Identifying Sugarcane Plantation using LANDSAT-8 Images with Support Vector Machines

2nd International Conference of Indonesian Society for Remote Sensing 2016

地球科学;计算机科学

Mulyono, Sidik^1 ; Nadirah^1

Agency for the Assessment and Application of Technology (BPPT), Indonesia^1

关键词: Enhanced vegetation index;    Environmental conditions;    Field campaign;    Normalized difference vegetation index;    Number of samples;    Training purpose;    Training sample;    Vegetation index;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/47/1/012008/pdf
DOI  :  10.1088/1755-1315/47/1/012008

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

The use of remote sensing has been highly beneficial in the identification and also mapping and monitoring of plantations. The identification of plantations includes the physiology, disease, environmental conditions, and also the production and time of harvesting. It can be done by doing satellite imagery classification. However, to reach the final result of identification, it could be carried out by getting the solid ground truth information. This paper will discuss about detection of sugarcane plantation in Magetan district of East Java province area by using LANDSAT-8 image with specific approach of phenology profile using EVI (Enhanced Vegetation Index) value from satellite data, as an alternative vegetation index to address some of the limitation of the NDVI (Normalized Difference Vegetation Index). Method of classification used for detecting sugarcane plantation is Support Vector machines (SVM), which is a promising machine learning methodology. It has the ability to generalize well even with limited training samples and complex data. A number of samples of phenology profile for training purpose using SVMs are obtained from the area that identified as sugarcane plantation during field campaign in 2015. The same manner is also done for the objects instead of sugarcane plantation with relatively the same number of samples. The result of the research shows that Remote Sensing is able to detect the sugarcane plantation cross the district with good accuracy.