Automatic extraction of large-scale aquaculture encroachment areas using Canny Edge Otsu algorithm in Google earth engine – the case study of Kolleru Lake, South India

Abstract

The aquaculture expansion has made significant contributions to global food security, socio-economic development and, if implemented sustainably, can help preserve stable coastal environments. This study explicitly details the rapid expansion of large-scale aquaculture growth across the Kolleru and Upputeru regions of South India. We developed a novel classification method for automated extraction of aquaculture ponds in the Kolleru zone using the Canny Edge-Otsu algorithm to segment and extract the ponds applied to SAR-VV images in Google Earth Engine. This approach enables the area estimation of dense aquaculture ponds are essential for monitoring and management purposes. The results indicated that this method could effectively map the aquaculture ponds and the overall accuracy achieved in 2020 for the Kolleru and Upputeru areas by 90.6% and 95.7%, respectively. The aquaculture maps of this study can help government organizations, resource managers, stakeholders, and decision-makers understand the dynamics and plan sustainable measures in this area.