Global scale remote sensing image water body intelligent extraction method based on multiple indexes

Abstract

The invention discloses a global scale remote sensing image water body intelligent extraction method based on multiple indexes. The invention belongs to the technical field of remote sensing image information mining, in particular to Landsat remote sensing image water body rapid extraction. A topographic index and normalized indices such as a water body, vegetation and architecture are comprehensively adopted, the problem of confusion between water body and mountain shadow, urban area, cloud shadow, ice and snow in the process of extracting satellite remote sensing images water body information. Moreover, the complexity of calculation is greatly simplified by using the index method, and the calculation speed of the algorithm can be effectively improved. At the same time, prior knowledge and a decision tree method are used to establish a rule model, thresholds are self-adaptively determined, strong universality is achieved, and the method can be well applied to the globally regional water information extraction. The method can rapidly, accurately and intelligently produce global epicontinental water products, and can be used for water resources monitoring, global change research, etc.

Description

technical field [0001] The technical field of the present invention is the mining of remote sensing image information, and specifically relates to the rapid and intelligent extraction of remote sensing image water bodies on a global scale. Background technique [0002] Water resources are an irreplaceable strategic resource for human survival and social development, and an indispensable and important basic information for global resource security and management, climate change research, and dynamic monitoring of the ecological environment. How to establish an effective method to scientifically, accurately and quickly extract the information of global land surface water bodies, and timely grasp the change information of land surface water bodies has become a key issue in the research of global changes such as water cycle. Because remote sensing technology has the characteristics of large-area synchronous observation, timeliness, comprehensiveness and comparability of data, an...

AI Technical Summary

Problems solved by technology

However, for global-scale land surface water mapping, the current sample library can only refer to existing land surface water products

However, water bodies are greatly affected by seasons, that is to say, compared with water bodies in different time phases, the samples are inaccurate to a certain extent, which easily leads to misclassification of results

Images