No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Sea surface simulation in the infrared modeling and validation - art. no. 62390J
Abstract
A physics based D simulation of sea surfaces is presented. The simulation is suitable for the pre-calculation of detector images for an IR camera. Synthetic views of a maritime scenario are calculated in the MWIR and LWIR spectral bands and the images are compared with data collected in a field trial. In our computer simulation the basic sea surface geometry is modeled by a composition of smooth wind driven gravity waves. Sea surface animation is introduced by time dependent control of the basic statistics. Choppy waves are included into the model to improve the realism of the rough sea. To predict the view of a thermal camera the sea surface radiance must be calculated. This is done with respect to the emitted sea surface radiance and the reflected sky radiance, using either MODTRAN or a semi-empirical model. Slope-shadowing of the sea surface waves is considered, which strongly influences the IR appearance of the sea surface near the horizon. MWIR and LWIR simulations are shown of sun glint as well as of whitecaps which depend upon wind velocity. For validation purposes appropriate data sets (images and meteorological data) were selected from field measurements. A simple maritime scenario including a floating foreground object has been prepared and views of two different thermal imagers, similar to those used in the field trials, have been simulated. The validation is done by visual inspection of measured and simulated images and in addition by numerical comparison based on image statistics. The results of the comparison are presented. For an accurate reflectance calculation it is necessary to consider the maritime sky. The model is improved by inclusion of a static two-dimensional cloud layer. The cloud distribution adjusted to measured data with respect, e.g. to power spectral density and temperature distribution.
... The total radiance L t at the infrared sensor can be approximated by [13] ...
... where Ω S is the solid angle of the sky, as seen from sea level; θ i is the incident zenith angle of the sky radiation; φ i is the incident azimuth of the sky radiation; and L sky θ i , φ i is the incident radiance of the sky. The calculation of the reflected sky radiance is as follows [13]: ...
... A fraction of the incoming solar radiation is scattered by the atmosphere before reaching the sea surface, which results in an angular distribution of the downwelling solar radiative power. Solar irradiance is transmitted through a cloudless atmosphere (transmissions) and is reflected at the sea surface [13]. The infrared radiance of the sea surface in the direction of reflection increases, and the phenomenon of "wave sparkling" appears on the sea surface. ...
In order to establish a more realistic radiation model of the sea surface, the effects of solar radiation, sky radiation, and atmospheric thermal radiation on sea surface radiation are taken into consideration, on the basis of which the infrared radiative transfer equation of the sea surface is deduced in this paper. A method for calculating the bidirectional reflection characteristics of the sea surface based on measured data is proposed according to the projection imaging of beam propagation. Based on the measurements of sea surface temperature, incident sky radiation, incident solar radiation, and radiance of sea crests at different times, the radiative transfer equation is used to retrieve the bidirectional reflectance of a midwave infrared sea surface. Meanwhile, the results of the method mentioned above are compared with the calculated results of Cox–Munk, Mermelstein, Wu, and Beckmann bidirectional reflection characteristics models. Research shows that the bidirectional reflectance at the wave crest of a sea surface increases gradually, when the solar incident zenith angle changes from 56.39° to 76.02° as well as the direction of observation remaining constant ( θ r = 80.0 ° ; ϕ r = 73.0 ° ). The reflection ability at the wave crest of the sea surface is strongest when the incident direction of the sun is close to the observation direction, which is in accordance with the law of reflection. The Cox–Munk model and Wu model are closer to our values when the solar incidence zenith angle is small ( θ i ≤ 65.93 ° ). On the other hand, the calculated values of the Mermelstein and Wu models are closer to the values in this paper when the solar incidence zenith angle is large ( θ i ≤ 65.93 ° ). In general, the error of the Beckmann model is a little greater than that of the other three models.
... When the steepness exceeds the limit value, the vertical acceleration on the wave crest crosses the limit acceleration, thus the vertical acceleration is also taken as a criterion [42], [47]. Schwenger et al. [49] rendered the whitecap, taking the vertical acceleration as a criterion. Unfortunately, the detail of the whitecap rendering method was not provided. ...
... Figure 7 shows the rendered results of the open source code offered by [45]. We note that the whitecap in Figure 7 is distinctly different from the actual whitecap in Figure 6. Figure 8 shows the rendered results of [43], [49]. Obviously, the rendered whitecap in Figure 8(a) enormously differs from the actual whitecap. ...
... Obviously, the rendered whitecap in Figure 8(a) enormously differs from the actual whitecap. The whitecap in Figure 8(b) is similar to our method, but [49] did not provide details of the whitecap rendering method. Figure 9 shows curves of the whitecap coverage. ...
