Home | Contact ST  

Feature Article

3D Seafloor Mapping With Automated Data Analysis
The Generation and Application of 3D Color Reconstructions For Quantitative Algorithm-Based Analysis

AUTHORS:
Feature Author
Adrian Bodenmann
Project Researcher
Feature Author
Blair Thornton
Associate Professor
Feature Author
Tamaki Ura
Professor
Underwater Technology Research Center
Institute of Industrial Science
The University of Tokyo
Tokyo, Japan


The AUV Tuna-Sand with the equipment for recording data to generate 3D reconstructions.
Imaging of the seafloor has been one of the major applications of underwater robots over the past decade. Different methods of combining individual photos into continuous maps have evolved, and methods to process images in 3D are becoming more common, making it possible to visualize lifelike reconstructions of the bottom of the ocean.

While 3D reconstructions provide a natural medium to show the seafloor and present a reference to overlay data measured by other sensors, these data can also be directly analyzed by algorithms to extract scientifically useful information.

A method based on laser scanning, which generates high-resolution bathymetry maps, was chosen and extended to match color information to generate 3D reconstructions. This approach is relatively simple in terms of its hardware and software requirements because the required images can be obtained by a single camera, and all calculations are feed-forward with no feature-based matching.


Hardware and Mapping Method
The underwater vehicle, which can be an AUV or ROV, is equipped with a camera, a sheet laser, lights and navigation sensors such as a Doppler velocity log (DVL) with an inertial navigation system (INS) compass and a depth meter. A computer in a pressure-tight housing saves time-stamped images from the camera and logs the navigational measurements.

The lights are directed to illuminate only a part of the camera’s field of view, and the sheet laser projects a laser line in the unilluminated area, perpendicular to the forward direction of the vehicle. The camera is mounted at an angle and with an offset to the sheet laser, and the images taken of the laser line reveal the topography of the seafloor. The algorithm analyzes all images in post-processing and extracts the laser line with subpixel resolution, based on which shape of the seafloor is calculated. Each image provides one line of bathymetry points, and by combining lines of data from multiple images taken while the vehicle moved forward, a bathymetry map of the scanned area is generated.

In the second step, the color of each bathymetry point is determined. To do so, a corresponding image is selected where the spot on the seafloor appears in the illuminated section of the image based on the vehicle’s navigation data. Through vector-based calculations and by taking the vehicle’s position and orientation into account, the pixel coordinates are calculated, and the RGB values of the pixel corresponding to each bathymetry point are assigned.

Special attention is paid to accurate color reproduction. As light travels through water, different wavelengths are attenuated by different amounts, where red is absorbed more severely than blue. The color of a spot on the seafloor as it appears in the camera image depends not only on the actual color of the seafloor but also on the distance and the angle from the light’s axis (the center of the light cone is brighter than the outside). The nature of the mapping methods allows for accurate color mapping since the path that the light took can be calculated for every single pixel, and hence the shift in brightness and the color balance can be corrected before the RGB values are assigned to each bathymetry point.

Finally, the point cloud of colored bathymetry points is meshed and written to a file that can be displayed with a suitable viewer.


Results from Deployment at Kagoshima Bay
The mapping system has been deployed at sea several times using the AUV Tuna-Sand developed at the Underwater Robotics & Application Laboratory at the University of Tokyo and the work-class ROV Hyper-Dolphin, operated by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC).

The system, mounted on the AUV Tuna-Sand, was used to map a 33-by-23-meter zone during a survey of a hydrothermally active area at 200 meters depth in southern Japan’s Kagoshima Bay. The area, located at a latitude of 31 degrees 40 minutes north and a longitude of 130 degrees 46 minutes east, lends itself well to mapping as it is mostly flat apart from a single hydrothermal chimney in the mapped region. To continue this article please click here.



Adrian Bodenmann is a project researcher at the Underwater Technology Research Center of the University of Tokyo. He received a master’s in microengineering from the École Polytechnique Fédérale de Lausanne. He presently works on the seafloor mapping project and the development of a deep-sea-going AUV.


Blair Thornton is an associate professor at the Underwater Technology Research Center of the University of Tokyo. He received a Ph.D. in underwater robotics from the University of Southampton. His research interests involve the development of in-situ sensors for integrated acoustic, visual and chemical marine surveys.


Tamaki Ura is the director of the Underwater Technology Research Center and is a professor at the University of Tokyo. He founded the Underwater Robotics and Application Laboratory and has developed various innovative ocean-going AUVs, which have performed successful dives to hydrothermal and gas hydrate fields.





-back to top-

-back to to Features Index-

Sea Technology is read worldwide in more than 110 countries by management, engineers, scientists and technical personnel working in industry, government and educational research institutions. Readers are involved with oceanographic research, fisheries management, offshore oil and gas exploration and production, undersea defense including antisubmarine warfare, ocean mining and commercial diving.