The sex ratio of the rice stem borer, Chilo suppressalis (Walker), is important for assessing reproductive potential and outbreak risk, but automated sex classification in field trap images is hindered by small targets, variable postures, subtle morphological differences and complex backgrounds. An edge–cloud collaborative system was developed for fine-grained sex classification of the rice stem borer. The edge terminal performs image acquisition, target detection, ROI extraction, and foreground enhancement, while the cloud platform conducts sex classification and result management. To improve small-target localization, YOLO-CSNet was constructed by integrating a global attention mechanism and adaptive spatial feature fusion into YOLO11n. Within detection-constrained ROIs, Haar-like features and an AdaBoost discriminator were used to suppress tray textures, shadows and non-target regions. For sex classification, DRS-ViT combines convolutional patch embedding, a discriminative-region soft-weighting module and a KANsformer encoder to represent weak sex-related cues and model global structural relationships. YOLO-CSNet achieved precision, recall, mAP@0.5, and mAP@0.5:0.95 values of 94.36%, 93.19%, 95.32%, and 73.95%, respectively. DRS-ViT achieved accuracy, precision, recall, and F1-score values of 93.93%, 94.68%, 93.29%, and 93.91%, respectively. In a temporally independent field deployment conducted at one Shandong site from May to June 2026, the system achieved an image-upload success rate of 91.7% and an end-to-end accuracy of 87.9%. The field results support the feasibility of the workflow under the tested conditions. Future work will extend field validation across multiple sites and seasons to evaluate system generalizability.

Traditional manual measurement of garlic bulb phenotypic traits is inefficient, subjective, poorly reproducible, and may cause sample damage. To improve the adaptability of three-dimensional reconstruction to garlic bulb morphology and grading-related parameter extraction, this study developed a non-destructive phenotypic measurement workflow based on multi-view image-based three-dimensional reconstruction. Four garlic materials with distinct bulb morphologies and epidermal characteristics were used to demonstrate the feasibility of the reconstruction workflow, and 40 Lanling white-skinned garlic bulbs were used for quantitative accuracy validation. Multi-view images were acquired using a high-resolution camera, a motorized turntable, and a controlled illumination system. Three-dimensional models were reconstructed using ContextCapture, and the resulting point clouds were processed in CloudCompare through cropping, denoising, downsampling, and pose correction. Maximum longitudinal diameter, maximum transverse diameter, and volume were extracted from the processed point clouds according to GB/T 45244-2025 (Grades and Specifications of Garlic) and validated against manual reference measurements. The coefficients of determination for maximum longitudinal diameter, maximum transverse diameter, and volume were 0.9935, 0.9909, and 0.9924, respectively, with RMSE values of 0.0529 cm, 0.0520 cm, and 0.8874 cm³, and MAPE values of 0.6647%, 0.7765%, and 1.9149%. Additional MAE, bias, confidence interval, and Bland–Altman analyses further supported the agreement between model-derived and manual reference measurements. These results demonstrate the feasibility of multi-view image-based three-dimensional reconstruction for non-destructive garlic bulb phenotypic measurement and provide a methodological basis for future grading-related assessment and three-dimensional phenotyping of bulbous horticultural crops.