Neural network classification of multibeam backscatter and bathymetry data from Stanton Bank (Area IV)

The paper presents an approach to automated seabed classification that incorporates spatially coincident bathymetric and backscatter data collected in multibeam surveys. The classification algorithm is a self-organising artificial neural network that can be used as a rapid classifier of grids of bathymetry (and attributes such as slope and roughness) and backscatter strength (and textures), or in a mode that uses both datasets at beam level to construct high spatial resolution classifications that preserve angular information in the backscatter. The latter mode requires processing of backscatter angular responses in a manner consistent with the essential physics of acoustic scattering from the seafloor.     

Developments in the application of multibeam sonar backscatter for seafloor habitat mapping

Human impacts on the seafloor environment have reached unprecedented levels. To facilitate ocean management and mitigate these impacts, there is a need to improve our understanding of seabed habitats. Recent developments in acoustic survey techniques, in particular multibeam echosounders (MBES), have revolutionised the way we are able to image, map and understand benthic ecosystems. Using MBES, it is now cost-effective to image large areas of the seafloor, and such surveys provide baseline data from which thematic maps of the seabed environment, including maps of benthic habitat, can be derived and interpreted in conjunction with in situ ground-truthing data. This paper provides an overview of recent developments in the application of MBES for seafloor habitat studies, with a focus on the use of backscatter data for surficial geology and habitat mapping. In March 2006, a MBES backscatter workshop brought together a number of international research teams/groups working on novel methods for interpretive/classification routines for segmentation of the backscatter into acoustic classes with the potential to facilitate the delineation of seabed geological and habitat characteristics. This paper introduces the common data set used as part of the workshop, sets out the research context in which the different studies were conducted, and outlines the main themes of the papers presented in Section 6.     

Acquisition and processing of backscatter data for habitat mapping - Comparison of multibeam and sidescan systems

Often marine habitat surveys use multibeam bathymetry systems to model the seafloor. This describes the morphology but not the terrain lithology or substrate. Backscatter imagery helps the interpreter to better classify the physical environment that may support a particular biological community. In this contribution, the acquisition performance of both multibeam and sidescan sonar backscatter imagery are contrasted and examples shown. The logistical factors affecting the two systems during surveying are discussed and data from both systems compared.Choice of systems for habitat mapping is discussed. A relative cost analysis of the various survey systems is presented with varying resolution and coverage. The size and shape of the footprint, and thus resolution, is variable dependant on many factors, including ship speed, data processing and sampling. The resolution and formation of imagery is important, but high sampling rates are shown not to be a complete solution as over-sampling can present a false impression of high resolution. However, it is suggested that backscatter imagery should have least 7 bit sensitivity to aid visual and digital inspection. Habitat mapping examples are shown using multibeam backscatter and sidescan sonar, where the processing has been optimised for backscatter imagery. A key question is how much of high resolution bathymetry data is essential for habitat mapping, and whether backscatter imagery can provide more of the information required at a higher resolution than a bathymetric morphology map.     

Textural analyses of multibeam sonar imagery from Stanton Banks, Northern Ireland continental shelf

The mapping of marine habitats mainly relies on acoustic techniques and there is a clear need for reliable classification methods supplementing the interpreter with as much quantitative information as possible. This article presents textural analyses of multibeam sonar imagery from Stanton Banks, on the continental shelf off Northern Ireland. TexAn, originally developed for the textural analysis of sidescan sonar imagery, was tested over an area of ∼72 km² surveyed in 2005 by the European MESH project. The multibeam imagery is affected by several artefacts, including strong uncorrected angular variations in some tracks, and the acquisition of some tracks with very different aspects. The results from unsupervised classification of the imagery, using K-Means, match well the interpretations that can be made using concurrent bathymetric data and visual observations acquired in a later cruise. Textural analyses successfully detect faint trawlmarks and distinguish between the different types of seafloor, including variations within sediments, rocky outcrops and gullied terrains.     

