Agricultural remote sensing image analysis method and system

Abstract

The invention provides an agricultural remote sensing image analysis method and system. The method comprises the steps of obtaining an agricultural remote sensing image; inputting an agricultural remote sensing image to an improved U-Net neural network model, and outputting the ground feature types corresponding to all pixels of the agricultural remote sensing image, wherein the improved U-Net neural network model is obtained after training is carried out on the agricultural remote sensing image with ground feature type labels, the improved U-Net neural network model comprises multiple parallel branch models, each branch model is used for extracting different ground feature features of the agricultural remote sensing image, and each branch model is a U-Net deep convolutional neural network. The method effectively solves the problems that a traditional method is high in data acquisition cost and poor in segmentation precision, and especially pixel-level segmentation is difficult to achieve, and provides important technical and method bases for agricultural quantitative remote sensing analysis, crop planting area remote sensing mapping and the like; and the method has important reference significance and value for improving the agricultural remote sensing image analysis effect.

Description

technical field [0001] The invention relates to the technical field of deep learning and image processing, in particular to an agricultural remote sensing image analysis method and system. Background technique [0002] As a type of image that records the electromagnetic wave characteristics of various ground objects, remote sensing images have the advantages of fast, simple, macroscopic, non-destructive and objective, and are widely used in various aspects of agricultural production by countries all over the world. As one of the world's major agricultural countries, combined with various agricultural remote sensing images, it is possible to obtain various agricultural information including agricultural quantitative remote sensing analysis and crop planting area remote sensing mapping, and improve agricultural production efficiency, which is of great significance to my country's agricultural development. very important. [0003] The analysis of agricultural remote sensing ima...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a method and system for analyzing agricultural remote sensing images, aiming to solve the problems in the prior art that consume a lot of manpower and material resources, and that manual classification is easily affected by subjectivity, has large errors, and affects classification efficiency

Images