Optical turbulence on underwater image degradation in natural environments
TL;DR: This study presents the first attempts in quantifying the level of image degradation due to optical turbulence in natural waters in terms of modulation transfer functions using measured turbulence dissipation rates.
Abstract: It is a well-known fact that the major degradation source on electro-optical imaging underwater is from scattering by particles of various origins and sizes. Recent research indicates that, under certain conditions, the apparent degradation could also be caused by the variations of index of refraction associated with temperature and salinity microstructures in the ocean and lakes. The combined impact has been modeled previously through the simple underwater imaging model. The current study presents the first attempts in quantifying the level of image degradation due to optical turbulence in natural waters in terms of modulation transfer functions using measured turbulence dissipation rates. Image data collected from natural environments during the Skaneateles Optical Turbulence Exercise are presented. Accurate assessments of the turbulence conditions are critical to the model validation and were measured by two instruments to ensure consistency and accuracy. Optical properties of the water column in the field were also measured in coordination with temperature, conductivity, and depth. The results show that optical turbulence degrades the image quality as predicted and on a level comparable to that caused by the particle scattering just above the thermocline. Other contributing elements involving model closure, including temporal and spatial measurement scale differences among sensors and mitigation efforts, are discussed.
Citations
More filters
TL;DR: This paper constructs an Underwater Image Enhancement Benchmark (UIEB) including 950 real-world underwater images, 890 of which have the corresponding reference images and proposes an underwater image enhancement network (called Water-Net) trained on this benchmark as a baseline, which indicates the generalization of the proposed UIEB for training Convolutional Neural Networks (CNNs).
Abstract: Underwater image enhancement has been attracting much attention due to its significance in marine engineering and aquatic robotics. Numerous underwater image enhancement algorithms have been proposed in the last few years. However, these algorithms are mainly evaluated using either synthetic datasets or few selected real-world images. It is thus unclear how these algorithms would perform on images acquired in the wild and how we could gauge the progress in the field. To bridge this gap, we present the first comprehensive perceptual study and analysis of underwater image enhancement using large-scale real-world images. In this paper, we construct an Underwater Image Enhancement Benchmark (UIEB) including 950 real-world underwater images, 890 of which have the corresponding reference images. We treat the rest 60 underwater images which cannot obtain satisfactory reference images as challenging data. Using this dataset, we conduct a comprehensive study of the state-of-the-art underwater image enhancement algorithms qualitatively and quantitatively. In addition, we propose an underwater image enhancement network (called Water-Net) trained on this benchmark as a baseline, which indicates the generalization of the proposed UIEB for training Convolutional Neural Networks (CNNs). The benchmark evaluations and the proposed Water-Net demonstrate the performance and limitations of state-of-the-art algorithms, which shed light on future research in underwater image enhancement. The dataset and code are available at https://li-chongyi.github.io/proj_benchmark.html .
697 citations
TL;DR: In this article, the current state of underwater optical imaging in the context of physics, technology, biology, and history is discussed and the future prospects for continuing advancements are also discussed.
Abstract: This paper discusses the current state of underwater optical imaging in the context of physics, technology, biology, and history. The paper encompasses not only the history of human's ability to see underwater, but also the adaptations that various organisms living in oceans or lakes have developed. The continued development of underwater imaging systems at military, commercial, and consumer levels portends well for both increased visibility and accessibility by these various segments. However, the fundamental limits imposed by the environment, as currently understood, set the ultimate constraints. Physics, biology, computer modeling, processing, and the development of technology that ranges from simple cameras and lights to more advanced gated and modulated illumination are described. The future prospects for continuing advancements are also discussed.
177 citations
TL;DR: A fiber-optic sensor based on a silicon Fabry-Pérot cavity, fabricated by attaching a silicon pillar on the tip of a single-mode fiber, for high-resolution and high-speed temperature measurement, suggesting a maximum frequency of ~2 kHz can be reached to address the needs for highly dynamic environmental variations such as those found in the ocean.
