Design and Implementation of an IoT-Based Aerial Agriculture Monitoring System

This work focuses on the aerial supervision of large agricultural landscapes using the Internet of Things (IoT) platform to detect crop health. The input module engaged in the IoT system (camera module) captures the aerial view of the cropland, where the processing unit (Raspberry Pi 4) extracts the R-G-B (Red-Green-Blue) values from the captured image. The extraction of R-G-B values is performed using the Python’s CV2 Module. The IoT system comprises a camera module as input unit, a processing module, and a motor unit (two DC motors) as the output module. Further, the crop condition is analyzed to detect the health condition of plants. For instance, four different conditions are included here, namely condition-I: healthy crop, condition-II: crop with disease-1, condition-III: crop with disease-2, and condition-IV: crop with disease-3, depending upon the different ranges/values of R-G-B. For a range of R1-R2, G1-G2, B1-B2, condition-I is detected, where the process observes a healthy crop condition. Similarly, for a range of R3-R4, G3-G4, B3-B4, processor detects the condition-II. In response to this (crop with disease-1), the processor activated the Motor-I with medicine-I (specific medicine to cure the disease-1). Moreover, in condition III, for the detected range of R5-R6, G5-G6, B5-B6, the processor activated Motor-II with medicine-II (specific medicine for detected disease-2). In our last considered condition (IV), for a range of R7-R8, G7-G8, B7-B8, the crop is detected with disease-3. For this condition, the mixture of medicine 1 and medicine 2 is considered for the cure. In response, the processor activated both the motors (motor I and II), to flow the mixture of medicines 1 and 2 simultaneously. In this work, only the designing aspect of the IoT system is under consideration, without focusing on specific crops, diseases, and medicine for the respective diseases.