Scopus Indexed Publications

Paper Details


Title
Assessing the effects of image alignment on bathymetry and zonation mapping of coral reefs using UAV
Author
, Mohammad Shawkat Hossain,
Email
Abstract

A comprehensive understanding of coral distribution and niche information, including bathymetry, and geomorphological zones is a fundamental issue in ecology. Unmanned aerial vehicles (UAVs) offer high resolution data, however the distorted ortho-image effect is considerable due to inaccurate orthorectification, which is mainly caused by image misalignment. How can the misaligned images be handled to create accurate DEM maps and bathymetry? There are numerous Structure from Motion (SfM) techniques for aligning UAV images, however their applicability is case specific. Coral habitat mapping is ultimately impacted by orthoimage quality, which has not been thoroughly examined in previous studies. Using Agisoft™ Metashape software, this study examined thirty combinations of image alignment setting (IAS) and key point limit (KPL), covering the two reefs located on Bidong island (located in the northeastern region of Peninsular Malaysia). The results of the completeness, geometric fidelity, data processing efficiency check, and statistical measures showed that: (1) IAS significantly affected orthoimage quality, while KPL had no discernible effects. The IAS settings were separated into three components: cubic, quadratic, and linear, using an orthogonal single degree of freedom contrast. The findings imply that the linear and quadratic components of the response function described it better than the cubic function. (2) Validated through relative water depth index (RWDI) derived from Sentinel-2 data and field data, the ‘high’ IAS and 5 k KPL found optimal for coral habitat mapping and classified patch (PC), branching (BC), and tabular corals (TC) into shallow, moderate and deep areas with higher accuracy. (3) Reef flat and fore reef zones were produced using these optimal alignment settings from DEM and these zones were represented by medium and deep RWDI, respectively. Bathymetric distribution and geomorphometric zonation will be useful in identifying the best strategic management to sustain the reef through real-time monitoring, restoration, and management.

Keywords
Journal or Conference Name
Remote Sensing Applications: Society and Environment
Publication Year
2025
Indexing
scopus