11:45 AM - 12:00 PM
[AOS20-05] Investigation of Sunglint Reduction Methods in Seawood Bed Mapping Using a Polarizing Filter
Keywords:seaweed bed monitoring, UAV, polarizing filter, sunglint
In recent years, significant variations in coastal seaweed bed distribution have been observed, with their decline severely impacting aquatic habitats and fisheries. Over the past 50 years, kelp forests have decreased in 38% of areas while increasing in only 27%, as reported in previous studies. Given the substantial interannual fluctuations in these beds, high-frequency, high-precision mapping is essential for monitoring their dynamics, conserving coastal ecosystems, and implementing effective management strategies.
Traditionally, submerged aquatic vegetation mapping has utilized satellite imagery, airborne optical sensors, acoustic devices, underwater cameras, and direct visual observations. However, UAVs have recently gained attention as a cost-effective alternative balancing spatial resolution and operational efficiency. In UAV-based underwater observations, sunglint—caused by sunlight reflecting off the water surface when the sensor’s viewing angle aligns with the reflection angle—poses a significant challenge. Studies indicate that sunglint degrades data quality and compromises analytical accuracy. Its severity depends on solar altitude, with an ideal range of 6.5° to approximately 40° suggested for reliable seaweed mapping under a 94° field of view. However, this constraint limits observation time, reducing the flexibility of large-scale monitoring.
This study aims to mitigate sunglint in UAV-based seaweed bed mapping using a polarizing filter, thereby easing observation time restrictions. To achieve this, it is necessary to clarify how sunglint reduction effectiveness varies across wavelengths depending on solar angle. An indoor experiment was conducted using an aquarium with submerged aquatic plants, where the angular dependence of the reflection spectrum was measured with a spectrometer (420–840 nm) to evaluate the extent of time constraint mitigation.
For the experiment, a halogen lamp simulated mirror-like reflection, with submerged aquatic plants placed in an indoor aquarium. The angles of the polarizing filter and halogen lamp’s illumination were independently varied, and reflection spectra from the water surface were measured. Additionally, as sunglint is affected by wind and waves, spectral differences between artificial wind conditions and still water were analyzed. Detailed results will be presented at the conference.
This study was partially supported by a grant from the Telecommunication Advancement Foundation for the project “Development of an Education and Training Method for Solving SDG Challenges Using Both ICT and Hands-on Approaches” (Shiga University of Medical Science, FY 2024), as well as by support from the NPO Super Scientist Program Plus.
Traditionally, submerged aquatic vegetation mapping has utilized satellite imagery, airborne optical sensors, acoustic devices, underwater cameras, and direct visual observations. However, UAVs have recently gained attention as a cost-effective alternative balancing spatial resolution and operational efficiency. In UAV-based underwater observations, sunglint—caused by sunlight reflecting off the water surface when the sensor’s viewing angle aligns with the reflection angle—poses a significant challenge. Studies indicate that sunglint degrades data quality and compromises analytical accuracy. Its severity depends on solar altitude, with an ideal range of 6.5° to approximately 40° suggested for reliable seaweed mapping under a 94° field of view. However, this constraint limits observation time, reducing the flexibility of large-scale monitoring.
This study aims to mitigate sunglint in UAV-based seaweed bed mapping using a polarizing filter, thereby easing observation time restrictions. To achieve this, it is necessary to clarify how sunglint reduction effectiveness varies across wavelengths depending on solar angle. An indoor experiment was conducted using an aquarium with submerged aquatic plants, where the angular dependence of the reflection spectrum was measured with a spectrometer (420–840 nm) to evaluate the extent of time constraint mitigation.
For the experiment, a halogen lamp simulated mirror-like reflection, with submerged aquatic plants placed in an indoor aquarium. The angles of the polarizing filter and halogen lamp’s illumination were independently varied, and reflection spectra from the water surface were measured. Additionally, as sunglint is affected by wind and waves, spectral differences between artificial wind conditions and still water were analyzed. Detailed results will be presented at the conference.
This study was partially supported by a grant from the Telecommunication Advancement Foundation for the project “Development of an Education and Training Method for Solving SDG Challenges Using Both ICT and Hands-on Approaches” (Shiga University of Medical Science, FY 2024), as well as by support from the NPO Super Scientist Program Plus.