The whitecap is an important oceanographic phenomenon. However, existing whitecap rendering methods do not successfully generate realistic whitecaps. To solve this problem, this paper presents a real-time whitecap rendering method applied to the visual system of a maritime simulator. The method takes the vertical acceleration on the wave crest as the criterion of whitecap generation. The Fourier coefficient of the vertical acceleration is provided, and a continuous mathematical model computing the whitecap coverage is built. The vertical acceleration is the variable of the model. The life time of the whitecap’s existence can be controlled by the parameter of the model, and the parameter is computed with the genetic algorithm. The average of the computed whitecap coverage is equal to the whitecap coverage computed by the stochastic method and is close to the whitecap coverage computed by the empirical formula. The whitecap coverage is used as the blending factor to blend the pixel color of the whitecap texture and that of the sea surface. The presented method has sound theoretical support, with small computational complexity. The rendered whitecap is closer to the description of the Beaufort wind force scale than before.
... The slope-shadowing factor S h considers the shadowing of the radiances caused by waves, essentially occurring at flat viewing angles. Details can be found in our previous paper [16], and its derivation is given in Ref. [17]; however, a short summary is given. We use a simple approximation of the shadowing derived by Saunders [17] that is a function of only the inclination of the line of sight, the viewing angle ϕ, and the rms slope value σ of the water facets, ignoring the slope dependencies on the upwind and crosswind direction: ...
... Since our model neglects higher-order effects, such as multiple reflections with second-order slopeshadowing and masking, and does not consider the sea surface BRDF, the radiance from regions of the sea surface near the horizon [Eq. (16)] is under predicted. To correct for these effects, we introduced a "shadow-weighting" factor η for the shadowing function S h . ...
... Ref. [16] shows that choosing η 0.1 leads to a good accordance of measured and simulated radiance profiles as a function of elevation angle ϕ, for both the LWIR and the MWIR. This factor is also used for the radiance calculation in the SWIR spectral band. ...
The knowledge of the spatial energy (or power) distribution of light beams reflected at the dynamic sea surface is of great practical interest in maritime environments. For the estimation of the light energy reflected into a specific spatial direction a lot of parameters need to be taken into account. Both whitecap coverage and its optical properties have a large impact upon the calculated value. In published literature, for applications considering vertical light propagation paths, such as bathymetric lidar, the reflectance of sea surface and whitecaps are approximated by constant values. For near-horizontal light propagation paths the optical properties of the sea surface and the whitecaps must be considered in greater detail. The calculated light energy reflected into a specific direction varies statistically and depends largely on the dynamics of the wavy sea surface and the dynamics of whitecaps. A 3D simulation of the dynamic sea surface populated with whitecaps is presented. The simulation considers the evolution of whitecaps depending on wind speed and fetch. The radiance calculation of the maritime scene (open sea/clear sky) populated with whitecaps is done in the short wavelength infrared spectral band. Wave hiding and shadowing, especially occurring at low viewing angles, are considered. The specular reflection of a light beam at the sea surface in the absence of whitecaps is modeled by an analytical statistical bidirectional reflectance distribution function (BRDF) of the sea surface. For whitecaps, a specific BRDF is used by taking into account their shadowing function. To ensure the credibility of the simulation, the whitecap coverage is determined from simulated image sequences for different wind speeds and compared to whitecap coverage functions from literature. The impact of whitecaps on the radiation balance for bistatic configuration of light source and receiver is calculated for a different incident (zenith/azimuth angles) of the light beam and is presented for two different wind speeds.
... In the example scenario shown in Fig. 3, the spectral analysis capability of the DAMA software is demonstrated for the Fig. 3 are consistent with the measurement provided in [25]. The components of sea, sky, and Sun radiances shown in Fig. 3(a)-(c), respectively, are multiplied by the transmittance in Fig. 3(e). ...
... The simulation results of the developed software DAMA are compared with those from the measurements provided in [25]. The simulation parameters are chosen in a way that would reflect the measurement scenarios as much as possible. ...
... The comparison of total radiance for the three sea surface models for varying observation zenith look angles with the measured total radiance is shown in Fig. 8. In [25], total radiance is measured for 3.4-5.3-and 8.0-8.75-μm ...
The performance of infrared (IR) surveillance systems is proportional to the ability of distinguishing the radiance of target from the background. In determining sea radiance as the background, experimental studies are incomplete and very expensive. The few simulation software packages that are available are limited to certain parameters and do not include all aspects of sea radiance. In this paper, simulation and analysis software, namely, InfrareD SeA Modeling and Analysis (DAMA), is developed to calculate the total radiance and its components seen by the observer. The DAMA software allows the user to define all possible parameters related to atmospheric conditions, sea surface conditions, and date and time of observation. The developed software includes all of the three major sea surface models in the IR band provided in the literature, namely, Cox and Munk, Mermelstein, and Shaw and Churnside models. In this sense, the DAMA software is unique in combining all possible atmospheric and surface parameters to provide components of total radiance. The DAMA software can be operated by a user-friendly graphical user interface to facilitate simulations and to analyze the outputs. For this paper, the software has been run for approximately 10 000 simulations to understand the behavior of maritime background in IR. The software is validated by the SEARAD software and measurement results. By this way, for the first time in the open literature, DAMA allows one to observe the behavior of total radiance and its components with respect to the variation of all possible input parameters.
