การจำแนกการใช้ประโยชน์ที่ดินและสิ่งปกคลุมดินของจังหวัดภูเก็ตจากภาพดาวเทียมแลนด์แซท 8 ด้วยคุณสมบัติทางชีพลักษณ์

Abstract

This study aims to test the utility of the decision tree classification using indices (i.e., Normalized Difference Vegetation Index (NDVI), Enhance Vegetation Index (EVI), Normalized Difference Water Index (NDWI), Land Surface Water Index (LSWI), and Soil-adjusted Vegetation Index (SAVI) exacted from Landsat 8 OLI multi-temporal data which representative of the phenology data of Land Use and Land Cover (LULC) in Phuket province, south of Thailand. The results showed that the multi temporal imagery classification was better than single date imagery classification, that the time series data aid with decision tree classification technique could provide the good classification results. The total testing accuracy and Kapa coefficients of 6 class were 98.42% and 97.92%, respectively. We hope that the methodology presented in this study can be used as a guideline for study in other areas.