Yuri Rzhanov

Automated segmentation of seafloor bathymetry from multibeam echosounder data using local Fourier histogram texture features

Abstract Patterns of seafloor topography represent regions of geomorphological feature types and the physiography governing the spatial distributions of benthic habitats. Topographic variability can be considered seafloor texture and can be remotely sensed by acoustic and optical devices. Benthic habitat delineations often involve distinctions based upon seafloor morphology and composition based upon acoustic data maps that are ground-truthed by optical imaging tools. Habitat delineations can be done manually, however, automation of the procedure could provide more objectivity and reproducible map products. Recently a technique using Fourier transforms (FT) to produce texture features called local Fourier histograms (LFH) has been used successfully to classify standard textures in grayscale images and automatically retrieve digital images from archives according to texture content [Zhou, F., Feng, J., Shi, Q., 2001. Texture feature based on local Fourier transform, ICIP Conference Proceedings, IEEE 0-7803-6725-1/01.]. We implemented a modified form of that approach by varying the spatial scales at which local Fourier histograms were calculated. A modified LFH texture feature classification technique was applied to multibeam echosounder (MBES) data from Piscataqua River, New Hampshire, USA, for automatic delineation of a seafloor topographic map into regions of distinct geomorphology and apparent benthic habitats. Automated segmentations were done by the LFH method on 1-m gridded MBES data, applying the local Fourier transform, used to generate the LFH, at spatial scales from 1 to 5 m. Seven seafloor texture classes were identified, corresponding to the primary substrate types and configurations in the study area as well as some previously unidentified regions and transitional zones. The texture regions serve as a physical habitat model for the seafloor, a basis for predicting benthic faunal inhabitants, their areal distributions, and serving as sampling strata for ground-truthing efforts.

Underwater video mosaicing for seabed mapping

This paper presents improved techniques and applications in mosaicing of underwater video images. The applications are of use to many marine scientists. High resolution seabed maps are created. The improvements in the processing relate to the removal of interframe variability and lighting, speed up of the mosaicing process and improved accuracy in the estimation of the transformation parameters. Results are presented for real data acquired under a variety of circumstances and scenes.

Acoustic estimates of methane gas flux from the seabed in a 6000 km2 region in the Northern Gulf of Mexico

Seeps of free methane gas escaping the seabed can be found throughout the ocean basins. To understand the role of methane gas seeps in the global carbon cycle—including both gas added to the atmosphere and that which is dissolved and potentially oxidized in the ocean volume—it is important to quantify the amount of methane escaping the seabed. Few large-scale mapping projects of natural methane seeps have been undertaken, however, and even among these, quantitative estimates of flux are rare. Here we use acoustic mapping techniques to survey 357 natural methane seeps in a large region (6000 km2) of the northern Gulf of Mexico and outline a general approach for methane seep mapping using a combination of multibeam and split-beam echo sounders. Using additional measurements collected with a remotely operated vehicle (ROV) together with the acoustic mapping results, we estimate the total gas flux within the 6000 km2 region to be between 0.0013 and 0.16 Tg/yr, or between 0.003 and 0.3% of the current estimates for global seabed methane seepage rates.

Deep-sea Geo-referenced Video Mosaics

Seafloor imagery is an important tool for scientists engaged in quantifying geological and biological processes operating on the deep-ocean floor. Large-scale mosaics of seafloor imagery have significant advantages over individual still photographs and video footage as they are able to capture large areas while retaining sufficient resolution to identify small-scale features. However, the process of constructing a mosaic is usually limited to a single panorama or transects from still imagery, and often requires manual control by the user. The goal of this paper is to describe a methodology that allows for utilization of navigation data for robust automatic construction and geo-referencing of video mosaics. This methodology is designed specifically for the submersible Alvin, but our methodology may be modified for use with other underwater vehicles and is broadly applicable to the creation of mosaics in a variety of seafloor settings

Measurement of micro-bathymetry with a GOPRO underwater stereo camera pair

A GO-PRO underwater stereo camera kit has been used to measure the 3D topography (bathymetry) of a patch of seafloor producing a point cloud with a spatial data density of 15 measurements per 3 mm grid square and an standard deviation of less than 1 cm A GO-PRO camera is a fixed focus, 11 megapixel, still-frame (or 1080p high-definition video) camera, whose small form-factor and water-proof housing has made it popular with sports enthusiasts. A stereo camera kit is available providing a waterproof housing (to 61 m / 200 ft) for a pair of cameras. Measures of seafloor micro-bathymetry capable of resolving seafloor features less than 1 cm in amplitude were possible from the stereo reconstruction. Bathymetric measurements of this scale provide important ground-truth data and boundary condition information for modeling of larger scale processes whose details depend on small-scale variations. Examples include modeling of turbulent water layers, seafloor sediment transfer and acoustic backscatter from bathymetric echo sounders.

