Health Monitoring System in Smart Greenbox for Chili Plant Using Convolutional Neural Network

Pests and viruses have an easy time attacking chili plants. To counter these threats, a routine monitoring procedure is necessary. This monitoring method is less efficient because it takes time to get to the area when conducted directly at the plantation. With the help of this research, monitoring will be quick and simple, accessible from anywhere at any time. The classification of plant health issues is another goal of this research. Using the ESP32 Cam module, the monitoring system in this study receives plant photos as input. The cloud platform will get the captured image. This picture will complete the process of determining the health of the plant. Convolutional neural networks (CNNs) are used in machine learning to carry out the identification process. AlexNet is the architecture utilized by CNN. The website will show the outcomes of the image capture and plant health detection processes. Users can access and monitor the webpage. 118 (20%) test data and 470 (80%) training data were used for the test. Based on the test findings, 100% accuracy and 0.0001434 loss were obtained. Using pictures from the ESP32 Cam to test the CNN algorithm's classification accuracy level, the results showed a precision value of 43% and an accuracy value of 63%.