Method and system for extracting global specific crop main producing area planting distribution

Abstract

The invention discloses a method and system for extracting global specific crop main producing area planting distribution and belongs to the field of agricultural remote sensing. The method provided by the invention comprises the following steps: obtaining multi-temporal data of MODIS data with medium and low resolution, generating a time sequence NDVI image, and carrying out S-G filtering on theimage;, and screening the image according to the phenological period of the specific crop to generate a new NDVI image; performing principal component transformation on the image, and selecting a principal component as a supervised classification input image; obtaining a sample of the specific crop by calculating a root-mean-square error of a growth curve and a standard growth curve of the NDVI image; and calling a support vector machine method to supervise and classify the input image, and constraining a supervised and classified result. According to the method, the NDVI time sequence changecurve of the specific crop is analyzed to obtain the growth curve, so that samples are convenient to select, and the influence of artificial experience and the time consumption of sample selection arereduced. The extraction efficiency of crop planting distribution can be improved, and meanwhile the extraction accuracy is improved.

Description

technical field [0001] The present invention relates to the field of agricultural remote sensing, in particular to a method and system for extracting the planting distribution of specific agricultural crops in the world. Background technique [0002] Understanding the planting distribution of crops can provide data reference and guidance for agricultural development. With the development of modern technology, it is aimed at a wide range of crop planting distribution. Traditional field surveys have been gradually replaced by remote sensing monitoring. Remote sensing monitoring is the long-distance detection of electromagnetic wave radiation and reflection characteristics through telemetry instruments on artificial satellites and other equipment. [0003] Remote sensing technology can provide effective monitoring for large-scale crop planting. At present, images with medium-to-high spatial resolution and high temporal resolution can better overcome the problem of mixed pixels ...

AI Technical Summary

Problems solved by technology

[0003] Remote sensing technology can provide effective monitoring for large-scale crop planting. At present, images with medium-to-high spatial resolution and high temporal resolution can better overcome the problem of mixed pixels in low-resolution remote sensing images. Low, it is difficult to reflect the phenological characteristics of crop dynamic changes, and it is difficult to distinguish multiple crop types with similar phenological periods in the same region for crops on a global scale

Therefore, it is necessary to use low-to-medium resolution multi-time series data for analysis. When using supervised classification methods for low-to-medium-resolution data, a lot of manual experience and sample selection time are required, and the efficiency of analysis and extraction is low.

Images