... The performance assessment of a specific device and an observer for a dedicated military application in maritime environment is of great importance [1]. Unlike the active RF detection systems, early design of passive sensors was possible in the infrared band and the first measurements for the analysis of infrared background radiance dates back to 1956 [2]. ...
... ( 100 (1) where x m and x s are the measured and the simulated data for the average respectively, and e is the error percentage value calculated using N measured data obtained at the same scenario. The absolute sign merges the positive and negative results to a same value, and to calculate the error bias correctly, error form without the magnitude operation should be used ...
Estimation of the land cover type and proportion of vegetation using NASA’s EOS/MODIS data is verified as a case study for Turkey where 7 locations were selected. The correlation between the NDVI/EVI data is analyzed. EVI is shown to be slightly more accurate than NDVI. Land cover types are determined for these 7 locations, and similar to other locations, İstanbul and Samsun are found to perfectly match the ‘forest type’. İzmir is found to be different from the currently available land cover types which are assumed to be due to its mixed canopy structure. Annual and monthly average measurements in various bands are analyzed and expected correlation with the number of days of precipitation
is observed.
... Our model is based on the model presented in [10], [11]. There exist different classes of waves for generating a realistic sea model. ...
... We first describe the method for computing the radiance leaving one sea facet n. The radiance leaving n for wavelength λ is computed using the following equation [10], [11], [13]. ...
For locating maritime vessels longer than 45 meters, such vessels are required to set up an Automatic Identification System (AIS) used by vessel traffic services (VTS). However, when a boat is shutting down its AIS, there are no means to detect it in open sea. In this paper, we use Electro-Optical (EO) imagers for non-cooperative vessel detection when the AIS is not operational. As compared to radar sensors, EO sensors have less complex system (lower cost and lower payload) and better computational processing load. EO sensors are mounted on LEO micro-satellites. We propose a simulator providing an estimate of sensor Receiver Operating Characteristics (ROC) curves in real-time and without computing the entire image received at the sensor. This simulator can help sensor manufacturers in optimizing the design of EO sensors.
... Our model is based on the model presented in [12, 13]. In realistic sea surface models, we consider three classes of waves: (1) capillarity waves with small wavelength (λ < 5 cm) influenced by viscosity and surface tension, (2) gravity waves that are wind-driven waves with wavelength λ > 5 cm and smaller than a few meters, (3) swells being waves with great wavelength, that is, λ is greater than a few meters (these waves originate due to the presence of wind. ...
... We first describe the method for computing the radiance leaving one sea facet n. The radiance R n (λ)[W/m 2 · sr μm] leaving n for wavelength λ is computed using the following equation [12, 13, 15]: ...
For locating maritime vessels longer than 45 meters, such vessels are required to set up an Automatic Identification System (AIS) used by vessel traffic services. However, when a boat is shutting down its AIS, there are no means to detect it in open sea. In this paper, we use Electro-Optical (EO) imagers for noncooperative vessel detection when the AIS is not operational. As compared to radar sensors, EO sensors have lower cost, lower payload, and better computational processing load. EO sensors are mounted on LEO microsatellites. We propose a real-time statistical methodology to estimate sensor Receiver Operating Characteristic (ROC) curves. It does not require the computation of the entire image received at the sensor. We then illustrate the use of this methodology to design a simple simulator that can help sensor manufacturers in optimizing the design of EO sensors for maritime applications.
... Böylece girdi parametrelerinin değişimlerinin toplam ışıma ve bileşenleri üzerindeki etkileri elde edilmektedir. DAMA ile güneşin doğmasının hemen sonrasında ve batmasına yakın zamanlarda deniz kenarında ufka doğru bakıldığında gözlenen güneş pırıltısı modellenebilmektedir. Geliştirilen program, Schwenger ve Repasi'nin çalışmalarında [7] verilen ölçüm sonuçları ile uyumlu çıktılar vermektedir ve çıktılar ölçüm sonuçları ile Schwenger ve Repasi'nin çalışmalarında verilen benzetim yazılımı çıktılarına göre daha gerçekçi bir şekilde örtüşmektedir. DAMA ve açık literatürde bulunan, kızılötesi bantta deniz ışımalarını hesaplayan ve kısıtlı özelliklere sahip olan SEARAD yazılımı ortak girdi parametreleri ile çalıştırıldığında çıktıların uyumlu olduğu görülmektedir. ...
