The Application of Faster Region based Convolutional Neural Network for Rice Plant Disease Identification Vani Krismo Anggoro(a*), Dwi Ratna Sulistyaningrum(a)
(a) Department of Mathematic, Faculty of Science and Data Analytics, Institut Teknologi Sepuluh Nopember
Kampus ITS Sukolilo, Surabaya 60111, Indonesia
*sayavanikrismo[at]gmail.com
Abstract
The rice plant is the highest consumed food commodity by Indonesians compared to other commodities. In cultivating rice plants, farmers will encounter pests or diseases that attack the plants. They include Blast, Wereng Coklat or Brown Spot, etc. The lack of knowledge about diseases that attack rice plants is the cause of delays in the diagnosis and treatment process, causing crop failure and decreased rice production. Therefore, a tool is needed to help farmers recognize rice plant diseases automatically. A digital image processing approach using deep learning can help the process of disease recognition in rice plants. One of the object detection methods with deep learning used in this study was Faster RCNN (Faster Region based Convolutional Neural Network). Faster RCNN can identify rice plant diseases quite well. The Faster RCNN model is trained using 160 training image data and several parameters to get the best model performance. Based on the results of training conducted by the Faster RCNN system, the best test accuracy obtained from 40 test image data is 93.4%
Keywords: deep learning- Faster RCNN- rice plant disease
