Geographic Object Based Image Analysis (GEOBIA) for Mangrove Canopy Delineation Using Aerial Photography

Abstract

Mangrove canopy cover is one of the important indicators for monitoring mangrove ecosystem's health. Measuring and monitoring mangrove canopy cover using direct field measurement is tedious and inefficient work. This study implements the advanced geographic object-based image analysis (GEOBIA) approach to a very high-spatial resolution aerial photograph taken from unmanned aerial vehicle (UAV). This new technology has ability to capture mangrove canopy information in the field rapidly and present it in a spatially explicit manner. The aims of this study are to (1) establish a rule set to discriminate mangrove canopy to other coastal objects from UAV images; and (2) map mangrove canopy and calculate its accuracy. The image data used in this study was acquired on 27 April 2018, covering part of dwarf Avicennia marina and Ceriops tagal stands in Karimunjawa Island, Jepara, Central Java. Based on this image the rule set was then developed to delineate mangrove canopy borders by considering the color and shape properties of the object of interest on the image. An iterative classification process was conducted to find the most operational rule set to map the targeted object. Once the map produced, the area-based accuracy assessment was then applied to assess the quality of the canopy delineation. The results of this study show that UAV image - although handicapped with limited spectral information - is a valuable input for GEOBIA to delineate and map mangrove canopy borders with high level of accuracy. This study contributes to the development of canopy delineation methods using the emerging remote sensing technology and mapping approach.