Sentinel-1A image classification for identification of garlic plants using decision tree and convolutional neural network

The Indonesian government launched a garlic self-sufficiency program by 2033 to reduce imports monitoring lands in several central areas. Remote sensing using satellite imageries can assist the mapping lands. A previous study has classified Sentinel-1A identify Sembalun Lombok Indonesia decision tree C5.0 algorithm with three scenarios data input and produced model an accuracy of 78.45% two attributes vertical-vertical (VV) vertical-horizontal (VH) bands. Therefore, this aims improve classification from study. This applied algorithms, convolutional neural network (CNN), new which used combinations attributes). results show that use as for not increase model's accuracy. While CNN shows it study's 7.91% because 86.36%. is expected help land identification area support programs