... Bu çalışma kapsamında yapılan analizler sonucunda tüm bakış eğim açıları, rüzgar hızları ve deniz yüzeyi modelleri için elde edilen grafiklerde güneş pırıltısı yolunun belirtilen özellikleri sağlanmaktadır. Bu çalışma kapsamında DAMA yazılımı ölçüm sonuçları ile de geçerlenmiştir [7]. ...
The success of infrared surveillance systems is proportional to distinguishing the radiance of the target and the background. In this study, a simulation program, namely, DAMA (InfrareD SeA Modeling and Analysis) software is developed to calculate the total radiance and its components seen by an observer looking towards the sea surface. Total radiance is composed of thermal sea radiance, reflected source (sun or moon) radiance, reflected sky radiance and path radiance seen by the observer. The sea surface models are directly effective on the calculation of reflected radiances. DAMA program includes Cox and Munk, Mermelstein and Shaw-Churnside models and models for six different saltiness levels. The inputs of the software are classified as observer, time, wavelength, medium, model selection and output file parameters. The software developed by modules gives the total radiance and its components as output. The program has been run for approximately 10,000 times to obtain the effects of input parameters on total radiance and its components and the outputs are examined. DAMA is validated by SEARAD software and measurement results. By this way, for the first time in open literature, a simulation program including all sea surface and saltiness models is developed and the behavior of the total radiance and its components are examined with respect to the variation of all input parameters.
... Perlin Noise can also be exploited to generate cloudy skies (see Figure 33). Schwenger has used a similar approach in [16]. Here again, a 3D cube can be used to generate animations of clouds. ...
In this paper, physically-based simulations of naval scenarios are presented. Simulating naval operations is a challenge since the maritime environment is complex, dynamic, and many physical effects have to be taken into account. This paper focuses on three different components: ship signatures, atmospheric propagation and sea modelling. To cope with these problems, the electro-optic simulator SAFIR has been used. Its interoperability capabilities allow us to exploit its models and/or other reference software. The user defines his own scenario (time of the day, atmospheric conditions, sea parameters, geometry and orientation of the ship, etc). A thorough knowledge of the 3D geometry of the ships, as well as the physical properties of its materials, is required. The vessel is meshed and an infrared simulation software (OSMOSIS from RMA), estimates its temperature at each node. The material properties govern both the reflectivity (known as BRDF) and the emissivity (ε) of the ship. The emitted radiance (not necessarily diffuse), and the reflected radiance are combined to determine the ship's signature. The atmospheric conditions strongly influence the net flux and the reflected radiance. We use MATISSE 1.4 (from ONERA), to compute the sun and sky irradiance. It is also used to calculate atmospheric effects along the line of sight. The sea surface geometry is modelled as a composition of waves at various frequencies. The animations of the sea surface and the ship are considered. Finally, our spectral ray-tracer naturally takes into account the reflection of the vessel on the sea surface. An innovative approach is used to reduce memory usage and computation time with the help of a view-dependent mesh. To conclude, MWIR and LWIR simulations are presented. Various models and parameters are used to describe each sensor. We will study the effect of the atmospheric and sensor parameters on the probability of detection. For a given scenario (atmosphere, sea, ship and sensor), the whole simulator provides the capability to estimate probabilities of detection, maximum range and other significant quantities. It is also a valuable tool for vessel signature reduction technique.
... Bu çalışma kapsamında yapılan analizler sonucunda tüm bakış eğim açıları, rüzgar hızları ve deniz yüzeyi modelleri için elde edilen grafiklerde güneş pırıltısı yolunun belirtilen özellikleri sağlanmaktadır. Bu çalışma kapsamında DAMA yazılımı ölçüm sonuçları ile de geçerlenmiştir [7]. ...
