Study on Extraction Methods of Winter Wheat Area Based on GF-1 Satellite Images

Abstract

Three GF-1 WFV images on March 16, 2014, April 9, 2014, and April 30, 2014 were selected to extract the planting area of winter wheat in Jianhu county of Jiangsu province. Vegetation indexes were extracted from the original spectrum data in order to extract winter wheat area with Maximum Likelihood Classifier (MLC), Support Vector Machine (SVM) and Classification and Regression Trees (CART). The extraction accuracy of wheat was verified through on-site GPS measurement of 5 ground samples area with the scale of 1km $\times$ 1km. The extraction accuracy of winter wheat area with SVM reached 84.138% on April 9 was the highest among three phases image. It indicated that the image on 9 April (booting stage) was the most suitable temporal for wheat identification. The GF-1 satellite image can be used for monitoring the cultivated area of wheat and it has higher accuracy and broad application prospects in the field of agriculture remote sensing monitoring.