Method for recognizing seawater polluted area based on high-resolution remote sensing image and device

Abstract

A method for recognizing polluted seawater areas based on high-resolution remote sensing images includes performing automatic sea and land classification on a remote sensing image by using a supervised learning algorithm. The classification result reaches a higher precision level by processized iterative clustering compared to conventional methods. The method further includes combining a chlorophyll-associated normalized difference vegetation index, a brightness-associated normalized difference water shadow index, a segmentation-based image interpretation thought and a human visual saliency-based mechanism in remote sensing interpretation by virtue of a chlorophyll concentration difference of a seawater polluted area and surrounding seawater and a brightness difference of pollutant shadows. The seawater polluted area is extracted by threshold segmentation. An area where the water quality is good and a heavily polluted area are extracted. A pollution transition area is further extracted. The method provides convenience and an accurate reference for prevention and control of marine pollution.