High-frequency volume and boundary acoustic backscatter fluctuations in shallow water

Volume and boundary acoustic backscatter envelope fluctuations are characterized from data collected by the Toroidal Volume Search Sonar (TVSS), a 68 kHz cylindrical array capable of 360 degrees multibeam imaging in the vertical plane perpendicular to its axis. The data are processed to form acoustic backscatter images of the seafloor, sea surface, and horizontal and vertical planes in the volume, which are used to attribute nonhomogeneous spatial distributions of zooplankton, fish, bubbles and bubble clouds, and multiple boundary interactions to the observed backscatter amplitude statistics. Three component Rayleigh mixture probability distribution functions (PDFs) provided the best fit to the empirical distribution functions of seafloor acoustic backscatter. Sea surface and near-surface volume acoustic backscatter PDFs are better described by Rayleigh mixture or log-normal distributions, with the high density portion of the distributions arising from boundary reverberation, and the tails arising from nonhomogeneously distributed scatterers such as bubbles, fish, and zooplankton. PDF fits to the volume and near-surface acoustic backscatter data are poor compared to PDF fits to the boundary backscatter, suggesting that these data may be better described by mixture distributions with component densities from different parametric families. For active sonar target detection, the results demonstrate that threshold detectors which assume Rayleigh distributed envelope fluctuations will experience significantly higher false alarm rates in shallow water environments which are influenced by near-surface microbubbles, aggregations of zooplankton and fish, and boundary reverberation.     

The use of multibeam backscatter angular response for marine sediment characterisation by comparison with shallow electromagnetic conductivity

In this empirically-driven research, multibeam backscatter angular response analysis is presented, together with shallow electromagnetic data and groundtruthing, to examine its suitability as a proxy for sediment characterisation. Backscatter angular curves extracted from Kongsberg EM1002 sonar (95–98 kHz), acquired in the Malin Basin to the northwest of Ireland, have been selected as a case study. Standard angular backscatter features and newly derived curvature features are examined and cross compared. Exhaustive statistical analysis has been performed on the data to elucidate the complex relationship between multibeam backscatter and sediment properties. Subtle subsurface sediment property gradients across the basin identified by the conductivity system are also captured by the newly derived backscatter features. The results reveal that Near-range backscatter is better suited for subsurface sediment characterisation in soft, fine-grained sediments than far-range. Furthermore, the analysis has constrained the optimum interval for such characterisation to in-between 4° and 16° for the parameters of this study. A number of shape features (slope, first derivative, second derivative and Fourier-smoothed least-squares-fitted curvature) have been examined, and their suitability discussed, in terms of sediment characterisation and, in particular, as potential proxies for delineating the boundary between sand- or silt-dominated sediment facies. Nonetheless, curvature features are found to be independent from average angular backscatter response, but outperform both first and second derivatives when correlating with conductivity in the central part of this case-study with fine-grained homogeneous sediments.     

A self-adaptive correction method with eliminating the influence of angular response for multi-beam acoustic intensity data

The multibeam backscatter strength(BS) is greatly affected by acoustic scattering mechanism among the process of multibeam acoustic data acquisition, which is the so-called angular response(AR). While, the AR effect has not been weakened by the acoustic hardware perfectly. There isn't a post-processing method to correct the BS on the complex seabed sediment correctly, especially in the high-incident-angle domain. An improved self-adaptive angular response correction model for the multibeam backscatter strength data was proposed.Firstly, the mean of the BS for continuous pings data at each resample angle was calculated to obtain the AR curve. Secondly, the AR model correction parameters were extracted by means of the second derivative approach on the fitting AR curve generated from the mean BS data curve using gauss fitting method. Finally, the BS segment processing correction model for a single Ping was present, which takes the high-incidence-angle domain into account. The experimental results indicated that the mean deviation among entire transmitting section was decreased by about 30% compared with the traditional model. Especially, in the high-incidencedomain, the mean deviation reduced by nearly 40% compared with traditional model, the standard deviation reduced by approximately 30% as well. The proposed method not only can weaken the influence of scattering mechanism, but also it can avoid these troubles such as BS inhomogeneity, abnormal correction in the central domain which suffered from the AR effect.Therefore, the reliability of multibeam backscatter strength is improved, and what is more, it can present the actual geomorphology of the seabed as much as possible.     

