Retrieval of Surface Waves Spectrum From UAV Nadir Video | IEEE Journals & Magazine | IEEE Xplore

Drone Video Processing Makes Wave Measurements More Accessible and Affordable

Scientists have developed an economical method for measuring ocean wave patterns using ordinary commercial drones, potentially transforming how we monitor coastal conditions. The research, published in IEEE Transactions on Geoscience and Remote Sensing, demonstrates that consumer-grade drones can effectively measure wave spectra by analyzing video footage of light reflecting off the water's surface.

"This is a game-changing approach that could significantly increase our wave monitoring capabilities at a fraction of the cost of traditional methods," explains Dr. Aviv Solodoch, lead author of the study from The Hebrew University of Jerusalem.

The researchers validated their technique through multiple sea trials off the Israeli coast, comparing drone measurements with data from traditional wave buoys and bottom-mounted sensors. The results showed strong agreement across different conditions, with wave height measurements accurate within 30 centimeters.

Application to Southern California:
This technology could be particularly valuable for Southern California's complex coastline, where wave conditions can vary significantly over short distances. While the region already has several wave buoys, drone-based measurements could:

1. Fill gaps between existing buoy locations to better understand local wave patterns
2. Provide rapid deployment during critical events like storms or high surf
3. Monitor specific coastal features like reefs or harbor entrances
4. Support surf forecasting with higher spatial resolution data
5. Help assess beach erosion risks by measuring waves closer to shore

The method's low cost (using drones priced at several thousand dollars) and operational simplicity make it feasible for local agencies, universities, and coastal managers to supplement existing wave monitoring networks. This could be especially useful for studying how waves interact with Southern California's numerous coastal structures and varying bathymetry.

However, the researchers note some limitations - the drones can only measure relatively short-period waves, and local flight regulations would need to be considered. Despite these constraints, the technology offers promising opportunities for enhancing coastal monitoring in Southern California.

Validation

The data has been validated in several ways:

1. Validation Through Direct Comparison:
- The researchers conducted four separate sea trials off the Israeli coast
- They compared their drone measurements against:
  - Bottom-mounted acoustic Doppler current profilers (ADCP)
  - Wave buoys
  - Wave reanalysis data from the Mediterranean Sea

2. Specific Validation Results:
- Significant Wave Height (SWH) measurements:
  - Deviation was less than 30% or 0.15m in all cases
  - In 3 out of 4 cases, deviation was less than 7% or 0.05m
- Wave Period measurements:
  - Agreed within 15% or 0.5 seconds
- Wave Direction measurements:
  - Within 40° in all cases
  - Within 30° in 3 out of 4 cases

3. Quality of Validation:
- The deviations between drone and traditional measurements were within the typical range of differences seen between different types of wave sensors (like between buoys and ADCPs) in previous studies
- The researchers used robust statistical methods and considered measurement uncertainties
- They conducted trials under different conditions and at different distances from shore

However, it's important to note some limitations:
- The validation was conducted only in Mediterranean Sea conditions
- The longest trial period was relatively short compared to traditional wave monitoring
- High winds and breaking wave conditions weren't fully tested
- The technology hasn't yet been validated in Pacific Ocean conditions like those found off Southern California

The researchers acknowledge that further validation would be beneficial, particularly in different ocean conditions and during more extreme weather events.

Mix of Sea and Swell

This is an important limitation of the current research. Looking carefully at the paper, there are several key concerns about applying this technology to California's complex wave conditions:

1. Technical Limitations:
- The method is limited by drone altitude and field of view
- At typical operating heights (50-120m), it can only measure waves with periods up to about 5-6 seconds
- California regularly experiences:
  * Long-period groundswell (12-20 second periods)
  * Wind waves (4-8 seconds)
  * Mixed sea states with multiple wave trains

2. Validation Gap:
- The study was conducted in the Mediterranean Sea which typically has:
  * Shorter wave periods
  * Limited fetch
  * No significant groundswell
- They did test one case with two wave trains (3s and 6s periods)
- But this doesn't approach the complexity of California conditions

3. Specific California Challenges:
- Common mixed conditions like:
  * South swell + northwest wind waves
  * Long-period northwest groundswell + afternoon sea breeze chop
  * Multiple simultaneous groundswells
- These complex wave patterns might make it harder to:
  * Separate different wave components
  * Accurately measure wave heights
  * Determine dominant wave directions

The authors mention this limitation indirectly when discussing the need for future validation in different conditions. For California applications, this technology might best serve as a supplement to existing buoy networks, perhaps focusing on measuring local wind waves or monitoring specific nearshore areas, rather than as a replacement for buoys that can measure the full spectrum of wave conditions.

Additional validation studies would be needed in Pacific conditions before relying on this method for operational wave monitoring in California.

