Multi-feature identification method based on high-resolution remote sensing image

Abstract

The invention discloses a multi-feature identification method based on a high-resolution remote sensing image and relates to the field of remote sensing image processing, in particular to a multi-feature recognition method based on a high-resolution remote sensing image. According to the identification method based on multiple features of the high-resolution remote sensing image, remote sensing multi-feature data serve as an input source of a neural network, multi-scale feature information of the remote sensing image is constructed, extracted and fused, an auxiliary loss function is added to improve the accuracy of a model, and the identification precision of the remote sensing image is improved. According to the invention, remote sensing image information can be better mined to improve the recognition capability of the deep convolutional network for the remote sensing image; an auxiliary loss function is set to assist the training of the network of the invention, so that the identification precision of the network can be improved; the network structure can extract and fuse different scale information of the remote sensing image, and can screen feature information beneficial to remote sensing image identification, so that the identification precision of the remote sensing image is improved; compared with a fusion method, the overall precision of remote sensing image identification can reach 1.4%.

Description

technical field [0001] The invention relates to the field of remote sensing image processing, in particular to a recognition method based on multiple features of high-resolution remote sensing images. Background technique [0002] Remote sensing image recognition is a basic problem in the field of remote sensing research. Through the characteristic information of the spectrum and texture of the ground objects, different ground objects are identified, and the classification of real ground object labels for each pixel in the image is realized. process. With the continuous development of remote sensing technology, high-resolution remote sensing images are used more and more frequently in various fields, and the requirements for remote sensing image classification technology are getting higher and higher. However, with the improvement of the resolution of remote sensing images, the spectra of different surface object types overlap each other, and the phenomenon of "same object ...

AI Technical Summary

Problems solved by technology

However, with the improvement of the resolution of remote sensing images, the spectra of different types of ground objects overlap each other, and the phenomenon of "same object with different spectra and different objects with same spectrum" is more serious, which seriously restricts the interpretation accuracy of high-resolution remote sensing images.

Although high-resolution remote sensing images can provide rich ground object information, but the amount of data is huge, the traditional remote sensing processing technology has not been able to fully mine the detailed information of the ground features, resulting in the phenomenon of "rich data, poor information"

Images