The sea surface statistical models are directly related with the calculation of the total radiance reflected from the sea surface seen by the observer. Three major work which model the sea surface slope statistics are Cox and Munk, Mermelstein and Shaw-Churnside models. The reflected sea radiance observed by an observer is composed of path radiance, reflected sky radiance, reflected sun radiance and blackbody emission of the sea surface. By this work, the total radiance observed by the observer can be calculated for different sea surface slope models and saline sea environments. The outputs of MATLAB program are in good agreement with the previous studies
A 3D simulation of the dynamic sea surface populated with whitecaps is presented. The simulation considers the dynamic evolution of whitecaps depending on wind speed and fetch. It is suitable for imaging simulations of maritime scenarios. The calculation of whitecap radiance is done in the SWIR spectral band by considering wave hiding and shadowing, especially occurring at low viewing angles. Our computer simulation combines the 3D simulation of a maritime scene (open sea/clear sky) considering whitecaps with the simulation of light from a light source (e.g. laser light) reflected at the sea surface. The basic sea surface geometry is modeled by a composition of smooth wind driven gravity waves. The whitecap generation is deduced from the vertical acceleration of the sea surface, i.e. from the second moment of the wave power density spectrum. To predict the view of a camera, the sea surface radiance must be calculated for the specific waveband with the emitted sea surface radiance and the specularly reflected sky radiance as components. The radiances of light specularly reflected at the windroughened sea surface without whitecaps are modeled by considering an analytical statistical sea surface BRDF (bidirectional reflectance distribution function). A specific BRDF of whitecaps is used by taking into account their shadowing function. The simulation model is suitable for the pre-calculation of the reflected radiance of a light source for near horizontal incident angles where slope-shadowing of waves has to be considered. The whitecap coverage is determined from the simulated image sequences for different wind speeds and is compared with whitecap coverage functions from literature. A SWIR-image of the water surface of a lake populated with whitecaps is compared with the corresponding simulated image. Additionally, the impact of whitecaps on the radiation balance for a bistatic configuration of light source and receiver is calculated for different wind speeds.
A 3-D simulation of the polarization-dependent reflection of a Gaussian shaped laser beam on the dynamic sea surface is presented. The simulation considers polarized or unpolarized laser sources and calculates the polarization states upon reflection at the sea surface. It is suitable for the radiance calculation of the scene in different spectral wavebands (e.g. near-infrared, SWIR, etc.) not including the camera degradations. The simulation also considers a bistatic configuration of laser source and receiver as well as different atmospheric conditions. In the SWIR, the detected total power of reflected laser light is compared with data collected in a field trial. Our computer simulation combines the 3-D simulation of a maritime scene (open sea/clear sky) with the simulation of polarized or unpolarized laser light reflected at the sea surface. The basic sea surface geometry is modeled by a composition of smooth wind driven gravity waves. To predict the input of a camera equipped with a linear polarizer, the polarized sea surface radiance must be calculated for the specific waveband. The s- and p-polarization states are calculated for the emitted sea surface radiance and the specularly reflected sky radiance to determine the total polarized sea surface radiance of each component. The states of polarization and the radiance of laser light specularly reflected at the wind-roughened sea surface are calculated by considering the s- and p- components of the electric field of laser light with respect to the specular plane of incidence. This is done by using the formalism of their coherence matrices according to E. Wolf [1]. Additionally, an analytical statistical sea surface BRDF (bidirectional reflectance distribution function) is considered for the reflection of laser light radiances. Validation of the simulation results is required to ensure model credibility and applicability to maritime laser applications. For validation purposes, field measurement data (images and meteorological data) was analyzed. An infrared laser, with or without a mounted polarizer, produced laser beam reflection at the water surface and images were recorded by a camera equipped with a polarizer with horizontal or vertical alignment. The validation is done by numerical comparison of measured total laser power extracted from recorded images with the corresponding simulation results. The results of the comparison are presented for different incident (zenith/azimuth) angles of the laser beam and different alignment for the laser polarizers (vertical/horizontal/without) and the camera (vertical/horizontal).
A wake will appear behind the ship with long continuous time when the ship sail in the sea. The apparent wake can be detected with infrared imaging device. By analyzing infrared imaging characterization of the wake, many parameters of the ship can be gained in principle, such as ship's dimensions, course and speed. Combined JONSWAP wave spectrum with Kelvin wake model based on hydrodynamics, the simulation of Kelvin wake in the maritime environment is achieved. Meanwhile, a physics based model of infrared imaging for ship wake is proposed. Then the infrared characterization of ship Kelvin wake in different detection conditions is obtained, considering sky background radiation, atmospheric transmission, and other environmental factors. The results show that ship speed range can be estimated by using the infrared imaging characterization of ship wake. The detection and recognition probability of ship can be improved with appropriate detection azimuth.
A 3D simulation of the reflection of a Gaussian shaped laser beam on the
dynamic sea surface is presented. The simulation is suitable for the
pre-calculation of images for cameras operating in different spectral
wavebands (visible, short wave infrared) for a bistatic configuration of
laser source and receiver for different atmospheric conditions. In the
visible waveband the calculated detected total power of reflected laser
light from a 660nm laser source is compared with data collected in a
field trial. Our computer simulation comprises the 3D simulation of a
maritime scene (open sea/clear sky) and the simulation of laser beam
reflected at the sea surface. The basic sea surface geometry is modeled
by a composition of smooth wind driven gravity waves. To predict the
view of a camera the sea surface radiance must be calculated for the
specific waveband. Additionally, the radiances of laser light specularly
reflected at the wind-roughened sea surface are modeled considering an
analytical statistical sea surface BRDF (bidirectional reflectance
distribution function). Validation of simulation results is prerequisite
before applying the computer simulation to maritime laser applications.
