Answers are grounded in this paper and cite their sources.
Early Shoot Borer (ESB) pests attack sugarcane crops and cause yield loss. An ESB framework was proposed to identify and control the ESB on time with expert guidance to reduce the yield loss. An expert-validated custom ESB dataset was created for four ESB symptoms (deadheart, borehole, tillers, and larva). The ESB framework integrates the fine-tuned YOLOv8m-seg, YOLOv8m object-detection, and TabNet models. The image-based deep-learning model decisions were fused with the ESBTabNet model. An interactive mobile application was developed that enables farmers to contact experts to obtain recommendations and control ESB attacks on time. The ESB-Deadheart model achieved 90.25% and 96.58% of the mAP for mask and bounding box detections, respectively. The ESB-Tillers, ESB-Borehole, and ESB-Larva models achieved 98.93%, 96.59%, and 88.40% of mAP for bounding box detections, respectively. The TabNet model achieved an average precision of 92.50%. The ESB control mobile application’s end-to-end delay, success rate, and effective throughput were tested at 10 different geographical locations, and their average values were recorded as 0.08s, 2.43%, 91.92 Mbps on 5G, and 0.07s, 4.95%, 87.14 Mbps on 4G, respectively. The mobile application helps farmers purchase suitable insecticides with the required quantity and saves soil, cane health, and cost. The proposed ESB framework can improve yield by controlling ESB attacks over time. Impacts: On-time expert recommendations can guide farmers in improving sugarcane yield and quality by controlling ESB attacks. Expert recommendations about insecticides, such as suitable type and correct quantity, help farmers save cane health, soil health, and cost.
No takes yet. Share an insight, caveat, or question.
S et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: