Crop remote sensing classification method based on multi-temporal SAR data and multi-spectral data

Abstract

In order to solve the problem of low precision of crop remote sensing classification, the invention provides a crop remote sensing classification method based on multi-temporal SAR data and multi-spectral data, belonging to the field of agricultural technology. The invention comprises the following steps: S1, respectively acquiring multi-spectral data in the crop growing period and SAR data with obvious characteristics at different times in the crop growing period; S2, extracting the cultivated land range according to the multi-spectral data with prominent cultivated land information; 3, selecting that VV polarization data band combination of different time in the SAR data to obtain the multi-temporal SAR data, and combining the multi-temporal SAR data with the single-temporal multi-spectral data respectively; 4, creating a training sample; S5, the crops in the study area are classified by maximum likelihood method with the combination of multi-temporal SAR data and single-temporal multi-spectral data bands, multi-temporal SAR data and single-temporal multi-spectral data, which are masked by the extracted cultivated land range, combined with the spectral characteristic mean and covariance of the training samples.

Description

technical field [0001] The invention relates to a method for improving the accuracy of crop remote sensing classification, in particular to a crop remote sensing classification method based on multi-temporal SAR data and multi-spectral data, which belongs to the field of agricultural technology. Background technique [0002] Since the identification and monitoring of crops in the same growth period by multi-spectral data still faces the problem of "same spectrum and foreign objects", and is easily affected by cloudy and rainy weather, it is impossible to obtain available monitoring data during the critical growth period of crops. Based on this situation, radar data has become a usable data source because it can work around the clock and around the clock. And one of the most important contents of remote sensing monitoring of agricultural conditions is the timely and objective forecast of crop planting area and yield. Accurate identification of different crops is a very import...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem of low crop remote sensing classification accuracy caused by low image acquisition rate in crop remote sensing classification monitoring, the present invention provides a crop remote sensing classification method based on multi-temporal SAR data and multi-spectral data that is fast, accurate and improves the overall crop classification accuracy

Images