Seafloor video mapping: modeling, algorithms, apparatus

This paper discusses a technique used for construction of high-resolution image mosaic from a video sequence and the synchronously logged camera attitude information. It allows one to infer geometric characteristics of the imaged terrain and hence improve the mosaic quality and reduce the computational burden. The technique is demonstrated using numerical modeling and is applied to video data collected on Rainsford Island, Mass. Calculation of the transformation relating consecutive image frames is an essential operation affecting reliability of the whole mosaicing process. Improvements to the algorithm are suggested, which significantly decrease the possibility of convergence to an inappropriate solution.

Sensor-assisted video mosaicing for seafloor mapping

This paper discusses a proposed processing technique for combining video imagery with auxiliary sensor information. The latter greatly simplifies image processing by reducing complexity of the transformation model. The mosaics produced by this technique are adequate for many applications, in particular habitat mapping. The algorithm is demonstrated through simulations and hardware configuration is described.

Deep-sea image processing

High-resolution seafloor mapping often requires optical methods of sensing, to confirm interpretations made from sonar data. Optical digital imagery of seafloor sites can now provide very high resolution and also provides additional cues, such as color information for sediments, biota and divers rock types. During the cruise AT11-7 of the Woods Hole Oceanographic Institution (WHOI) vessel R/V Atlantis (February 2004, East Pacific Rise) visual imagery was acquired from three sources: (1) a digital still down-looking camera mounted on the submersible Alvin, (2) observer-operated 1-and 3-chip video cameras with tilt and pan capabilities mounted on the front of Alvin, and (3) a digital still camera on the WHOI TowCam (Fornari, 2003). Imagery from the first source collected on a previous cruise (AT7-13) to the Galapagos Rift at 86/spl deg/W was successfully processed and mosaicked post-cruise, resulting in a single image covering area of about 2000 sq.m, with the resolution of 3 mm per pixel (Rzhanov et al., 2003). This paper addresses the issues of the optimal acquisition of visual imagery in deep-sea conditions, and requirements for on-board processing. Shipboard processing of digital imagery allows for reviewing collected imagery immediately after the dive, evaluating its importance and optimizing acquisition parameters, and augmenting acquisition of data over specific sites on subsequent dives. Images from the deepsea power and light (DSPL) digital camera offer the best resolution (3.3 Mega pixels) and are taken at an interval of 10 seconds (determined by the strobes recharge rate). This makes images suitable for mosaicking only when Alvin moves slowly (/spl Lt/1/4 kt), which is not always possible for time-critical missions. Video cameras provided a source of imagery more suitable for mosaicking, despite its inferiority in resolution. We discuss required pre-processing and image enhancement techniques and their influence on the interpretation of mosaic content. An algorithm for determination of camera tilt parameters from acquired imagery is proposed and robustness conditions are discussed.

Position, Orientation and Velocity Detection of Unmanned Underwater Vehicles (UUVs) Using an Optical Detector Array

This paper presents a proof-of-concept optical detector array sensor system to be used in Unmanned Underwater Vehicle (UUV) navigation. The performance of the developed optical detector array was evaluated for its capability to estimate the position, orientation and forward velocity of UUVs with respect to a light source fixed in underwater. The evaluations were conducted through Monte Carlo simulations and empirical tests under a variety of motion configurations. Monte Carlo simulations also evaluated the system total propagated uncertainty (TPU) by taking into account variations in the water column turbidity, temperature and hardware noise that may degrade the system performance. Empirical tests were conducted to estimate UUV position and velocity during its navigation to a light beacon. Monte Carlo simulation and empirical results support the use of the detector array system for optics based position feedback for UUV positioning applications.

Modeling Uncertainty in Photogrammetry-Derived National Shoreline

Tidally-referenced shoreline data serve a multitude of purposes, ranging from nautical charting, to coastal change analysis, wetland migration studies, coastal planning, resource management and emergency management. To assess the suitability of the shoreline for a particular application, end users require reliable estimates of the uncertainty in the shoreline position. Previous studies on modeling uncertainty in shoreline mapping from remote sensing data have focused on airborne light detection and ranging; to date, these methods have not been extended to aerial imagery and photogrammetric shoreline mapping, which remains the primary shoreline mapping method used by the National Geodetic Survey. The aim of this article is to develop and test a rigorous total propagated uncertainty model for shoreline compiled from both tide-coordinated and non-tide-coordinated aerial imagery using photogrammetric methods. The uncertainty model is developed using data from a study site in northeast Maine. For the study area, the standard uncertainty was found to be ∼3.2–3.3 m, depending on whether the imagery was tide coordinated or not. The uncertainty model developed in this paper can easily be extended from the study area to other areas and may facilitate estimation of uncertainty in inundation models and marsh migration models.