Quick extraction method for water body

Abstract

The invention relates to the field of water extraction, in particular to a quick water extraction method. The water body quick extraction method is applied to a remote sensing image, and comprises the following steps: S1, obtaining a normalized water body index; s2, obtaining a normalized blue water body index; s3, selecting a plurality of water body feature points and non-water body feature points, and constructing an optimal classification line; s4, acquiring a fused vertical water body index according to the optimal classification line; and S5, presetting a threshold value, and extracting the water body according to the threshold value and the fused vertical water body index. The technical scheme of the invention has the beneficial effects that according to the water body quick extraction method provided by the invention, the difference between the building and the water body can be enhanced by obtaining the normalized blue water body index, and the difference between the element gray values between the water body and the non-water body can be further enhanced by obtaining the fused vertical water body index; fine water bodies are quickly recognized and extracted, the water body extraction accuracy is improved, and the water body extraction speed is increased.

Description

technical field [0001] The invention relates to the field of water extraction, in particular to a rapid extraction method for water. Background technique [0002] The water body index method uses the multi-band information of remote sensing images to construct a band operation function that distinguishes between water bodies and non-water bodies, so as to achieve the purpose of extracting water bodies. [0003] However, the mainstream water body index method used has the following problems: ①High-resolution satellite images contain rich ground object information, and the traditional extraction method is easily disturbed by information such as dark buildings, building shadows, and bare soil, resulting in small rivers, small fish, etc. It is difficult to extract ponds and small pit ponds; ② high-resolution satellite images only have red band (R), green band (G), blue band (B), and near-red band (NIR), but most of the existing water index calculations must use Short-wave infra...

AI Technical Summary

Problems solved by technology

[0003] However, the mainstream water body index method used has the following problems: ①High-resolution satellite images contain rich ground object information, and the traditional extraction method is easily disturbed by information such as dark buildings, building shadows, and bare soil, resulting in small rivers, small fish, etc. It is difficult to extract ponds and small pit ponds; ② high-resolution satellite images only have red band (R), green band (G), blue band (B), and near-red band (NIR), but most of the existing water index calculations must use Short Wave Infrared (SWIR)

Images