Nonuniform Media Research Articles

Multi-Attention DenseNet: A Scattering Medium Imaging Optimization Framework for Visual Data Pre-Processing of Autonomous Driving Systems

The vision system is important for almost all kinds of autonomous driving systems. However, visual data interfered by scattering media, such as smoke, haze, water, and other non-uniform media will be degraded seriously, showing the characteristics of detail loss, poor contrast, low visibility, or color distortion. These characteristics can significantly interfere with the reliability of autonomous driving systems. In real environments the image degradation mechanism is complex, and the estimation of degradation parameters is difficult. This issue remains to be solved. In this study, we employed dense blocks as the framework and introduced the attention mechanism to our model from four dimensions: Multi-scale Attention, Channel Attention, Structure Attention, and ROI (region of interest) Attention. With the help of the training data provided by the weakly supervised model, the proposed method achieved excellent performance in the task of scattering medium imaging optimization in different scenes. Comparative experiments show that the proposed method is robust, and is superior to other state-of-the-art methods in image dehazing, and underwater image enhancement tasks. It is of great significance to improve the reliability of autonomous driving systems in underwater and severe weather environments.

A semiempirical approach to low-energy cosmic ray propagation in the diffuse interstellar medium

Context. We investigate the ionization of the diffuse interstellar medium by cosmic rays by modeling their propagation along the wandering magnetic fields using a Monte Carlo method. We explore how particle trapping and second-order Fermi processes affect the ionization of the medium. Aims. We study how low-energy comic rays propagate in turbulent, translucent molecular clouds, and how they regulate the ionization and both lose and gain energy from the medium. Methods. As a test case, we used high spatial resolution (0.03 pc) CO maps of a well-studied high latitude translucent cloud, MBM 3, to model turbulence. The propagation problem is solved with a modified Monte Carlo procedure that includes trapping, energization, and ionization losses. Results. In the homogeneous medium, trapping and re-energization do not produce a significant effect. In the nonuniform medium, particles can be trapped for a long time inside the cloud. This modifies the cosmic ray distribution due to stochastic acceleration at the highest energies (∼100 MeV). At lower energies, the re-energization is too weak to produce an appreciable effect. The change in the energy distribution does not significantly affect the ionization losses, so ionization changes are due to trapping effects. Conclusions. Our Monte Carlo approach to cosmic ray propagation is an alternative method for solving the transport equation. This approach can be benchmarked to gas observations of molecular clouds. Using this approach, we demonstrate that stochastic Fermi acceleration and particle trapping occurs in inhomogeneous clouds, significantly enhancing their ionization.

Constant intensity acoustic propagation in the presence of non-uniform properties and impedance discontinuities: Hermitian and non-Hermitian solutions.

Propagation of sound through a non-uniform medium without scattering is possible, in principle, if the density and acoustic compressibility assume complex values, requiring passive and active mechanisms, also known as Hermitian and non-Hermitian solutions, respectively. Two types of constant intensity wave conditions are identified: in the first, the propagating acoustic pressure has constant amplitude, while in the second, the energy flux remains constant. The fundamental problem of transmission across an impedance discontinuity without reflection or energy loss is solved using a combination of monopole and dipole resonators in parallel. The solution depends on an arbitrary phase angle that can be chosen to give a unique acoustic metamaterial with both resonators undamped and passive, requiring purely Hermitian acoustic elements. For other phase angles, one of the two elements must be active and the other passive, resulting in a gain/loss non-Hermitian system. These results prove that uni-directional and reciprocal transmission through a slab separating two half spaces is possible using passive Hermitian acoustic elements without the need to resort to active gain/loss energetic mechanisms.

Experimental and numerical investigation on the urea-deposit formation at different severities in selective catalytic reduction systems

Complex Exhaust Aftertreatment System (EATS) designs with Selective Catalytic Reduction (SCR) system can suffer deposit problems due to the liquid film formation in critical regions. Experimentally validated numerical prediction methods can help detect these problematic areas to eliminate deposit formation problems in product development and ongoing product development stages of EATS. This study investigates the spray-wall interaction, the liquid film formation at different severities, the liquid film evaporation by injecting Urea Water Solution (UWS) into an impingement plate in a non-uniform flow medium under different operating conditions, experimentally and numerically. The experimental set-up includes a backward facing step flow through an impingement plate. During experiments, after a specified time, the UWS injection was stopped, and a non-uniform liquid film evaporation due to non-uniform flow was also investigated, separately. Experimental and finite volume based numerical studies show that the liquid film layer formed in the stagnant flow regions has greater risks in terms of deposit formation than the other regions in an EATS. The wall cooling occurred mostly on the spray footprint portion of the impingement plate due to the liquid film formation during UWS injection, the liquid film in this part was rapidly evaporated after UWS injection was stopped and no deposit formation was observed in this part for each case under study. Experimental and numerical studies show that deposit formation occurred in the stagnant regions where the liquid film did not evaporate. The numerical results agree well with the experimental results and they demonstrate the predictive capability of the numerical model for the deposit formation.

