การสร้างดัชนีแอนแทรคโนสและการคาดการณ์ผลผลิตกาแฟพันธุ์อาราบิก้าจากการเกิดโรคแอนแทรคโนสด้วยดาวเทียม Sentinel-2

Abstract

Anthracnose Disease outbreak in Arabica coffee. It has caused great damage to Thai coffee farmers and the agricultural economic system. The aims of this study are separated into three objectives. Firstly, this research aims to analyze the physical factors that affect the occurrence of Anthracnose Disease in Arabica coffee. Secondly, to create an Anthracnose Disease Index in Arabica coffee using Sentinel-2 and Finally, aims to predict the yield of Arabica coffee using the Sentinel-2 satellite in Anthracnose Disease outbreaking. Physical factors from satellite imagery indices are analyzed about environmental conditions to the occurrence of fungi that cause Anthracnose Disease to be used to model Anthracnose Disease predictions. For the creation of an Anthracnose Disease Index analyzed in combination with satellite imagery of bands from spectral-signature analysis to obtain a test index sample suitable for Anthracnose indexation, and to use physical factors to analyze Arabica coffee yield prediction with multiple linear regression equations to evaluate coffee yield in the next harvesting by comparing it with actual yield data from field surveys. The study found that the Land Surface Temperature Index (LST) is a physical factor associated with temperature that affects the occurrence of fungi that cause Anthracnose. It is a physical factor that is associated with temperature that affects the occurrence of fungi that cause Anthracnose Disease. When temperatures are high and there is drought, fungi can thrive, infesting resident plants. The Normalize Difference Moisture Index (NDMI) can explain the spread of Anthracnose Disease through fungal spores in wind and rain that cause air humidity, and elevation affects the growth and plant disease resistance of Arabica coffee plants. Anthracnose indexing section It was found that 2 indices are test index 0E and test index 0F, which are test index 0E that is suitable for analyzing the probability of Anthracnose Disease and index 0F which is suitable for analyzing the probability of not occurring Anthracnose Disease and analysis of yield prediction of Arabica coffee. It was found that the Leaf Area Index (LAI) and the Normalized Difference Moisture Index (NDMI) correlated with yield prediction. When analyzing the data error from the Root Mean Square Error (RMSE), it was found that the mean squared error was 76.61, which means that the yield calculated from the production yield prediction model equation has a deviation from the actual yield of 76.61 kg per rai.