Integrated Monitoring of Pests and Diseases in Paddy Plants Based on Convolutional Neural Network Mike Yuliana, Jalu Tirtabuana
Politeknik Elektronika Negeri Surabaya
Abstract
Pests and diseases are a serious threat to rice production which can cause huge losses for farmers. Early detection and monitoring of pests and diseases of rice plants is an important key in reducing production losses and maintaining plant health. The paper aims to develop an integrated system for monitoring pests and diseases in rice plants based on deep learning using a Convolutional Neural Network (CNN) with image processing results to be uploaded to the web. This research also collects image data from rice plants infected with pests and diseases and will be used to train a CNN model that can recognize and classify types of pests and diseases in rice plants. The trained CNN model will be integrated with a web application. This web application will allow users, such as farmers or agricultural experts, to discover the health condition of rice plants in agrarian areas where testing has been conducted. Developing an integrated pest and disease monitoring system for rice plants based on a Convolutional Neural Network, which uses image processing with the results uploaded to the web, will make an important contribution in supporting efforts for early detection and monitoring of pests and diseases in rice plants. The results obtained during the experiments in this paper are the smallest loss value in the training model testing was obtained with a learning rate of 0.001, batch size of 64, and epoch 20, yielding a value of 0.1876 for pest and disease detection.
Keywords: Convolutional Neural Network, pest and disease detection, farmers, learning rate, rice plants
