Information extraction of Ulva Prolifera from coastal landscape using UAV m ultispectral remote sensing images

Since 2007, green tides(also called Ulva prolifera) occurred every summer in the Yellow Sea, causing ecological problems in the coastal environment of Shandong Peninsula . A large number of Ulva prolifera on shore will rot and stink if not handled in time,which seriously affects the tourism and the health of residents in coastal landscape. In order to improve the accuracy of monitoring green tide disasters, and to improve the efficiency of the cleaning up and disposal of Ulva prolifera at key prevention and control area, In this study, the high-precision image of UAV is used to monitor the green tide disaster in Yintan landscape of Rushan City. With the spectral characteristics of Ulva prolifera and coastal vegetation measured by spectroradiometer, four vegetation indices were used to analyze and identify the Ulva prolifera and coastal vegetation, and to verify the extraction of Ulva prolifera and coastal vegetation under different vegetation indices, and based on this extraction method, the biomass of coastal green tide algae was estimated. The results show that in the red-edge band,Ulva prolifera and coastal vegetation can be distinguished. MTCI(MERIS terrestrial chlorophyll index) is more suitale,with the accuracy of 91.3%, followed by SR_(redge),NDVI_(redge) and MSR_(redge),with the accuracy of 85.3%, 83.8% and 81.2%, respectively; Estimation model of biomass based on MTCI index showed that about 600 tons of Ulva prolifera were estimated in 300 m study area. An effective method for dynamic monitoring and cleaning up of green tide disaster is provided