一种消除角度响应影响的多波束声强数据自适应改正方法

针对多波束反向散射强度(Backscatter Strength,BS)数据在采集过程中受到声学散射机理而产生的角度响应(Angular Response,AR)影响,而目前声学硬件方面尚未完美解决且现有后处理改正方法在复杂海底底质环境下适应性较差,尤其在高入射角区域的改正效果甚为不理想的问题,为此,给出了一种基于散射强度的自适应角度响应改正模型。首先获取连续脉冲(Ping)平均散射强度数据的角度响应曲线;其次使用高斯拟合方法对角度响应曲线进行平滑拟合处理,进而对其解算二阶导数提取角度响应模型改正参数;最后给出顾及高入射角区域的单Ping反向散射强度数据的分段处理改正模型。实验结果表明,该方法与传统方法相比,整个发射扇区散射强度平均偏差精度约提高30V,尤其在影响较大的高入射角区域,平均偏差精度约提高40V,并且该区域的标准差精度也提高了近30%。该模型较好地解决了多波束在非正射情况下获取海底精准散射声强的问题,削弱了声波散射机理的影响;同时也解决了散射强度过渡不均衡、中央波束区域改正异常等问题。因此,提高了多波束反向散射强度的可靠性,可以真实地呈现出海底实际的地貌。      

A Bayesian approach to seafloor classification using multi-beam echo-sounder backscatter data

Seafloor classification using acoustic remote sensing techniques is an attractive approach due to its high-coverage capabilities and limited costs. The multi-beam echo-sounder (MBES) system provides high-resolution bathymetry and backscatter information with 100% coverage. In this paper, we present a seafloor classification method that employs the MBES backscatter data. The method uses the averaged backscatter data per beam. It, therefore, is independent on the quality of the MBES calibration. Also, its performance is insensitive to seafloor type variation along the MBES swathe and corrections for the angular dependence of the backscatter are not needed. The method accounts for the ping-to-ping variability of the backscatter intensity. It estimates both the number of seafloor types present in the survey area and the probability density function for the backscatter strength at a certain angle for each of the seafloor types. Application of the method to MBES backscatter data acquired in a well-known test area in the North Sea shows very good agreement with available ground truth. The method’s discriminatory performance for this area is demonstrated to be comparable to that of taking samples of the sediment. All seafloor types known to be present in the area are resolved for. Application of the method to the Stanton bank data set shows clearly separable areas that differ in seafloor composition.     

Quantification of a multibeam sonar for fisheries assessment applications

The acoustic theory is developed for a multibeam fisheries-type sonar employing a circular arc of transducer elements. Specifically, numerical relations for transmit and receive beam patterns are derived and methodologies set forth for the derivation of appropriately scaled acoustic target strength and acoustic volume backscattering strength from an ideally performing multibeam device. Predicted and measured beam characteristics of a realizable multibeam sonar, a Kongsberg Simrad-Mesotech SM 2000, are compared. Practical techniques for the extraction of calibrated acoustic volume backscattering strength from real systems are advanced.     

Multibeam volume acoustic backscatter imagery and reverberation measurements in the northeastern Gulf of Mexico

Multibeam volume acoustic backscatter imagery and reverberation measurements are derived from data collected in 200-m-deep waters in the northeastern Gulf of Mexico, with the Toroidal Volume Search Sonar (TVSS), a 68-kHz cylindrical sonar operated by the U.S. Navy's Coastal System Station. The TVSS's 360-degree vertical imaging plane allows simultaneous identification of multiple volume scattering sources and their discrimination from backscatter at the sea surface or the seafloor. This imaging capability is used to construct a three-dimensional representation of a pelagic fish school near the bottom. Scattering layers imaged in the mixed layer and upper thermocline are attributed to assemblages of epipelagic zooplankton. The fine scale patchiness of these scatterers is assessed with the two-dimensional variance spectra of vertical volume scattering strength images in the upper and middle water column. Mean volume reverberation levels exhibit a vertical directionality which is attributed to the volume scattering layers. Boundary echo sidelobe interference and reverberation is shown to be the major limitation in obtaining bioacoustic data with the TVSS. Because net tow and trawl samples were not collected with the acoustic data, the analysis presented is based upon comparison to previous biologic surveys in the northeastern Gulf of Mexico and reference to the bioacoustic literature.     

