Multi-Sensor Satellite Image Classification Using Random Forest Algorithm in Google Earth Engine: Application of CAS500-1 Image Sets
Year: 2024
Type: article
Abstract: This study conducted multi-sensor image classification by utilizing Google Earth Engine (GEE) and applying satellite imagery from Compact Advanced Satellite 500-1 (CAS500-1), Sentinel-1, and Sentinel-2.The Land Use and Land Cover classification was performed using the Random Forest algorithm provided by GEE.The study experimented with various combinations of input data, integrating CAS500-1, Sentinel-1, and Sentinel-2 imagery with Normalized Difference Vegetation Index (NDVI) data from CAS500-1.The study area focused on the Goryeong County region in Gyeongsangbuk-do, and the satellite imagery was acquired in early January 2023.The results of this study showed that the highest classified result (94.51%) in overall accuracy and Kappa coefficient (0.9342) were achieved when applying CAS500-1, Sentinel-1, Sentinel-2 imagery, and NDVI data.The NDVI data is believed to complement the CAS500-1 imagery, improving classification accuracy.This study confirmed that applying multi-sensor data can improve classification accuracy, and the high-resolution characteristics of CAS500-1 imagery are expected to enable more detailed analyses within GEE.
Source: 대한원격탐사학회지
Authors: Jeonghee Lee, Kwangseob Kim, Kiwon Lee
Institutions: Hansung University, Kyungmin University
Language: English
Subfield: Media Technology
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
Funder: Hansung University
OpenAlex ID: W4404029328