First Results on Wake Detection in SAR Images by Deep Learning

Spaceborne synthetic aperture radar (SAR) represents a powerful source of data for enhancing maritime domain awareness (MDA). Wakes generated by traveling vessels hold a crucial role in MDA since they can be exploited both for ship route and velocity estimation and as a marker of ship presence. Even if deep learning (DL) has led to an impressive performance boost on a variety of computer vision tasks, its usage for automatic target recognition (ATR) in SAR images to support MDA is still limited to the detection of ships rather than ship wakes. A dataset is presented in this paper and several state-of-the-art object detectors based on convolutional neural networks (CNNs) are tested with different backbones. The dataset, including more than 250 wake chips, is realized by visually inspecting Sentinel-1 images over highly trafficked maritime sites. Extensive experiments are shown to characterize CNNs for the wake detection task. For the first time, a deep-learning approach is implemented to specifically detect ship wakes without any a-priori knowledge or cuing about the location of the vessel that generated the wake. No annotated dataset was available to train deep-learning detectors on this task, which is instead presented in this paper. Moreover, the benchmarks achieved for different detectors point out promising features and weak points of the relevant approaches. Thus, the work also aims at stimulating more research in this promising, but still under-investigated, field.

Electromagnetic Scattering Model for Far Wakes of Ship with Wind Waves on Sea Surface

A comprehensive electromagnetic scattering model for ship wakes on the sea surface is proposed to study the synthetic aperture radar (SAR) imagery for ship wakes. Our model considers a coupling of various wave systems, including Kelvin wake, turbulent wake, and the ocean ambient waves induced by the local wind. The fluid–structure coupling between the ship and the water surface is considered using the Reynolds–averaged Navier–Stokes (RANS) equation, and the wave–current effect between the ship wake and wind waves is considered using the wave modulation model. The scattering model can better describe the interaction of the ship wakes on sea surface and illustrates well the features of the ship wakes with local wind waves in SAR images.

Prediction of cavity volumes on rotating blades and scale effects

Object and purpose of research. Pressure pulsations induced by cavitating blades substantially contribute to flowinduced loads and amplify structural vibration. These pulsations depend on oscillation of the volume of cavities over blades. Prediction of them usually involves model tests and there are three kinds of scale effects influencing the cavity volumes. The first one is associated with the non-uniform inflows. The second one is associated with the combined influence of the blade boundary layer and surface tension on the cavity surface. The third one is associated with the cavity buoyancy. Materials and methods. Because of complexity of blade flows, a qualitative analysis of similar unsteady non-uniform flows around 3D hydrofoils is useful. This paper presents such an analysis for a hydrofoil with the sections copied from a marine propeller blade. The inflows correspond to the wakes of a ship and of her model. Computations carried out using an analysis of viscous-inviscid interaction. Main results. The qualitative explanation of observed trends and scale effects is obtained due to this analysis. In particular, the role of pressure side cavitation in full scale conditions is pointed out. Conclusion. The difference of model and ship wakes results in the substantial difference in blade section angles of attack at the same blade loading. Therefore, in model tests the suction side cavitation is more extensive, whereas the pressure side cavitation may not appear, though it exists on full-scale ship propeller blade. This substantial scale effect has been usually out of previous considerations.