Turbulence generation by shock interaction with a highly nonuniform medium.

An initially planar shock wave propagating into a medium of nonuniform density will be perturbed, leading to the generation of postshock velocity perturbations. Using numerical simulations we study this phenomenon in the case of highly nonuniform density (order-unity normalized variance, σ_{ρ}/ρ[over ¯]∼1) and strong shocks (shock Mach numbers M[over ¯]_{s}≳10). This leads to a highly disrupted shock and a turbulent postshock flow. We simulate this interaction for a range of shock drives and initial density configurations meant to mimic those which might be presently achieved in experiments. Theoretical considerations lead to scaling relations, which are found to reasonably predict the postshock turbulence properties. The turbulent velocity dispersion and turbulent Mach number are found to depend on the preshock density dispersion and shock speed in a manner consistent with the linear Richtymer-Meshkov instability prediction. We also show a dependence of the turbulence generation on the scale of density perturbations. The postshock pressure and density, which can be substantially reduced relative to the unperturbed case, are found to be reasonably predicted by a simplified analysis that treats the extended shock transition region as a single normal shock.

Coherent Stimulated Radiation, Coherent Spontaneous Radiation and Coupled Photon Pairs in Non-Uniform Thermally Excited Gaseous Media: Calculation Procedure

This work continues the author’s study of the processes of interaction between a quantum electromagnetic field and a spatially non-uniform gaseous medium [1, 5]. Previously unknown optical effects caused by consideration of high-order quantum correlators were discovered. This is caused by the problem of calculation of high-order photonic correlators starting from the fourth order, which appears when the interaction between the quantum electromagnetic field is described using the Feynman diagrams, According to T.A. Matsubara’s technique applied nowadays, these correlators are just neglected. The author’s articles [1, 5] describe the calculation method allowing us to avoid thiscomplication. The method is based on the following. High-order quantum correlators appear when multi-particle problems are studied. If the number of particles is small (hydrogen atom), the problem disappears. Therefore, if we replace a multi-particle problem with a single-particle problem (even if this particle is rather complex), the problem will be solved. Such solution (by using the Γ-operators technique) is described in the said author’s articles. While the standard secondary quantisation technique uses operators of creation and destruction of separate particles, the Γ-operators technique deals with operators of creation and destruction of photon conglomerates (systems). It turns out that high-order correlators just do not appear when the Γ-operators technique is used. This article demonstrates a specific example of application of the Γ-technique.

Quantitative comparison of the DRESOR and Monte Carlo methods for calculating radiative heat flux

The Monte Carlo method (MCM) can be used to calculate the radiative heat flux and divergence with high accuracy, and the Distribution of Ratios of Energy Scattered by the medium Or Reflected by the boundary surface (DRESOR) method based on the MCM provides the radiation intensity with high directional resolution. However, the accuracy and efficiency of the DRESOR method for calculating the radiative heat flux and divergence have not been verified. Therefore, in this article, the DRESOR method is reformed to make it more suitable for calculating the radiative heat flux and divergence, and the DRESOR method and MCM are compared quantitatively. In the comparison, the influences of the Mean Optical Thickness per Element (MOTE), the surface and medium parameters, anisotropic scattering, and a nonuniform medium are considered. The results show that the figure of merit (FoM) of the reformed DRESOR method is 81.18 and 6.04 times that of the MCM for the surface element and space element, respectively when the MOTE is 0.17 for calculating the radiative heat flux and divergence. The larger the surface emissivity, and the smaller the scattering albedo, the better the performance of the DRESOR method than the MCM is. Meanwhile, the MOTE (mesh density), anisotropic scattering, and the nonuniform medium have similar influences on the DRESOR method and the MCM.

Single underwater image restoration based on descattering and color correction

Underwater images typically exhibit color distortion and low contrast due to the absorption and scattering of the non-uniform medium during the propagation of light underwater. To address these problems, an effective underwater image restoration method is presented in this paper. Firstly, we adopted an adaptive dark channel prior that can be applied to underwater images with different color tones. Then, the local water light with strong robustness is obtained by utilizing maximum filtering and Gaussian low-pass filtering. Additionally, a linear relationship between blurred and clear images in the dark channel was introduced combining with the attenuation coefficient compensation to estimate the transmission of r, g and b channels. Ultimately, we designed a new adaptive color compensation method by taking into account the attenuation difference of the three color channels. Simulation and experimental results show that the proposed method can effectively improve the clarity and contrast of the image and eliminate color distortion.

Towards machine learning for heterogeneous inverse scattering in 3D microscopy.

