Remote sensing image change detection method and system taking into account temporal and spatial changes of image objects

Abstract

The invention provides a remote sensing image change detection method and a remote sensing image change detection system taking into consideration spatial and temporal variations of image objects. The method comprises the following steps: extracting the spectrum features, texture features and edge features of two images of different time phases; gathering the features of the two time phases by taking each pixel as an initial object, and merging connected sub objects according to a merging rule of minimum heterogeneity growth to get an incomplete image object set; combining the features of mono-temporal images into a feature set by taking the incomplete image object set as an initial image, and merging connected sub objects of the images of different time phases according to the merging rule of minimum heterogeneity growth to get to get a complete image object set; establishing an image object corresponding relationship according to the hierarchy of the object sets; and detecting complete change or local change of image objects based on the object corresponding relationship of images in different periods. By using a multi-feature method, the spatial change of multi-temporal image objects is taken into consideration, and the accuracy of object-oriented change detection is improved.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image change detection, and relates to a remote sensing image change detection method and system taking into account temporal and spatial changes of image objects. Background technique [0002] Under the dual effects of natural forces and human activities, the earth is constantly changing every day. With the rapid development of human society and the increasing level of science and technology, human beings are more and more capable of transforming the natural world, and the speed of earth changes is also intensifying. In order to know the status and progress of the earth's development, master the earth's changes and their impact on human beings, quickly and accurately obtain the earth's change information through change detection, and scientifically and efficiently monitor the geographical conditions, it is important to improve human beings' ability to prevent and warn various disasters It...

AI Technical Summary

Problems solved by technology

For the unchanged object, due to the influence of projection difference, its position is displaced in different time phase images, and the unified image object division will inevitably cause the target in one or more time phases to be inconsistent with the image object

Therefore, it will cause false changes and reduce the accuracy of change detection.

Images