Independent Verification

Reviewing the paper's technical details:

Hardware Platforms:
- Two consumer drone models were tested:
  1. DJI Mavic 2 Zoom (used in Feb 8, 2022 cruise)
     * Used with minimal zoom (24mm)
     * Did not record camera angles in metadata
  2. DJI Phantom Pro 4 (used in other three trials)
     * Fixed lens (no zoom)
     * Recorded pitch, roll, yaw and height in video metadata

Video Configuration:
- 30 frames per second
- Video format and compression details not specified
- No raw video data mentioned as being publicly available

Processing:
- Built as an extension to the CopterCurrents MATLAB package
- Available upon request according to acknowledgments: "The new software can be received upon request"
- GitHub repository for base package: https://github.com/RubenCarrascoAlvarez/CopterCurrents

Notably absent from the paper:
1. Complete camera settings (aperture, shutter speed, ISO)
2. Specific video encoding parameters
3. Access to test datasets
4. Public repository of the processing code
5. Detailed processing parameters for reproducibility

For independent verification, one would need:
- The actual processing code (not currently public)
- Sample video datasets (not provided)
- Complete camera configuration details (not fully specified)
- Processing parameters used in the validation (not fully detailed)

I'd say the paper lacks sufficient detail and public artifacts for complete independent verification of the results. While the base CopterCurrents package is available, the specific extensions and modifications for wave spectrum analysis are not publicly accessible.

Retrieval of Surface Waves Spectrum From UAV Nadir Video | IEEE Journals & Magazine | IEEE Xplore

A. Solodoch, Y. Toledo, V. Grigorieva and Y. Lehahn, "Retrieval of Surface Waves Spectrum From UAV Nadir Video," in IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1-14, 2025, Art no. 4201914, doi: 10.1109/TGRS.2025.3536378.

 

Abstract: Sea surface wave spectrum measurements are necessary for a host of basic research questions and for engineering and societal needs. However, most measurement techniques require great investment in infrastructure and time-intensive deployment techniques. We propose a new approach to wave measurement from standard video footage recorded by low-cost unmanned aerial vehicles (UAVs). We address UAV nadir imagery, which is particularly simple to obtain operationally. The method relies on the fact that the optical contrast of surface gravity waves is proportional to their steepness. We present a robust methodology of regularized inversion of the optical imagery spectra, resulting in retrieval of the 3-D wavenumber-frequency sea surface height (SSH) spectrum. The system was tested in several sea trials and in different bathymetric depths and sea state conditions. The resulting wave bulk parameters and spectral characteristics are in good agreement with collocated measurements from wave buoys and bottom-mounted acoustic sensors. Simple deployment, mobility, and flexibility in spatial coverage show the great potential of UAVs to significantly enhance the availability of wave measurements. 

keywords: {Optical surface waves;Sea surface;Surface waves;Sea measurements;Autonomous aerial vehicles;Optical reflection;Optical sensors;Cameras;Brightness;Optical imaging;Low-cost observing systems;ocean wave sensing;optical sensing;surface wave spectra;unmanned aerial vehicles (UAVs)}, 

URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10857442&isnumber=10807682

Retrieval of Surface Waves Spectrum from UAV Nadir Video

Aviv Solodoch

https://doi.org/10.36227/techrxiv.170854748.81779222/v1 :Feb 20, 2024

 Background of the study:The paper focuses on retrieving the surface wave spectrum from videos recorded by low-cost unmanned aerial vehicles (UAVs). Measuring surface wave spectrum is important for understanding air-sea interaction and for various applications, but traditional methods require expensive and complex equipment. The researchers propose a new approach to obtain the full 3D wavenumber-frequency spectrum of the sea surface height using standard UAV video footage.

Research objectives and hypotheses:
The main objective is to develop a robust methodology to retrieve the 3D wavenumber-frequency sea surface height spectrum from UAV nadir video data. The key hypothesis is that the optical contrast of surface gravity waves is proportional to their steepness, allowing the wave spectrum to be inferred from the video brightness spectrum.

Methodology:
The method relies on the fact that the optical contrast of surface gravity waves is proportional to their steepness. The researchers present a process to invert the optical imagery spectra and retrieve the 3D wavenumber-frequency sea surface height spectrum. This includes estimating the reference skylight brightness, accounting for non-uniform skylight gradients, and reducing the dimensionality of the spectrum while correcting for Doppler shifts.

Results and findings:
The researchers conducted four sea trials in different locations and conditions. They found that the UAV-derived bulk wave parameters, such as significant wave height, mean wave period, and mean wave direction, agreed well with collocated in-situ measurements from wave buoys and acoustic sensors. The UAV-derived frequency spectra also showed good agreement with the in-situ sensors within the overlapping frequency range.

Discussion and interpretation:
The good agreement between the UAV-derived and in-situ measurements validates the proposed methodology. The UAV system provides high-resolution surface elevation information in both time and space, allowing the construction of the full 3D wavenumber-frequency spectrum, which is not easily obtained from traditional in-situ sensors.

Contributions to the field:
The study presents a new approach to measure surface wave spectra using low-cost, consumer-grade UAVs, which can significantly enhance the availability of wave measurements globally. This is an important advancement, as traditional wave measurement techniques require expensive infrastructure and complex deployment.

Achievements and significance:
The proposed methodology allows UAVs to perform as high-quality wave spectrum sensors without any modifications, opening the door to a much higher volume of wave measurements worldwide. This can be transformative for scientific research and various applications, such as oil spill response, marine safety, and search and rescue operations.

Limitations and future work:
Further development is needed to increase the accuracy and applicability of the method, such as integrating skylight polarization measurements and extending the methodology to non-nadir imagery. The performance of the method in high wind and wave conditions also requires investigation through numerical simulations.