Abstract: We report a fiber-optic sensor based on a silicon Fabry-Perot cavity, fabricated by attaching a silicon pillar on the tip of a single-mode fiber, for high-resolution and high-speed temperature measurement. The large thermo-optic coefficient and thermal expansion coefficient of the silicon material give rise to an experimental sensitivity of 84.6 pm/°C. The excellent transparency and large refractive index of silicon over the infrared wavelength range result in a visibility of 33 dB for the reflection spectrum. A novel average wavelength tracking method has been proposed and demonstrated for sensor demodulation with improved signal-to-noise ratio, which leads to a temperature resolution of 6 × 10−4 °C. Due to the high thermal diffusivity of silicon, a response time as short as 0.51 ms for a sensor with an 80-µm-diameter and 200-µm-long silicon pillar has been experimentally achieved, suggesting a maximum frequency of ~2 kHz can be reached, to address the needs for highly dynamic environmental variations such as those found in the ocean.
162 citations
Posted Content•
TL;DR: The Underwater Image Enhancement Benchmark (UIEB) as mentioned in this paper is a large-scale real-world underwater image enhancement dataset, which consists of 950 realworld underwater images, 890 of which have the corresponding reference images.
Abstract: Underwater image enhancement has been attracting much attention due to its significance in marine engineering and aquatic robotics. Numerous underwater image enhancement algorithms have been proposed in the last few years. However, these algorithms are mainly evaluated using either synthetic datasets or few selected real-world images. It is thus unclear how these algorithms would perform on images acquired in the wild and how we could gauge the progress in the field. To bridge this gap, we present the first comprehensive perceptual study and analysis of underwater image enhancement using large-scale real-world images. In this paper, we construct an Underwater Image Enhancement Benchmark (UIEB) including 950 real-world underwater images, 890 of which have the corresponding reference images. We treat the rest 60 underwater images which cannot obtain satisfactory reference images as challenging data. Using this dataset, we conduct a comprehensive study of the state-of-the-art underwater image enhancement algorithms qualitatively and quantitatively. In addition, we propose an underwater image enhancement network (called Water-Net) trained on this benchmark as a baseline, which indicates the generalization of the proposed UIEB for training Convolutional Neural Networks (CNNs). The benchmark evaluations and the proposed Water-Net demonstrate the performance and limitations of state-of-the-art algorithms, which shed light on future research in underwater image enhancement. The dataset and code are available at this https URL.
123 citations
TL;DR: A Monte Carlo numerical simulation for computing the received power for an underwater optical communication system is discussed and validated and power loss between receiver and transmitter is simulated.
Abstract: A Monte Carlo numerical simulation for computing the received power for an underwater optical communication system is discussed and validated. Power loss between receiver and transmitter is simulated for a variety of receiver aperture sizes and fields of view. Additionally, pointing-and-tracking losses are simulated.
97 citations
Cites background from "Optical turbulence on underwater im..."
...[4] shows that small-scale turbulence does still exist....
[...]
References
More filters
TL;DR: In this article, a theoretical investigation of the spectrum of a turbulent fluid at large wave-numbers is presented, taking into account the two effects of convection with the fluid and molecular diffusion with diffusivity k. Hypotheses of the kind made by Kolmogoroff for the small-scale variations of velocity in a turbulent motion at high Reynolds number are assumed to apply also to small-size variations of θ.
Abstract: When some external agency imposes on a fluid large-scale variations of some dynamically passive, conserved, scalar quantity θ like temperature or concentration of solute, turbulent motion of the fluid generates small-scale variations of θ. This paper describes a theoretical investigation of the form of the spectrum of θ at large wave-numbers, taking into account the two effects of convection with the fluid and molecular diffusion with diffusivity k. Hypotheses of the kind made by Kolmogoroff for the small-scale variations of velocity in a turbulent motion at high Reynolds number are assumed to apply also to small-scale variations of θ.Previous contributions to the problem are reviewed. These have established that the spectrum of θ varies as , the result being given by (4.8). The same result is obtained, using essentially the same approximation about the velocity field, from a different kind of analysis in terms of velocity and θ correlations. Finally, the relation between this work and Townsend's model of the small-scale variations of vorticity in a turbulent fluid is discussed.
1,665 citations
TL;DR: In this article, the average resolution of very-long and very-short-exposure images is studied in terms of the phase and log-amplitude structure functions, whose sum is called the wave-structure function.
