Remote sensing image change detection method

Abstract

The invention discloses a remote sensing image change detection method. The method comprises the following steps: acquiring a first remote sensing image and a second remote sensing image shot at the same position at different moments; respectively marking the first remote sensing image and the second remote sensing image to form unlabeled sample data and labeled sample data; performing vector format conversion on the unlabeled sample data and the labeled sample data to form unlabeled sample vector data and labeled sample vector data; performing network structure search by using an evolutionary algorithm to obtain a plurality of target ladder grid models; performing semi-supervised and unsupervised training on each target ladder network model to determine an intermediate optimal ladder network model; and performing semi-supervised and unsupervised training on the middle optimal ladder network model to obtain a target optimal ladder network model, wherein an output result corresponding to the target optimal ladder network model is a detection result. The detection method is higher in universality, and a better remote sensing image change detection result can be obtained.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method for detecting changes in remote sensing images. Background technique [0002] Remote sensing image change detection (Change Detection, CD for short) is a technology that identifies the changing areas of the same area at different times by comparing and processing a pair of remote sensing images. With the development of remote sensing technology, the types of remote sensing images are gradually diversified, such as Synthetic Aperture Radar (SAR) images, optical images, heterogeneous images, and Very High Resolution (VHR) images. , multispectral images and hyperspectral images, etc., bring more abundant remote sensing image information. Nowadays, as more and more remote sensing image information can be fully utilized, the applications of remote sensing image change detection are also very common, such as surface change analysis, land cover change, nat...

AI Technical Summary

Problems solved by technology

[0005] 1. The training of labeled data and unlabeled data is relatively independent, and it is difficult to train or obtain a good remote sensing image information representation structure to spread to unlabeled data, thereby affecting the detection accuracy of remote sensing image changes;

[0006] 2. Changes in detection objects lead to large changes in the difference information between objects and the scale of data that needs to be processed. The fixed network structure has poor adaptability, which will also affect the detection accuracy of remote sensing image changes.

Images