Mapping inland aquaculture from orthophoto and LiDAR data using object-based image analysis

Previous studies demonstrated the use of satellite images, such as Satellite imaging radar, Landsat Thematic Mapper and WorldView-2, in able to map coastal aquaculture. With the advent of LiDAR technology in the Philippines, methods have been developed to extract fish ponds, fish corrals, fish pens and fish cages using height information (DSM and DTM) and RGB-derived layers from orthophoto taken during the flight mission. In this paper, the effectiveness of LiDAR and orthophoto in mapping freshwater culture facilities found in Tikub Lake Tiaong, Quezon, Philippines was explored. Moreover, Canny's edge detection algorithm was used to delineate the boundaries of fish pens and fish cages. Using object-based image analysis and applying rule-based classification, the aquaculture extraction was met, enabling the detection of the fish cages together with the cottages or built-up used by fishermen in the lake. The output map can then be used as a baseline information in managing aquaculture areas specially that these anthropogenic activities have caused devastating fish kills in the Philippines.