Abstract: A theoretical foundation is developed for relating the statistics of wave distortion to optical resolution. The average resolution of very-long- and very-short-exposure images is studied in terms of the phase- and log-amplitude-structure functions, whose sum we call the “wave-structure function.” Those results which are comparable are in agreement with the findings of Hufnagel and Stanley who studied the average modulation transfer function of long-exposure images. It is found that the average short-exposure resolution can be significantly better than the average long-exposure resolution, but only if the wave distortion does not include substantial intensity variation.
1,525 citations
TL;DR: In this article, a range-gated laser system was used for underwater camera image enhancement in a remotely operated vehicle (HYSUB 5000) with a 10 cm-diam zoom lens.
Abstract: A careful analysis of a scattering and absorption database of the waters off the coasts of Canada shows that a laser-assisted camera system will have a significantly improved viewing performance over conventional systems. The laser underwater camera image enhancer system is a range-gated laser system that can be mounted on a remotely operated vehicle. The system uses a 2-kHz diode-pumped frequency-doubled Nd:YAG laser as an illumination source. The light is collected by a 10-cm-diam zoom lens. The detector is a gated image intensifier with a 7-ns gate and a gain that is continuously variable from 500 to 1,000,000. The system has been tested in a water tank facility at Defence Research Establishment Valcartier and has been mounted on the HYSUB 5000 remotely operated vehicle for sea trials. In the strongly scattering waters typical of harbor approaches, this system has a range of from three to five times that of a conventional camera with floodlights.
171 citations
"Optical turbulence on underwater im..." refers background in this paper
...Most research has focused on reducing the impact of particle scattering by means of discriminating scattering photons involving polarization, range gating, modulation, and by means of restoration via deconvolution [1–6]....
[...]
TL;DR: The technique includes an exact correction for the frequency passband of the finite-element differentiation required to obtain the line spread function from the edge spread function and is found to be consistent with an MTF obtained using a bar pattern test phantom.
Abstract: In this paper we describe a technique for determining the modulation transfer function (MTF) of an imaging system from an experimentally obtained edge profile. The technique includes an exact correction for the frequency passband of the finite-element differentiation required to obtain the line spread function from the edge spread function. This correction has been ignored by investigators in the past and is required whenever finite-element differentiation is used rather than analytic differentiation of a model fitted to the edge response data. The magnitude of the MTF correction is approximately 11% at f = fc/2 and approximately 57% at f = fc, where fc = fs/2 is the maximum frequency reproducible without aliasing with a sampling rate of fs. The correction is performed in the spatial frequency domain by multiplying the uncorrected MTF by 1/sinc (pi f/2fc). A computer simulation is presented to demonstrate the effect and the correction procedure. An experimental MTF of an x-ray image intensifier system obtained using this technique is found to be consistent with an MTF obtained using a bar pattern test phantom.
158 citations
TL;DR: A model is presented to include the effects of both particle and turbulence on underwater optical imaging through optical transfer functions to help quantify the limiting factors under different circumstances.
Abstract: It is commonly known that underwater imaging is hindered by both absorption and scattering by particles of various origins. However, evidence also indicates that the turbulence in natural underwater environments can cause severe image-quality degradation. A model is presented to include the effects of both particle and turbulence on underwater optical imaging through optical transfer functions to help quantify the limiting factors under different circumstances. The model utilizes Kolmogorov-type index of refraction power spectra found in the ocean, along with field examples, to demonstrate that optical turbulence can limit imaging resolution by affecting high spatial frequencies. The effects of the path radiance are also discussed.
128 citations
"Optical turbulence on underwater im..." refers background or methods in this paper
...This is necessary, however, in order to contain all of the variations caused by the optical turbulence [15]....
[...]
...The SUIMmodel [15,22] is used to incorporate the impacts of optical turbulence at 8....
[...]
...The simple underwater imaging model (SUIM) [15] was developed to address this issue, and it has been shown in theory that, on average, the relative contribution of different components [3,16] in underwater imaging applications can be expressed in terms of the optical transfer function (OTF) as...
[...]
...R0 is the characteristic seeing parameter [15]....
[...]