A Binary Change Detection Method for Remote Sensing Imagery Based on Feature Deviation Alignment

Abstract

The invention discloses a binary change detection method of remote sensing images based on feature deviation alignment, comprising: step 1: constructing a dual-temporal remote sensing image binary change detection data set and performing preprocessing; step 2: constructing a binary change detection method based on feature deviation alignment Binary change detection model, and given the bitemporal remote sensing image to obtain the change area prediction result and the auxiliary prediction map of the change area; step 3: use the real change area label results and the predicted change area results and the change area auxiliary prediction map to calculate the main loss respectively Function and auxiliary loss function, update the model by backpropagating the gradient according to the loss, stop the training until the loss value converges, save the model structure and model weight; step 4: use the model weight trained in step 3 to predict the test set data. The invention can effectively solve the misdetection phenomenon of changing regions caused by factors such as multi-angle shooting, too many high-rise buildings, or large terrain fluctuations in dual-temporal remote sensing images.

Description

[0001] field of invention [0002] The invention belongs to the field of remote sensing image processing, relates to the field of computer deep learning, and in particular relates to a binary change detection method of remote sensing images based on feature deviation alignment. Background technique [0003] The goal of the binary change detection task is to locate the change area and non-change area of ​​the area given two dual-temporal remote sensing images of the same area. With the development of deep learning technology, methods based on twin convolutional neural networks have achieved higher accuracy than traditional methods in binary change detection tasks, which usually use an encoder-decoder architecture, that is, twin convolutional neural networks with shared weights are first used. The network encoder extracts the low-level features and high-level features of the bitemporal image respectively, and then uses the decoder to obtain the difference features and gradually ...

AI Technical Summary

Problems solved by technology

[0004] However, when the binary change detection method based on Siamese convolutional neural network obtains difference features in high-dimensional feature space, it needs to ensure that the extracted high-dimensional features are aligned in the original bitemporal image as much as possible, otherwise it will be due to The problem of feature deviation appears false detection area

However, due to the existence of registration errors and downsampling layers, there will inevitably be a problem of feature deviation between the high-dimensional features of bitemporal images.

The existing binary change detection methods often directly introduce the semantic segmentation model or make certain improvements on the basis of the semantic segmentation model, without considering the problem of feature deviation

Images