High-resolution remote sensing image change detection method based on twin convolutional neural network

Abstract

The invention belongs to the technical field of optical remote sensing image change detection, and particularly relates to a high-resolution remote sensing image change detection method based on a twin convolutional neural network. The method comprises the following steps: 1, making a sample data set; 2, constructing a twin convolutional neural network model; 3, training a twin convolutional neural network model; 4, model prediction is carried out, and a change detection initial result is output; and 5, carrying out post-processing on the initial result to obtain a final change detection result. According to the method, the convolutional neural network technology is applied to the field of high-resolution remote sensing image change detection, and the twin convolutional neural network fusing high-dimensional features and low-dimensional features is constructed to train and autonomously extract the features layer by layer, so that the problem of information omission possibly occurring in manual calculation and feature selection processes is avoided, and the change detection precision is improved.

Description

technical field [0001] The invention belongs to the technical field of optical remote sensing image change detection, and in particular relates to a high-resolution remote sensing image change detection method based on a twin convolutional neural network. Background technique [0002] Change detection is an important research direction in the field of remote sensing, and it has a wide range of applications in civilian and military applications, such as military target monitoring, battlefield intelligence analysis, land inspection, disaster assessment, urban planning and other fields. The change detection of remote sensing images is the process of quantitatively analyzing and determining the changes of ground features from remote sensing images of different periods, which involves the description of the type of change, distribution and change information, that is, it is necessary to determine the types of ground features before and after the change, and analyze the changes. A...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problems existing in the prior art, the present invention provides a high-resolution remote sensing image change detection method based on a twin convolutional neural network, applying the convolutional neural network technology to the field of high-resolution remote sensing image change detection, and constructing a fusion high- and low-dimensional The feature twin convolutional neural network is trained layer by layer to extract features independently, avoiding the problem of information omission that may occur in the process of manual calculation and feature selection, and improving the accuracy of change detection

Images