Application of multispectral remote sensing technology in surface water body extraction

This paper addresses an improved water-body extraction method by the multispectral remote sensing image of four-bands. Firstly, traditional multispectral water body extraction technologies are reviewed, and their features are compared. It shows that finding the most difference of the reflection ratio of the water body and the background is the core to improve the identification rate. Furthermore, combing data's feature of GF-01 and GF-02 satellites, an improved normalization index method by introducing the weighted average of the blue and green light bands was proposed. Finally, the comparative results illustrating the improvement of water body extraction are provided and analyzed.