EDGE-INTELLIGENT WALL-MOUNTED IOT SURVEILLANCE SYSTEM FOR REAL-TIME STRAWBERRY DISEASE DETECTION
Abstract
Plant diseases need to be identified early to help in safeguarding crop yield and to encourage sustainable farming methods. In the process of strawberry farming, early detection is usually followed by a rapid spread of the disease and a high cost. In this paper, the author will describe a wall-mounted edge-intelligent IoT surveillance system that is capable of being deployed to constantly monitor the real-time status of strawberry plants in field settings. The suggested system substitutes manual inspection with an automated fixed position monitoring model with an integrated embedded vision hardware and lightweight deep learning. A SqueeNet-based classification network is trained in a manner that it is capable of identifying various diseases of a strawberry plant and executes effectively on resource-limited edge devices. In order to facilitate real-time operation, the UDP-based communication mechanism is used to relay detection results in the least amount of latency. An on-site alert unit which is an LCD display and a buzzer is also built into the system to allow the farmers to be able to obtain immediate feedback and make appropriate action. The proposed architecture, unlike solutions of agricultural monitoring that rely on clouds, integrates local edge processing in order to minimize energy consumption, network reliance, and response time. The experimental results demonstrate that the system is effective with regards to a reliable classification accuracy, constant performance, and rapid detection response in real-world conditions. The findings indicate that the suggested solution can be deployed in the smart farming and precision farming solutions in large-scale applications.