Remote sensing image water extraction method and device based on RGBW characteristic space diagram cutting algorithm

Abstract

The present invention discloses a remote sensing image water extraction method based on an RGBW characteristic space diagram cutting algorithm. A water index method and an image classification method are combined, firstly the water index method is utilized to carry out preliminary extraction of water, then the preliminarily extracted water body is used as the initial target of the diagram cutting algorithm, in the constructed RGBW characteristic space, the diagram cutting algorithm is used to carry out fine extraction of water, finally the fine processing of a water boundary is carried out adaptively according to the water index and color characteristic information. The invention also discloses a remote sensing image water extraction device based on an RGBW characteristic space diagram cutting algorithm. The accurate extraction of the information of water with a complex surrounding environment can be realized and the extraction effect of the water boundary is good. In addition, human intervention is not needed in the water extraction process of the invention, the full automation and full intelligence are can be realized, and the automatic batch processing of a large number of remote sensing image data is facilitated.

Description

technical field [0001] The invention relates to a remote sensing image water body extraction method and device based on an RGBW feature space graph cut algorithm, and belongs to the technical field intersecting remote sensing technology and water conservancy technology. Background technique [0002] Remote sensing technology has the advantages of wide observation range, large amount of information, fast information acquisition, short update cycle, and saving manpower and material resources. The automatic extraction technology of water body information in remote sensing images has become an important tool for water resource investigation, utilization, water dynamic change monitoring, water conservancy planning evaluation, etc. A powerful tool for research. Water bodies, land, vegetation, etc. have different reflection, absorption, and transmission characteristics of solar radiation, and the differences are more obvious in remote sensing images. This characteristic is conduciv...

AI Technical Summary

Problems solved by technology

[0006] The boundary pixels of lakes, reservoirs, rivers and other water bodies are mainly composed of beaches and water bodies. Due to human activities, plankton growth, seasonal water level fluctuations, etc., the boundary areas of water bodies are complex and diverse, which increases the difficulty of water body extraction.

In the same image, the imaging characteristics of different water body units or their respective physical and chemical characteristics or the influence of the surrounding environment may not be balanced, and the water quality and sediment content inside and outside the water body are quite different. The water body index method or image classification method is effective for extracting the main part of the water body, but there are problems such as difficult determination of the water body index threshold, unsatisfactory water body boundary processing, misclassification of shoals into water bodies, and wrong extraction.

Images