Development of a method to detect coral bleaching using airborne hyperspectral sensor

In recent years, coral bleaching is becoming one of the environmental problems all over the world. In order to monitor this bleaching, remote sensing is expected as a effective tool. Although existing multispectral imager can detect coral bleaching, it cannot detect whether coral became death or living after bleaching. In this study, we used airborne hyperspectral imager and classified bottom type to detect coral death or living. The classification was conducted by ISODATA classification using bottom index and first derivative imageries calculated from hyperspectral data. We verified the accuracy of classification with thirty ground truths and found that twenty-four points were correctly classified and six points were misclassified.