Generative-adversarial-network-based multi-spectral image change detection method

Abstract

The invention discloses a generative-adversarial-network-based multi-spectral image change detection method so that problems of low detection precision and high noise sensitive of the existing methodcan be solved. The method comprises: step one, setting structures and objective functions of a discriminant classification network D and a generative network G as well as a distance coefficient lambdabetween the image generated by the generative network G and a real image; step two, acquiring a difference chart ID of images of two different time phases; step three, dividing the ID to obtain an initial change detection result, classifying the two different time phases into marked data and unmarked data based on the result, and forming a training set; step four, forming a classification networkW by the discriminant classification network D and the generative network G, and training the classification network by using the training set to obtain a trained discriminant classification networkD'; and step five, inputting the images of the two different time phases into the discriminant classification network D' to obtain a final change detection result. The method has advantages of high detection precision and strong robustness and can be applied to image understanding or pattern recognition.

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 image multispectral images, which can be used for pattern recognition or target detection. Background technique [0002] Image change detection is a technique that can identify areas of change between images of the same region at different times. With the rapid development of remote sensing technology, multispectral images with high resolution have become easy to obtain. Change detection in high-resolution multispectral images has been given more attention. [0003] At present, the multispectral image change detection method widely used in disaster assessment, video detection and other fields is based on the method of image difference map, which can be divided into three steps: 1. Preprocessing the multispectral images in different phases, Mainly including noise removal and registration; 2. Generate difference maps of multi...

AI Technical Summary

Problems solved by technology

However, most of these methods deal with SAR images, and high-resolution multispectral images have more spectral channels. When this method is applied to high-resolution multispectral images, the detection accuracy is low, and the detection results contain many noise problem

Images