Random forest-based multi-scale stratified sampling high-resolution satellite image change detection method

Abstract

The invention discloses a random forest-based multi-scale stratified sampling high-resolution satellite image change detection method. The method adopts an idea for an object, multi-scale stratified sampling is carried out to automatically acquire multi-scale training samples, and sub-scale samples and current-scale samples are combined; a training sample spectrum is then extracted, and texture and shape features are fused together to form feature space, the sample combination and the corresponding feature space are inputted to a random forest to train multiple change classifiers, and a classifier with the minimum out-of-bag error parameter is selected as a change detection classifier for change detection. In comparison with the traditional method, according to the multi-scale stratified sampling method disclosed in the invention, multi-scale feature information is considered, training samples are automatically added to a change area and a non-change area while the manual workload is not increased, the training sample feature generalization ability is improved, classification change detection is carried out, the method is simple, the operability is strong, and the scalability is good.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, and relates to a high-resolution satellite image change detection method, in particular to a high-resolution satellite image change detection method based on random forest multi-scale layered sampling. Background technique [0002] Change detection technology in remote sensing images is widely used in land use / cover change, disaster assessment, urban sprawl monitoring and other fields. It can be mainly divided into pixel-based and object-based change detection methods. Although the pixel-based remote sensing image information extraction method is fast, the information extraction in high-resolution images has its inherent limitations. It only relies on the spectral information of ground objects to cause serious salt and pepper noise, and does not make good use of high-resolution images. The rich spatial information and semantic information of high-resolution images are lik...

AI Technical Summary

Problems solved by technology

[0003] At present, some methods based on object-oriented change detection often only use single-scale segmentation to obtain objects, and multi-scale information cannot be used. In fact, high-resolution images have multi-scale characteristics, and single-scale objects are different from other Objects on higher or lower scales are related to each other, and there will inevitably be over-segmentation or incomplete segmentation when using a single scale to segment images

Images