MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202641018836 A) filed by Nandha Engineering College, Erode, Tamil Nadu, on Feb. 19, for 'farmeye: smart pest detection system using yolov8 and yolov11 with automated sms notification.'
Inventor(s) include Dharun Kumar K B; Jobika V; Kouslkan M; and Satheesh Kumar A.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to an intelligent pest detection and monitoring system developed for precision agriculture applications. More particularly, the invention provides an automated, real-time crop pest identification system that integrates deep learning-based object detection with threshold-driven decision support and automated farmer notification mechanisms. The system is designed to minimize manual field inspection, reduce crop losses, and optimize pesticide usage through data-driven monitoring. The proposed system comprises a high-resolution image acquisition module, such as a camera installed in agricultural fields, greenhouses, or smart farming environments. Captured crop images are processed using a deep learning object detection algorithm, specifically YOL0v8 or YOLOv II, which is trained to identify and classify various pest species. The model performs real-time inference, generating bounding boxes around detected pests along with confidence scores and class labels. A filtering mechanism is applied to eliminate low-confidence detections, thereby improving detection accuracy and reliability. Following detection, the system incorporates a pest counting module that calculates the total number of detected pest instances within a given image frame. The counted value is then compa~ed against a predefined threshold parameter stored within the system. This threshold represents the acceptable pest density limit for a specific crop type. Based on this comparison, the system categorizes infestation severity into multiple levels, such as low infestation, moderate infestation, or critical infestation. The infestation status is visually displayed through a graphical user interface, enabling easy interpretation by farmers or agricultural supervisors. In addition to visual alerts, the invention further includes an automated notification modttle configured to send Short Message Service (SMS) alerts or other digital notifications when the detected pest count exceeds the critical threshold. This real-time alert mechanism enables timely intervention measures, such as targeted pesticide spraying or biological pest control, thereby preventing large-scale crop damage. The system may also incorporate a data logging and storage module to maintain historical pest detection records. These records can be analyzed over time to identify pest trends, seasonal infestation patterns, and environmental correlations. Such data-driven insights assist farmers in implementing predictive and preventive agricultural strategies. By combining artificial intelligence-based detection with rule-based decision logic, the invention significantly enhances monitoring efficiency, reduces dependency on manual labor, and minimizes excessive pesticide application. This results in cost savings, improved crop yield, enhanced crop quality, and environmentally sustainable farming practices. The invention is adaptable to various crop types and scalable for deployment in small-scale farms as well as large agricultural fields, making it a practical and efficient solution for modem smart fam1ing systems."
Disclaimer: Curated by HT Syndication.