一种声图直航段数据自动提取及镶嵌的方法

声图镶嵌是利用成像声纳获得声图进行测深、水下地形地貌测绘的重要应用方向。常见的成像声纳多要求声纳载体做直线运动,而实际应用中载体有大量转弯的情况。为此提出了一种声图直航段数据自动提取方法,该方法首先利用Douglas-Peucker算法对声纳航迹进行抽稀处理,之后利用相邻抽稀点之间的距离判断数据是否为直航段,最后将直航段图像提取出来。获得直航段数据后,利用地理信息对声图进行镶嵌处理。对千岛湖湖底地貌图像进行了处理,获得了较好的镶嵌结果。     

Improving riverbed sediment classification using backscatter and depth residual features of multi-beam echo-sounder systems

Riverbed and seafloor sediment classification using acoustic remote sensing techniques is of high interest due to their high coverage capabilities at limited cost. This contribution presents the results of riverbed sediment classification using multi-beam echo-sounder data based on an empirical method. Two data sets are considered, both taken at the Waal River, namely Sint Andries and Nijmegen. This work is a follow-up to the work carried out by Amiri-Simkooei et al. [J. Acoust. Soc. Am. 126(4), 1724-1738 (2009)]. The empirical method bases the classification on features of the backscatter strength and depth residuals. A principal component analysis is used to identify the most appropriate and informative features. Clustering is then applied to the principal components resulting from this set of features to assign a sediment class to each measurement. The results show that the backscatter strength features discriminate between different classes based on the sediment properties, whereas the depth residual features discriminate classes based on riverbed forms such as the "fixed layer" (stone having riprap structure) and riverbed ripples. Combination of these two sets of features is highly recommended because they provide complementary information on both the composition and the structure of the riverbed.     

Automated acoustic seabed classification of multibeam images of Stanton Banks

Dividing sidescan images into regions that have similar seabeds is often done by expert interpretation. Automated classification systems are becoming more widely used. This paper describes techniques, based on image amplitudes and texture, that lead to useful and practical automated segmentation of multibeam images. Seabed (or riverbed or lakebed) type affects amplitudes and texture, but so do system operating details and survey geometry. Effects of the last two must be compensated to isolate the effects of seabed type. Images from multibeam surveys are accompanied by bathymetric data from which grazing angles of all sonar footprints can be calculated. By compiling tables of amplitude against range and grazing angle, systematic changes in amplitude with these two variables can be removed consistently. Classification, based on a large number of features, is done in image space to avoid artifacts common in mosaics. Unsupervised segmentation requires clustering, in which records are divided into their natural classes. An objective clustering method using simulated annealing assigns points to classes based on their Bayesian distances from cluster centres. Stanton Banks is a rocky area 100 km north of County Donegal, Ireland, that rises about 100 m above the ocean floor at 180 m. Multibeam images and data from an 80-km² survey were classified into regions of acoustic similarity. Assigning labels of physical properties to these regions requires non-acoustic ground truth, which was obtained from a series of 105 photographs. Photographic geological assignments were found to correlate well with the acoustic classes.     

Experimental verification of an interpolation algorithm for improved estimates of animal position

This article presents experimental verification of an interpolation algorithm that was previously proposed in Jaffe [J. Acoust. Soc. Am. 105, 3168-3175 (1999)]. The goal of the algorithm is to improve estimates of both target position and target strength by minimizing a least-squares residual between noise-corrupted target measurement data and the output of a model of the sonar's amplitude response to a target at a set of known locations. Although this positional estimator was shown to be a maximum likelihood estimator, in principle, experimental verification was desired because of interest in understanding its true performance. Here, the accuracy of the algorithm is investigated by analyzing the correspondence between a target's true position and the algorithm's estimate. True target position was measured by precise translation of a small test target (bead) or from the analysis of images of fish from a coregistered optical imaging system. Results with the stationary spherical test bead in a high signal-to-noise environment indicate that a large increase in resolution is possible, while results with commercial aquarium fish indicate a smaller increase is obtainable. However, in both experiments the algorithm provides improved estimates of target position over those obtained by simply accepting the angular positions of the sonar beam with maximum output as target position. In addition, increased accuracy in target strength estimation is possible by considering the effects of the sonar beam patterns relative to the interpolated position. A benefit of the algorithm is that it can be applied "ex post facto" to existing data sets from commercial multibeam sonar systems when only the beam intensities have been stored after suitable calibration.     

