Paddy Growth Stages Classification based on Hyperspectral Image using Decision Tree and Naive Bayes

Febri Maspiyanti (Informatics Engineering, Pancasila University, Jakarta)
Ghinanti Pratiwi (Informatics Engineering, Pancasila University, Jakarta)
Jullend Gatc (Information System, Kalbis Institute, Jakarta)

JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi)
Vol 4, No 3 (2021): JTKSI

Hyperspectral imaging is one of remote sensing technology that gather information from a wide spectrum of electromagnets called spectral bands, with the aim of finding objects, identifying materials, or detecting processes. In an effort to calculate the amount of rice crops can be harvested within a certain periode of time, we need to accurately predict the growing phase of paddy plant at that time. In determining the phase of the rice plant with high accuracy value, need to be supported with the selection of appropriate algorithms, and also the features selection. In this study, a comparison between the Decision Tree and Naive Bayes methods to classify the nine phases of rice growth based on hyperspectral image achieve accuracy value of 91.67% and 83% respectively. Based on the accuracy result, our new proposed method improved 6,38% accuracy compare to our previous research.