Research On Multi-band Remote Sensing Image Change Detection Based On Time-Space Optimization Fusion

Remote sensing change detection is not only an important way to study global changes,but also an important basis for human beings to make scientific decisions in the sustainable use of resources.The use of remote sensing technology for change detection is the most economical and effective way to obtain land use change information.Therefore,research on the change detection method based on remote sensing technology has always played an important role in the field of remote sensing application research.In this paper,TM,OLI,MSI and other data sources are used to study the change detection method of fusion multi-band remote sensing images.This article uses a combination of remote sensing image processing techniques,geographic information systems,numerical analysis,mathematical statistics and other multidisciplinary knowledge to address several key issues in remote sensing image change detection techniques,including the construction of multi-temporal difference images and the extraction of multi-band difference information.,Threshold segmentation,detection accuracy evaluation methods,and so on,from a variety of aspects such as theory,methods,experiments were studied.The innovation of this article mainly has the following two aspects:1?Propose a multi-temporal difference image construction algorithm of spectral product fusion.The algorithm combines the results of the difference between the multi-temporal remote sensing images and the ratio results to obtain a difference image.Firstly,the two remote sensing images at different time phases after preprocessing are separately used for difference operation and ratio operation to obtain difference image and ratio image.Then,using the proposed fusion algorithm,the final difference image is calculated.One of the advantages of this algorithm is that the effect of image noise on the result is weakened to a certain extent,and the edge feature is retained.The second advantage is that the algorithm has a higher degree of automation.Experiments show that the algorithm can improve the quality of differential images to a great extent,and is generally applicable to mid-low resolution and high resolution multi-band images and panchromatic band images.2.The method of principal component transformation is used to fuse multi-band change information.Commonly used change detection methods generally cannot use multi-band difference information for single-band images.However,multi-band superposition will cause a large amount of information redundancy,increase calculation difficulty,and slow down the calculation speed.In this paper,the principal component analysis method is applied to the fusion of multi-band differential images,which not only highlights the main change information,but also reduces the amount of calculation and improves the time efficiency of the algorithm.Through experiments,this method has obviously improved the detection accuracy compared with the traditional change detection algorithm.