Precise, non-destructive detection of fruit maturity is a cornerstone of modern precision agriculture, directly impacting harvest scheduling and post-harvest quality control. In the case of strawberries (Fragaria × ananassa), in-field automated assessment is persistently hampered by the fruit’s diminutive size, subtle physiological colour transitions, and frequent occlusion by foliage. To overcome these limitations, we developed SMLO-YOLO, a specialised lightweight vision system designed to reliably detect different maturity stages on edge devices under complex orchard conditions. The proposed architecture incorporates a Cross-Scale Aggregation Neck (HDP-Neck) driven by entropy-guided dynamic sampling, which effectively concentrates computational resources on fruit regions while filtering background noise. Additionally, we introduce a Shape-aware Intersection-over-Union (ShapeIoU) loss and a Boundary- and Class-aware Knowledge Distillation (BCKD) strategy to specifically address the challenge of detecting overlapping clusters and low-maturity fruits. Validation on custom datasets collected from commercial orchards in Sichuan and Shanxi demonstrated that the final SMLO-YOLO model, after BCKDloss-based knowledge distillation, achieved an mAP50 of 92.4% at an inference speed of 256.41 FPS, with 6.49 M parameters and 15.0 GFLOPs. These metrics indicate that the system successfully balances high-throughput detection with the non-harvestable low-maturity fruits of agricultural robotics, offering a robust tool for objective, real-time maturity monitoring.

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.

Reliable multi-object tracking of fish is essential for individual-level behavior analysis and intelligent aquaculture monitoring. However, degraded underwater imagery, ambiguous boundaries in dense groups, and high inter-individual appearance similarity often lead to missed detections, association errors, and identity switches. We propose BF-DETR-DAPS-BoT-SORT, a tracking-by-detection framework integrating fish detection, identity representation, and trajectory association. BF-DETR incorporates a Fish Morphology Enhancement Module (FMEM) to model local texture, directional structure, and contextual information, and a Boundary-Contrast Separation Module (BCSM) to distinguish adjacent individuals. We further develop the Detail-Aware Part Similarity Network (DAPS), which combines underwater texture enhancement, dual-axis local representations, and part-similarity learning to capture subtle appearance differences under changes in swimming direction. DAPS features are integrated into BoT-SORT for data association. On task-specific datasets constructed from continuous underwater videos, BF-DETR achieved mAP@0.5 and mAP@0.5:0.95 values of 96.63% and 77.53%, respectively. The complete framework achieved MOTA, HOTA, and IDF1 scores of 93.1%, 75.8%, and 91.2%, respectively, with 13 identity switches. Compared with BoT-SORT, HOTA and IDF1 increased by 1.4 and 2.1 percentage points, while identity switches decreased from 18 to 13. These results indicate improved trajectory continuity and identity consistency in complex underwater scenes, supporting reliable individual-level fish behavior analysis.