Remote sensing image scene classification method and device

Abstract

The invention discloses a remote sensing image scene classification method and device, which alleviate the interference of redundant information and the influence of intra-class diversity and improvethe classification accuracy of remote sensing image scene classification. The method comprises the steps of obtaining a scene data set of a remote sensing image, wherein the scene data set comprises atraining set and a test set; preprocessing the remote sensing images of the training set through a data enhancement technology; improving a preset residual error network structure through a self-attention mechanism to obtain a self-attention residual error network; training the self-attention residual error network by using the training set, and optimizing parameters of the self-attention residual error network according to the cross entropy loss and the center loss; and classifying the images of the test set according to the optimized self-attention residual network to obtain a remote sensing image scene classification result.

Description

technical field [0001] The invention relates to the technical field of remote sensing, in particular to a remote sensing image scene classification method and device. Background technique [0002] In recent years, with the rapid development of remote sensing technology and sensor systems, remote sensing image data continue to emerge. Remote sensing image scene classification is an important means of remote sensing image analysis and interpretation, and has a wide range of applications in the fields of urban planning, land resource management, and military investigation. However, the spatial layout of remote sensing scenes is complex, and the classification is easily disturbed by redundant backgrounds. The diversity of targets makes it more difficult to distinguish their categories, which makes the classification task quite challenging. [0003] With the development of deep learning, convolutional neural networks are also used by researchers to extract advanced semantic feat...

AI Technical Summary

Problems solved by technology

[0004] However, since the targets in the remote sensing scene are usually small and scattered, the redundant background occupies most of the image space, and not all the information in the scene is useful for classification

However, most of the existing remote sensing image scene classification methods extract features from the entire image, but do not make full use of the most critical target information in the image, and are easily disturbed by redundant information.

In addition, the intra-class diversity of remote sensing image scenes also brings great difficulty to classification, and factors such as different seasons, locations, and sensors may make scenes of the same category have greater differences.

The interference of redundant information and the influence of intra-class diversity will lead to a decrease in the classification accuracy of remote sensing image scene classification

Images