Maize recognition and accuracy evaluation with GF-1 WFV sensor data

Abstract

As part of the "High-Resolution Earth Observation System," many major projects are being implemented. The first optical satellite (GF-1) in the high-resolution satellite series has completed in-orbit tests and entered the stage of data acquisition. GF-1 owns high resolution and information of wide field view sensor (WFV sensor) and the panchromatic and multispectral sensor (PMS sensor). In this study, GF-1 WFV sensor data with a resolution of 16 m, integrated with Landsat-8 and RapidEye data were selected to recognize maize in Xuchang using Support Vector Machine (SVM) and Spectral Angle Mapper (SAM) method. The results showed that the precision of classification varies greatly among WFV sensors. In particular, WFV3 was of the highest accuracy to identify crops and planting area with accuracy higher than Landsat-8 and close to RapidEye. With regard to WFV1 and WFV4, the application effect was worse and less viable to identify species of complex autumn crops. In brief, the classification accuracy of SVM classifier is better than SAM classifier. It can be also concluded that SVM is more suitable for the identification of crops and planting area of extraction in the study area.