Stochastic Uncertainty in a Dam-Break Experiment with Varying Gate Speeds

Uncertainties inherent in gate-opening speeds are rarely studied in dam-break flow experiments due to the laborious experimental procedures required. For the stochastic analysis of these mechanisms, this study involved 290 flow tests performed in a dam-break flume via varying gate speeds between 0.20 and 2.50 m/s; four pressure sensors embedded in the flume bed recorded high-frequency bottom pressures. The obtained data were processed to determine the statistical relationships between gate speed and maximum pressure. The correlations between them were found to be particularly significant at the sensors nearest to the gate (Ch1) and farthest from the gate (Ch4), with a Pearson’s coefficient r of 0.671 and −0.524, respectively. The interquartile range (IQR) suggests that the statistical variability of maximum pressure is the largest at Ch1 and smallest at Ch4. When the gate is opened faster, a higher pressure with greater uncertainty occurs near the gate. However, both the pressure magnitude and the uncertainty decrease as the dam-break flow propagates downstream. The maximum pressure appears within long-period surge-pressure phases; however, instances considered as statistical outliers appear within short and impulsive pressure phases. A few unique phenomena, which could cause significant bottom pressure variability, were also identified through visual analyses using high-speed camera images. For example, an explosive water jet increases the vertical acceleration immediately after the gate is lifted, thereby retarding dam-break flow propagation. Owing to the existence of sidewalls, two edge waves were generated, which behaved similarly to ship wakes, causing a strong horizontal mixture of the water flow.

Non-Linear Modeling of Detectability of Ship Wake Components in Dependency to Influencing Parameters Using Spaceborne X-Band SAR

The detection of the wakes of moving ships in Synthetic Aperture Radar (SAR) imagery requires the presence of wake signatures, which are sufficiently distinctive from the ocean background. Various wake components exist, which constitute the SAR signatures of ship wakes. For successful wake detection, the contrast between the detectable wake components and the background is crucial. The detectability of those wake components is affected by a number of parameters, which represent the image acquisition settings, environmental conditions or ship properties including voyage information. In this study the dependency of the detectability of individual wake components to these parameters is characterized. For each wake component a detectability model is built, which takes the influence of incidence angle, polarization, wind speed, wind direction, sea state (significant wave height, wavelength, wave direction), vessel’s velocity, vessel’s course over ground and vessel’s length into account. The presented detectability models are based on regression or classification using Support Vector Machines and a dataset of manually labelled TerraSAR‑X wake samples. The considered wake components are: near‑hull turbulences, turbulent wakes, Kelvin wake arms, Kelvin wake’s transverse waves, Kelvin wake’s divergent waves, V‑narrow wakes and ship‑generated internal waves. The statements derived about wake component detectability are mainly in good agreement with statements from previous research, but also some new assumptions are provided. The most expressive influencing parameter is the movement velocity of the vessels, as all wake components are more detectable the faster vessels move.

A Retinex-based Enhancement method for Ocean Remote Sensing Image

In order to effectively monitor important sea areas, one of the key issues is the detection of small dynamic targets such as ships. Besides monitoring the hull itself, the monitoring of small targets can also be achieved by detecting ship wakes, which means the sea image with clear texture feature is required. This paper firstly reviews the multiscale retinex (MSR) method, commonly used to enhance the image contrast. Then, it proposes a novel contrast enhancement algorithm base on the subband-decomposed multiscale retinex (SDMSR) method. Experimental results show that our proposed method can make a remarkable enhancement effect for ocean remote sensing images with clouds and whitecap, etc.

Wave Patterns of Gravity–Capillary Waves from Moving Localized Sources

We study wave patterns of gravity–capillary waves from moving localized sources within the classic setup of the problem of ship wakes. The focus is on the co-existence of two wave systems with opposite signatures of group velocity relative to the localized source. It leads to the problem of choice of signs for phase functions of the gravity (“slow”) and capillary (“fast”) branches of the dispersion relation: the question generally ignored when constructing phase patterns of the solutions. We detail characteristic angles of the wake patterns: (i) angle of demarcation of gravity and capillary waves—“the phase Mach” cone, (ii) angle of the minimal group velocity of gravity–capillary waves—“the group Mach” cone, (iii, iv) angles of cusps of isophases that appear after a threshold current speed. The outer cusp cone is naturally associated with the classic cone of Kelvin for pure gravity waves. The inner one results from the effect of capillarity and tends to the “group Mach” pattern at high speeds of current. Amplitudes of the wave patterns are estimated within the recently proposed approach of reference functions for the problem of propagation of packets of linear dispersive waves. The effect of shape is discussed for elliptic reference sources.