its practical feasibility for edge-based inspection.