Rice yield estimation method combining satellite image and MODIS data

Abstract

The invention relates to a rice yield estimation method combining a satellite image and MODIS data. The method comprises the following steps: carrying out preprocessing operation on an original satellite image; selecting a region of interest on the preprocessed satellite image, carrying out classification and rice extraction on the region of interest by adopting three methods of supervised classification based on manual visual interpretation, object-oriented classification based on rules and decision tree classification based on expert knowledge, wherein classified types include forest land, rice, water and construction land; taking an intersection region of rice extraction results obtained by the three classification methods as a rice region, and extracting an EVI index mean value image in the MODIS images of the rice region in different periods; taking the rice EVI mean value as an independent variable, rice yield statistical data as a dependent variable, carrying out regression analysis, and constructing a statistical regression estimation model. The problem that the data volume is insufficient when a high-resolution image is used for multi-time-sequence index extraction in a specific period is solved, and the problem that low-resolution extraction precision is low is solved.

Description

technical field [0001] The invention relates to the field of rice yield estimation, in particular to a rice yield estimation method combined with satellite images and MODIS data. Background technique [0002] As one of the most important food crops in the world, rice yield data is of great significance. Traditional rice production estimation is slow, heavy workload and high cost. At present, most of the channels for obtaining rice production data come from the data collected by the statistical department at all levels, the data collected by the agricultural department, and the data calculated by the meteorological department according to the climatic conditions. The traditional rice production estimation is based on the above data and according to the physical structure of rice. It is difficult to estimate the rice yield in a large area by using statistical mathematics to estimate the rice yield in a small area through the sample survey of the crop area. [0003] Compared ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the technical problems existing in the prior art, the present invention provides a rice yield estimation method combining satellite images and MODIS data, which solves the problem of low resolution and low extraction accuracy of rice yield estimation data in the prior art

Images