Detection of bacterial leaf blight disease using RGB-based vegetation indices and fuzzy logic

• Main Authors: Aziz, Nor Hafiza, Haron Narashid, Rohayu, Razak, Tajul Rosli, Anshah, Siti Aminah, Talib, Noorfatekah, Abd Latif, Zulkiflee, Hashim, Norhashila, Zainuddin, Khairulazhar
• Format: Conference or Workshop Item
• Published: IEEE 2023
• Online Access: http://psasir.upm.edu.my/id/eprint/37737/; https://ieeexplore.ieee.org/document/10087429

Paddy planting becomes the primary source of income and livelihood for paddy farmers, especially small-scale farmers and landless laborers. Unfortunately, rice production has been threatened by paddy disease. The bacteria leaf blight disease (BLB) is one of Malaysia’s most significant paddy diseases, causing substantial harm to rice production. This study aims to determine the bacteria leaf blight (BLB) disease from the utilized techniques of RGB-Based Vegetation Indices and Fuzzy Logic on the Unmanned Aerial Vehicle (UAV) images during the first paddy season 2022 in Perlis. In this study, the RGB-based indices of Normalized Green Red Different Index (NGRDI) and Green Leaf Index (GLI) were applied to the UAV Images captured at 20m altitudes. Then the fuzzy logic classification technique was applied to identify the BLB disease severity which consists of healthy and infected paddy leaves with the acceptable accuracy of 90.16%. Based on the classified BLB severeness with fuzzy logic, the result shows that the NGRDI was more significant to identify paddy disease in the area. In contrast, the GLI index is more significant to identify the non-paddy area. The NGRDI and GLI index ranges for BLB were found between -0.054 to 0.092 and 0.005 to 0.222. For more improvement of the study, the multispectral UAV Image should be applied to increase the accuracy of paddy disease detection like BLB and the images will also be taken and verified in other paddy plots with the aid of a spectroradiometer.