Cultivated crop type extraction method and system, storage medium and electronic equipment

Abstract

The invention discloses a cultivated crop type extraction method and system, a storage medium and electronic equipment. The invention discloses a cultivated land crop type extraction method. The method comprises the following steps: obtaining cultivated land parcel vector data of a target area; Constructing a radar time sequence image by using the obtained original radar data; Superposing the radar time sequence image and the cultivated land parcel vector to form radar time sequence characteristics; Obtaining sample data of a plurality of crops of the cultivated land parcel in the target area,marking the type of the crops, and generating a crop sample of the cultivated land parcel in the target area; Utilizing the deep learning network model to train the radar time sequence characteristics of the samples collected in the target area to obtain a crop type model; Inputting radar time sequence images and vector data of cultivated land parcels, and performing machine extraction on crop types of the input data by utilizing a crop type model; By combining the advantages of radar image all-weather monitoring and optical image ground object boundary clearness, the defect of single observation of the radar image all-weather monitoring and the optical image ground object boundary clearness is overcome, large-area crop type extraction in cloudy and rainy areas is achieved, and managementof various crops is facilitated.

Description

technical field [0001] The invention belongs to the intersecting field of remote sensing and agriculture, specifically relates to a method for extracting cultivated land crop types, and is mainly applied to the automatic extraction of cultivated land crop types of sub-meter level remote sensing images. Background technique [0002] Satellite remote sensing technology can obtain large-scale, multi-temporal image data, which provides a strong technical support for crop monitoring. The current crop classification technology based on remote sensing images is mainly the crop classification of optical images. The classification principle is that different types of crops show different spectral characteristics in each band of remote sensing images. Through the difference in spectral characteristics, different crops can be identified. Types of. However, optical remote sensing satellites are affected by clouds and revisit cycles, and crop classification is limited. It is difficult t...

AI Technical Summary

Problems solved by technology

Then there is coherent speckle noise in the radar image, and it is necessary to perform multi-look and filter noise removal operations on it during preprocessing. At the same time, this denoising operation will make the boundary of the ground object blurred, so it becomes a limit to the classification accuracy of the radar image. The main reason

Images