A method and system for crop yield spatialization

Abstract

The invention discloses a method and a system for spatializing crop yield. The method comprises the following steps: acquiring NDVI varying with time from a remote sensing image; determining the planting range by a CART classification and regression tree algorithm; dividing planting areas; acquiring statistical yields of crops by region; determining the growing season of crops, including multiplesingle-time phases of crops and different phenological phases formed by the combination of single-time phases; according to the NDVI of each single phase and phenological period in the growing season,analyzing the correlation between the statistical yield of crops and the NDVI of the growing season by linear regression analysis; recording the NDVI variable of the growing season with the highest correlation with the statistical yield of crops and the smallest root mean square error as the NDVI variable of the target growing season; according to the proportion of the NDVI of each pixel to the total NDVI of the region in the target growing season, determining the spatial yield of each pixel. The invention can reflect the spatial pattern and the dynamic change of the crop yield, and providesa reference for the optimization of the crop planting structure.

Description

technical field [0001] The invention relates to the field of crop yield statistics, in particular to a method and system for spatializing crop yield. Background technique [0002] Social statistics are one of the important indicators that reflect the social development of a country or region, and are an important basis for the country or region to formulate laws and policies and macro-control management. Traditional social statistics can no longer meet people's needs. Traditional social statistics are generally collected and sorted out using administrative divisions (provincial, prefectural, county, township, etc.) as statistical units, with low spatial resolution, lack of description of data spatial distribution characteristics, and incomplete data However, it cannot express the spatial differences within administrative divisions in detail, which makes it difficult to analyze the overlapping of social statistical data and other data, which affects the collection of effecti...

AI Technical Summary

Problems solved by technology

Traditional social statistics can no longer meet people's needs

Traditional social statistics are generally collected and sorted out using administrative divisions (provincial, prefectural, county, township, etc.) as statistical units, with low spatial resolution, lack of description of data spatial distribution characteristics, and incomplete data Shortcomings such as inability to express spatial differences in administrative divisions in detail, making it difficult to analyze the overlapping of social statistical data and other data, affecting the collection of effective information, and seriously hindering the extraction of effective information

Images