The extensive cultivation scale of sugar beet seedlings has resulted in the necessity for accurate identification and monitoring of the seedling count, a task which has become crucial and highly challenging in the sugar industry. However, sugar beet seedlings in UAV aerial photography scenarios are mostly small targets with complex backgrounds. Existing general detection models not only have insufficient detection accuracy, but also struggle to balance computational efficiency and resource consumption. To meet the practical needs of field monitoring, this paper proposes the LDH-RTDETR, a sugar beet seedling detection model that balances high accuracy and light weight. This model uses LSNet for feature extraction to reduce size, adds a deformable attention (DAttention) module to capture fine-grained seedling features, and adopts HS-FPN to improve multi-scale feature fusion in the neck network. Experimental results show that the improved model significantly outperforms the original RT-DETR model, with a 3.6% increase in accuracy, a 2.1% increase in mAP50, a recall rate of 86.0%, and a final model size of only 43.3 MB, thus achieving an effective balance between accuracy and model size. This study’s improved model offers an efficient solution for large-area identification and counting of sugar beet seedlings, and is highly significant for advancing the automation of sugar crop field management and agricultural digital transformation.

Accurate 3D semantic perception is essential for robotic operations in multi-tier caged-pigeon houses, where dense cage-mounted facilities, repetitive layouts, local geometric similarity, and dynamic objects make static-facility segmentation difficult. To address these issues, this study proposes GSR-PointNet++, a point cloud semantic segmentation method with dynamic suppression. A Hierarchical Dynamic Suppression Front-End (HDS-Front) is introduced before static map construction to reduce dynamic interference caused by pigeon activity and human movement. Based on the processed static point clouds, a semantic segmentation dataset for multi-tier caged-pigeon houses is constructed. GSR-PointNet++ is developed on the PointNet++ backbone and incorporates a Geometric Structure-Relationship Module (GSRM), which models local normal consistency and spatial relationships to enhance feature representation in structurally repetitive scenes. Geometry-aware Prototype Consistency Learning (GPCL) is further introduced during training to improve class separability and prediction stability. Independent testing on the annotated dataset collected from the second pigeon house showed that the proposed method achieved a mean intersection over union (mIoU) of 96.29% ± 0.25% and an overall accuracy (OA) of 98.62% ± 0.13% for the four target facility classes, outperforming PointNet++ by 5.92 and 2.54 percentage points, respectively. These results indicate that the proposed method improves point cloud semantic segmentation for static facilities and has potential to support robotic perception in structured livestock environments.

In modern intelligent food processing, the unpredictable variability in pomelo orientation on high-speed conveyors poses a significant challenge to automated grading and precision peeling operations. To address this, a deep learning-based method is proposed for the real-time detection of pomelo posture. Firstly, a pomelo posture dataset was constructed to support model training and validation. Secondly, to balance the extraction of posture features from uniform fruits with the low-power constraints of edge deployment, a domain-specific architectural optimization is presented. Building on the YOLOv8n framework, the proposed model synergistically integrates specialized modules. A lightweight GhostHGNetV2 foundation is utilized to significantly reduce computational redundancy while maintaining the resolution required to detect key anatomical landmarks. To overcome spatial confusion and capture multi-scale global appearance information, a multi-path coordinate attention (MPCA) module is introduced. Furthermore, the SlimNeck architecture and VoVGSCSP module streamline multi-scale feature fusion via one-time aggregation, effectively preventing computational bottlenecks. This design optimizes the computational efficiency of the model while maintaining detection accuracy. Experimental results demonstrate that compared with the baseline YOLOv8n model, the proposed method increased the mAP50 accuracy by 3.67% while reducing parameter count and computational load by 17.5% and 23.3%, respectively. Additionally, it achieved a processing speed of 19.3 FPS on the Jetson Orin Nano 6G edge platform. This research provides a critical technical foundation for the recognition of pomelo posture, enabling subsequent orientation rectification and fostering the development of streamlined, automated pomelo processing lines.