Irrigated farmland identification method based on remote sensing vegetation canopy moisture index

Abstract

The invention discloses an irrigation farmland identification method based on a remote sensing vegetation canopy moisture index, and particularly relates to an irrigation identification method based on a regression relationship between a remote sensing vegetation canopy moisture index difference value of a farmland and a nearby forest and average annual precipitation. Due to the effect of rainfall, the irrigation threshold value of each area is reduced along with the increase of the average annual precipitation. Therefore, the determination of the irrigation threshold value can realize slidingvaluing in space, i.e., the smaller the irrigation threshold value is in the place with the larger average annual precipitation, the larger the irrigation threshold value is in the place with the smaller average annual precipitation. By designing a linear regression equation and fully utilizing the linear regression relationship between the difference value of the moisture indexes of the irrigated farmland pixel and the nearby forest pixel and the average annual precipitation, the change of the irrigation threshold in space can be realized, so that the identification of the irrigated farmlandis realized. The method has the characteristics of no dependence on priori knowledge, good robustness, high classification precision, strong recognition capability and the like.

Description

technical field [0001] The present invention relates to the technical field of agricultural remote sensing, and more specifically, relates to an irrigation identification method based on the regression relationship between the difference of moisture index of remote sensing vegetation canopy and annual precipitation in farmland and nearby forests. Background technique [0002] Accurate spatial distribution information of irrigated crops is an important basis for ensuring food security, and is crucial to the adjustment of agricultural industrial structure and the estimation of food production. Especially in arid regions, irrigation is an important condition for crop production. In China, only 40% of the total arable land is irrigated, but it produces 74% of the country's grain. Therefore, it is of great significance to develop a fast and accurate method for monitoring irrigation distribution. Due to the limitations of data availability and computing power, high-precision opt...

AI Technical Summary

Problems solved by technology

Therefore, the survey data cannot accurately and timely reflect the spatial distribution of irrigated and non-irrigated areas; (2) The United States Geological Survey (USGS) and the International Water Management Institute (IWMI) use unsupervised classification methods to analyze global irrigation and rainfed areas. Regions were classified to produce a Global Land Cover Type Map (GLCC) and a Global Irrigation Map (GIMA)

This method can still be used in the case of sparse local information; but this method is difficult to control the number of coverage types

In addition, the coverage type of the classification combination is affected by subjective factors; (3) Ozdogan and Gutman (2008) used the decision tree supervised classification method to draw a map of the irrigation and non-irrigation areas in the United States, which has the advantage of being easy to control the training samples , but its classification accuracy is usually significantly affected by the training samples

Images