For validation purposes data (images and meteorological data) were
selected from field measurements, using a 660nm cw-laser diode to
produce laser beam reflection at the water surface and recording images
by a TV camera. The validation is done by numerical comparison of
measured total laser power extracted from recorded images with the
corresponding simulation results. The results of the comparison are
presented for different incident (zenith/azimuth) angles of the laser
beam.
This paper describes an experiment-based model to simulate 3D infrared
target in sea background. The geometric model of dynamic sea waves are
based on gravity wave theory, while the sky model use SkyDome technology
which takes the sky as a dome covers the land. To create the infrared
images of sea waves and sky, the radiance at the detector are
calculated. To acquire the radiance of target radiation, a spectrometer
is settled to measure the radiant emitted by interested targets.
Meanwhile, the radiant of sky is also measured to provide reference
data. Furthermore, the spectrometer is used to measure the atmospheric
transmission rate which is compared to the values calculated by MODTRAN.
The optical system is simulated based on OTF theory. The detector noise
is expressed as an equivalent Gaussian white noise. Finally, the
simulation images are compared with the practical images. It has been
proved that this experiment-based model produces infrared images with
high reality.
A D simulation of the reflection of a Gaussian shaped laser beam on the dynamic sea surface is presented. The simulation is suitable for both the calculation of images of SWIR (short wave infrared) imaging sensor and for determination of total detected power of reflected laser light for a bistatic configuration of laser source and receiver at different atmospheric conditions. Our computer simulation comprises the D simulation of a maritime scene (open sea/clear sky) and the simulation of laser light reflected at the sea surface. The basic sea surface geometry is modeled by a composition of smooth wind driven gravity waves. The propagation model for water waves is applied for sea surface animation. To predict the view of a camera in the spectral band SWIR the sea surface radiance must be calculated. This is done by considering the emitted sea surface radiance and the reflected sky radiance, calculated by MODTRAN. Additionally, the radiances of laser light specularly reflected at the wind-roughened sea surface are modeled in the SWIR band considering an analytical statistical sea surface BRDF (bidirectional reflectance distribution function). This BRDF model considers the statistical slope statistics of waves and accounts for slope-shadowing of waves that especially occurs at flat incident angles of the laser beam and near horizontal detection angles of reflected irradiance at rough seas. Simulation results are presented showing the variation of the detected laser power dependent on the geometric configuration of laser, sensor and wind characteristics.
The sea surface is an important background of sea surface targets and its infrared radiation characterization is the basis of infrared detection and recognition of sea surface targets. The model of linear wave theory and JONSWAP spectrum was used to achieve shape and display of ocean wave. Real-time infrared characterization simulation model was established for the sea surface, considering the solar radiance, the sky background radiance and the effects of atmospheric transmission. The sea surface was divided into many wavelet facets and the radiance of every facet was calculated on the basis of infrared simulation model. Three typical time, 9:00, 12:00, 15:00 in one day were selected and the radiance of every facet about 3~5μm and 8~12μm waveband was respectively calculated. Real-time infrared simulation images of sea surface were generated by converting wavelet facets radiance value to image gray value. The results show that the sea surface infrared images of 3~5μm waveband display the bright spots of reflected solar irradiance, and the spots appear the same direction with the solar azimuth while the impact of solar irradiance in 8~12μm waveband isn't usually required, and the method can effectively simulate the sea surface infrared characterization and provide the basis for infrared detection and recognition of sea surface targets.
The paper introduces the segmentation method of binary image and grey level image into the design of simulated pattern painting. They are region segmentation method based on seeds and region segmenting and merging method. Firstly according to the design characteristic of simulated pattern painting, it carries on the chromatic image segmentation regarding chromatic aberration as distance. Secondly carry on mathematical morphology treatment to layers of different colors. Thirdly it forms multiple spots of simulated pattern painting based on Matlab programming. We get the spots of simulated pattern painting using computer to carry on the color segmenting and clustering. We can find that the color, dimension and shape are unanimous or very similar to the spots on the background. It accords with the design principle of simulated pattern painting.
Expressions are obtained for the infrared radiance of a rough ocean surface by a method closely following that of Cox and Munk. For normal viewing, average radiance is found to be practically independent of roughness, but for oblique viewing blackness increases with roughness. Shadowing is treated in a simple approximate manner and estimates of surface radiance are made for glancing angles of viewing: an infrared horizon is predicted. The theoretical results, which are closely substantiated by observations, reveal the dish-shaped character of the ocean and indeed of any rough surface.