Model-based seafloor characterization employing multi-beam angular backscatter data--a comparative study with dual-frequency single beam

Sediment geoacoustic inversion results are estimated employing a multi-beam (MB) echo-sounding system operable at 95 kHz. To characterize the western continental shelf of India (off Goa) seafloor, MB backscatter signals were acquired along with grab sediment samples. The substrate type and roughness of the site were estimated using the composite roughness scattering model with the measured backscatter values. The seafloor parameters, namely mean grain size (M(φ)); roughness spectrum strength (w(2)) and exponent (γ(2)); and sediment volume parameter (σ(2)), for coarse and fine grain sediments are estimated by employing the MB system. These parameters have also been estimated at two other frequencies (33 and 210 kHz) and are compared to the ground truth data to provide sufficient support in validating the model results and increasing the understanding of the shelf seafloor processes. Distinct interclass separations between the sediment provinces are evident from the estimated mean grain size M(φ) and water-sediment interface roughness w(2). The seafloor parameters for coarse and fine grain sediments derived from the 95 kHz MB data are consistent with the sediment sample data as well as with the inversion results obtained using backscatter data at 33 and 210 kHz from the same locations.     

Inversion of penetrable obstacles from far-field data on narrow angular apertures

Reported in this paper are reconstructions of shape and material parameters of two-dimensional, homogeneous, acoustic, penetrable obstacles of arbitrary cross-sections which are immersed in an infinite, homogeneous ambience. Reconstructions are based on the far-field scattering patterns of multiple incident plane acoustic waves. For remotely acquired data, practical necessities require not only that the entire data collection region be less than 2 pi, but also that each received "sees" the object over as narrow an angular aperture as possible. The inversions presented here were obtained under such conditions. Two types of data were used for each incidence namely, "near-monostatic" fields of narrow angular apertures (the narrowest aperture reported is 2 degrees), and a "duostatic" geometry consisting of backscatter plus one other receiver angle. These data sets were acquired for a series of incident angles. The theoretical formalism for inversion is algebraic in nature, requires no integral equation, and possesses a number of advantages for the implementation of a Gauss-Newton type of inversion that was used in this study. Moreover, the algorithm is shown to be inheritently parallelizable.     

南海北部负跃层环境下海底参数声学反演

利用2015年南海北部声学实验的实验数据,对海底声学参数进行反演。在综合分析声速起伏及海底模型对海底反演影响的基础上,通过选择等效的海水声速剖面和海底模型,改进了多参量联合地声反演方法,使得其不仅能解决反演中的多值性问题,还能适用于负跃层起伏环境下的声学反演。即当用匹配场反演海底声速和用传播损失反演海底衰减时,如果温跃层内有内波等随机起伏,可使用传播路径上平均声速剖面和单层等效海底进行声场计算。反演得到的海底声速和密度结果与海底采样测量符合较好,拟合给出了海底衰减系数随频率的非线性经验关系式。反演结果可为南海北部声传播规律研究与应用提供海底参数。      

海底表层沉积物声速的环境因素影响特性

应用声速通用模型开展理论分析,结合温度-压力可控实验测量分析,研究地声反演、海底原位声学测量和采样样品声学测量3种主要的海底表层沉积物声学特性测量方法中,存在的环境因素影响机制和特性.理论计算与实验测量研究具有一致性,揭示温度和静水压力对海底表层沉积物的压缩波声速的影响机制主要是通过影响孔隙海水的密度、黏度、体积弹性模...     

Application of acoustic reflection tomography to sonar imaging

Computer-aided tomography is a technique for providing a two-dimensional cross-sectional view of a three-dimensional object through the digital processing of many one-dimensional views (or projections) taken at different look directions. In acoustic reflection tomography, insonifying the object and then recording the backscattered signal provides the projection information for a given look direction (or aspect angle). Processing the projection information for all possible aspect angles enables an image to be reconstructed that represents the two-dimensional spatial distribution of the object's acoustic reflectivity function when projected on the imaging plane. The shape of an idealized object, which is an elliptical cylinder, is reconstructed by applying standard backprojection, Radon transform inversion (using both convolution and filtered backprojections), and direct Fourier inversion to simulated projection data. The relative merits of the various reconstruction algorithms are assessed and the resulting shape estimates compared. For bandpass sonar data, however, the wave number components of the acoustic reflectivity function that are outside the passband are absent. This leads to the consideration of image reconstruction for bandpass data. Tomographic image reconstruction is applied to real data collected with an ultra-wideband sonar transducer to form high-resolution acoustic images of various underwater objects when the sonar and object are widely separated.