Method for detecting image change interactively

Abstract

The invention discloses a method for detecting image change interactively. The method comprises the following steps of: constructing change vectors of pixels of two colored images which are shot at different times in the same site by a local window of the image and space coordinates of the pixels; performing clustering analysis on changed pixels and non-changed pixels which are labeled by users respectively; constructing a spline function by taking a clustering center as a reference point; determining undetermined coefficients of the spline function by taking the clustering center as a training sample in a regression estimation method; and mapping pixels which are not labeled by utilizing the spline function which is obtained by regression estimation, and judging whether the pixels belong to the changed pixels or non-changed pixels according to a mapping result. By the method, the problems of low detection accuracy, uncertainty generated during the determination of a change area and non-change area of the image and the like in the prior art are solved, so that the method has a bright application prospect.

Description

technical field [0001] The invention relates to the technical field of image processing and pattern recognition, in particular to the field of interactive image change detection. Background technique [0002] Image change detection refers to identifying the area where the location has changed by analyzing two images taken at different times at the same location. Image change detection technology is an important means to obtain information on changes in ground objects and objects in areas of interest, and has very important applications in many fields such as land resource management, environmental monitoring, natural resource monitoring, and disaster monitoring. [0003] After years of research, image change detection technology has been greatly developed. However, there currently does not exist a general change detection method that can fully automatically detect regions that change between two images. The fundamental reason is that it is difficult to automatically and ac...

AI Technical Summary

Problems solved by technology

In fact, the complexity and diversity of visual objects bring great difficulties to automatically define changing and non-changing classes

This difficulty will make it difficult to select appropriate feature extraction methods and classification methods, thereby making it difficult to guarantee the accuracy of change detection

Images