The directional distribution of radiant flux reflected from roughened surfaces is analyzed on the basis of geometrical optics. The analytical model assumes that the surface consists of small, randomly disposed, mirror-like facets. Specular reflection from these facets plus a diffuse component due to multiple reflections and/or internal scattering are postulated as the basic mechanisms of the reflection process. The effects of shadowing and masking of facets by adjacent facets are included in the analysis. The angular distributions of reflected flux predicted by the analysis are in very good agreement with experiment for both metallic and nonmetallic surfaces. Moreover, the analysis successfully predicts the off-specular maxima in the reflection distribution which are observed experimentally and which emerge as the incidence angle increases. The model thus affords a rational explanation for the off-specular peak phenomenon in terms of mutual masking and shadowing of mirror-like, specularly reflecting surface facets.
A method is developed for interpreting the statistics of the sun’s glitter on the sea surface in terms of the statistics of the slope distribution. The method consists of two principal phases: (1) of identifying, from geometric considerations, any point on the surface with the particular slope required for the reflection of the” sun’s rays toward the observer; and (2) of interpreting the average brightness of the sea surface in the vicinity of this point in terms of the frequency with which this particular slope occurs. The computation of the probability of large (and infrequent) slopes is limited by the disappearance of the glitter into a background consisting of (1) the sunlight scattered from particles beneath the sea surface, and (2) the skylight reflected by the sea surface.
The method has been applied to aerial photographs taken under carefully chosen conditions in the Hawaiian area. Winds were measured from a vessel at the time and place of the aerial photographs, and cover a range from 1 to 14 m sec⁻¹. The effect of surface slicks, laid by the vessel, are included in the study. A two-dimensional Gram-Charlier series is fitted to the data. As a first approximation the distribution is Gaussian and isotropic with respect to direction. The mean square slope (regardless of direction) increases linearly with the wind speed, reaching a value of (tan16°)² for a wind speed of 14 m sec⁻¹. The ratio of the up/ downwind to the crosswind component of mean square slope varies from 1.0 to 1.9. There is some up/downwind skewness which increases with increasing wind speed. As a result the most probable slope at high winds is not zero but a few degrees, with the azimuth of ascent pointing downwind. The measured peakedness which is barely above the limit of observational error, is such as to make the probability of very large and very small slopes greater than Gaussian. The effect of oil slicks covering an area of one-quarter square mile is to reduce the mean square slopes by a factor of two or three, to eliminate skewness, but to leave peakedness unchanged.
An analytical approach of the two-dimensional emissivity of a rough sea surface in the infrared band is presented. The emissivity characterizes the intrinsic radiation of the sea surface. Because the temperature measured by the infrared camera depends on the emissivity, the emissivity is a relevant parameter for retrieving the sea-surface temperature from remotely sensed radiometric measurements, such as from satellites. This theory is developed from the first-order geometrical-optics approximation and is based on recent research. The typical approach assumes that the slope in the upwind direction is greater than the slope in the crosswind direction, involving the use of a one-dimensional shadowing function with the observed surface assumed to be infinite. We introduce the two-dimensional shadowing function and the surface observation length parameters that are included in the modeling of the two-dimensional emissivity.
Incorporating polarization parameters into the lighting model can enhance the physical realism of images rendered with a ray tracer. Polarization effects can be important in certain scenes, and the difference in rendering even simple scenes with and without proper treatment of polarization can be striking. All light waves possess a state of polarization, which changes almost every time light reflects off a material surface. A single reflection partially polarizes and may even completely polarize previously unpolarized light. Polarization influences the rendering of a scene because the reflected radiant intensity depends largely on the incident light waves's polarization state. E. Wolf's (1959) coherence matrix formalism of polarization has been incorporated into the Torrance-Sparrow reflectance model. This combination allows elegant quantitative derivations of the altered polarization state of light upon reflection in a ray tracer. Comparisons of identical scenes rendered with a conventional ray tracer and the ray tracer presented incorporating a polarization model show that the present method renders specular interobject reflections more accurately with respect to reflected radiance and color.< >
The emission and reflection properties of a two-dimensional (2D)
Gaussian rough sea surface are investigated. The emissivity and
reflectivity study is of importance for accurate measurement of the
temperature distribution of a wind-roughened water surface by infrared
thermal imaging. The radius of curvature of the capillary waves being
much larger than the wavelength involves the fact that their statistical
model is based on the first order geometrical-optics method. In this
paper, the first order geometrical optics approximation is considered
because the higher order approximations can be neglected compared to the
first order approximations. Indeed the total reflected electromagnetic
field is damped by the multiple reflections on the surface. They
introduce the 2D shadowing function, and the observed surface length in
the azimuthal direction with respect to the wind direction. Their
emissivity and reflectivity model is obtained from recent works. In
order to use the 1D (1D) shadowing function, assume that the average
slope of the surface is smaller than unity. Assuming an infinite
observed surface length as determines only the emissivity with the 2D
shadowing function. Applying their emissivity and reflectivity model,
the relation determining the apparent ocean surface temperature is
given, in the considered wavelength band, according to the following
parameters: wind speed, wavelength, polarization, intrinsic surface
temperature, atmospheric transmission coefficient as evaluated by the
Taylor and Larmor model, receiver location, and the camera field of view
In the development and analysis of sophisticated IR detection and recognition systems it is necessary to have a priori knowledge of the background clutter. The spatial power spectrum and the spatial autocorrelation functions are used to analyze the spatial structure of ground- based infrared cloudy sky images. The experimental results obtained for ground-based IR cloudy sky images do not fit the analytical model of the Wiener spectrum that is frequently used to describe natural clutter sources in the infrared. The spatial structure of ground-based cloudy sky IR images was found to be dependent on the percentage cloud cover in the image. A corrected model is developed that relates the spatial structure of cloudy sky images to the percentage cloud cover in the image.
In the development and analysis of sophisticated detection systems, the ability to simulate background clutter provides a useful tool in assessing system performance. Infrared cloudy sky images can be generated using a technique based on the physical parameters of the background. Power spectra and radiance distribution functions for ground-based IR cloudy sky images in 3 to 5 (mu) and 8 to 12 (mu) spectral windows were determined experimentally in our previous research. The empirical power spectra and radiance distribution functions are used here as a basis to generate computer simulations of cloudy sky images. We require that the power spectrum and radiance distribution functions of the simulated image be in agreement with the functions that we obtained experimentally in our previous research. Realistic cloudy sky images were obtained. The simulated cloudy sky images retain the radiance distribution and power spectra functions of real cloudy sky IR images.
The first‐order probability densities of height and slope on the portions of a rough surface illuminated jointly by rays in two arbitrary directions have been evaluated for a surface generated by a stationary normal random process. The dependence of the statistical properties of the apparent surface on the angles of incidence and observation is illustrated. For the case of backscatter, a simple calculation of the “shadowing functions,” valid at very small grazing angles, and a more detailed approximation, valid at the larger angles, are shown to be in excellent agreement with the computer‐simulation results of Brockelman and Hagfors.
The statistical behavior of ground-based IR cloudy sky images are analyzed to acquire a priori knowledge of the background necessary for the development of sophisticated infrared target detection and recognition systems. Infrared cloudy sky images can be automatically segmented into their related groups of interest, according to the statistical radiance distribution, by implementing specialized image processing techniques. Once the images have been segmented, the cloud cover, statistical distribution, and other parameters of interest are readily obtained. It was found that, in the 8- to 12-micrometers spectral window, cloudy sky images must be divided into at least five regions of interest, and in the 3- to 5-micrometers spectral window, a distinction must be made between shaded cloud images and sunlit cloud images. Shaded cloud images have only three regions of interest, whereas sunlit cloud images are more complex and have at least five regions of interest. If each region is approximated by a Gaussian distribution, then the normalized cross-correlation function of the measured data with the multinormal function gives a value in excess of 0.9.
Present systems simulate sea surfaces either in the visible or in the IR band. A physics based 3D simulation of sea surfaces for the calculation of images for multiband cameras is presented here. Dynamic sea surfaces, composed of smooth wind-driven gravity waves, are generated by means of time dependent statistical models. In addition, choppy waves are modeled to improve the realism of the rough sea. The appearance of the sea in the visible and thermal bands is modeled. Sea surface radiance in the IR band is calculated with respect to the reflected sky radiance and the emitted sea surface radiance. Sun glint simulations in the visible and IR are presented. Polarization effects were incorporated to enhance the physical realism. As an example for an application a real-time animation of a sea surface with floating foreground objects is shown. The simulated images of the sea surface are in good accordance with real images.© (2003) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
It is possible to generate synthetic images of the sea by creating a sea relief and by calculating its image using raytracing techniques. The sea surface is generated using sea wave statistics which are wind speed and direction dependent. A known power spectrum of the sea wave is used to filter white noise and the filtered noise is used as the sea relief. After that, the image of the sea is calculated using a ray-tracing technique which considers the emitted radiance of the sea, the specular reflection of the sky radiance and the reflection of the sun radiance (which is the sun irradiance scattered by the atmosphere). The resulting sun glint has been compared with the Cox and Munk model in order to make final adjustments to the simulation parameters. Thus, this model is able to produce images with well calibrated radiance and good sun glint distribution.© (1990) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
In the context of the backscattering of waves from a random rough surface, a theoretical model is used to investigate the geometrical self-shadowing of a surface described by Gaussian statistics. Expressions are derived for various shadowing probabilities as functions of the parameter characterizing surface roughness and of the angle of incidence of the illuminating beam. The theoretical shadowing functions compare closely with those obtained experimentally from a recent computer simulation of a Gaussian surface.