Light propagating through a nonuniform medium scatters as it interacts with particles with different refractive properties such as cells in the tissue. In this work we aim to utilize this scattering process to learn a volumetric reconstruction of scattering parameters, in particular particle densities. We target microscopy applications where coherent speckle effects are an integral part of the imaging process. We argue that the key for successful learning is modeling realistic speckles in the training process. To this end, we build on the development of recent physically accurate speckle simulators. We also explore how to incorporate speckle statistics, such as the memory effect, in the learning framework. Overall, this paper contributes an analysis of multiple aspects of the network design including the learning architecture, the training data and the desired input features. We hope this study will pave the road for future design of learning based imaging systems in this challenging domain.

Monitoring of microscopic behavior of nano-enhanced colloidal gas aphron in fractured and un-fractured non-uniform porous medium

The present article aimed to examine the pore-scale examination of a fluid capability to block pores and fractures deduced by return permeability and the blocking ability of a novel lightweight Colloidal Gas Aphron Nano-Fluid (CGANF) in heterogeneous fractured/un-fractured porous medium. Silica (S) and fumed silica (FS) nanoparticles, Sodium Dodecyl Sulfate (SDS) anionic surfactant, and Xanthan Gum (XG) bio-polymer were employed for following up CGANF bubbles' movements in incongruous micro-models’ inclusive crack. The transparent nature of the glass micro-model provides an intuition into the lenticular behavior of CGANF bubbles in the heterogeneous porous media. Differential pressure analysis through porous media during experiments demonstrated that opposition to the stream of CGANF bubbles increased as more CGANF fluid was injected. Also, the lamella division mechanism generated more small bubbles resulting in increasing differential pressure across porous micro-models. Contrasted to bigger CGANF bubbles, smaller CGANF bubbles had lesser movability. Therefore, they were key factors for blocking pores and throats of porous media and changed the orientation of bigger bubbles in cracks toward the matrix. Observations revealed that CGANF micron-bubbles built up through the porous media could set up an impressive snag to control loss of filtrate, and permeability of models was reverted almost to its primary permeability when saturation fluid was reinjected into micro-models. For the detailed analysis in this study, the maximum percentages of reverted permeability for fractured and un-fractured non-uniformed micro-models were attained 91.81% and 90.70%, respectively. Also, by increasing XG concentrations from 0.286 to 0.571% w/v, return permeability percentage increased from 84.4% to 87.4%. There was an optimum concentration of nanoparticles at 0.0571% w/v of silica and fumed silica, and consequently, returned permeability and infusion pressure of CGANF fluid reached their maximum values. Moreover, the stability of the CGANF fluid increased to 8.15% and 12.6% by adding silica and fumed silica nanoparticles, respectively, to the CGA fluid formulation without nanoparticles. This article presents a novel insight into porous media for analyzing blockage capability and controlling porous media damage created via CGANF-based fluids in non-uniformed environments. • Enhancing the behavior and properties of CGANF fluids. • Use of silica and fumed silica nanoparticles in CGANF fluids. • Controlling filtration loss by forming an effective seal using CGA/CGANF fluids. • Use of heterogeneous transparent glass micromodels for monitoring the bubbles flowin the pores and fracture by microscope. • Analysis of return permeability by high accurate pressure drop data through micromodel.

Micro-mechanismof multiscale dynamic fracture

The shock propagation in non-uniform medium is studied theoretically and experimentally. A set of experiments on shock loading of high-strength low-alloyed AB2 steel is performed. The "trigger" mechanism for switching the dynamic fracture from one scale to another is shown to be the resonance interaction of plastic flow oscillations and mesoscopic structural elements. There is a threshold strain rate at which the defect (decrease) of mass velocity at the plateau of compressive pulse increases abruptly. Under strain rates higher threshold value, a multitude of short transcrystalline cracks is formed, which results in decay of shock wave. Keywords: Shock loading, multiscale fracture, transcrystalline cracks.

Target Echo Detection Based on the Signal Conditional Random Field Model for Full-Waveform Airborne Laser Bathymetry

Airborne laser bathymetry (ALB) systems with digital full-waveform signal collection can obtain corresponding temporal positions from several backscattering surfaces by laser beam irradiation. This information can help describe the multi-elevation structures of the target and explore the echo signal attenuation response in different nonuniform mediums during laser propagation. Therefore, a full-waveform echo signal is quite practical for integrated water-land detection. However, the wavelength used in the ALB system is generally in the visible band range of 470~580 nm, and the received signal is constantly interfered with by many nontarget factors, such as imperfections in the receiving channel or the strong scattering from the transmission medium. The conventional processing method transforms nontarget interference into noise point cloud filtering or classification extraction, enabling the detection of a single surface or regular geometry. The accuracy of the identification and extraction for multi-elevation target surfaces echo signal is always reduced due to the significant noise signal intensity. We proposed a signal component detection method by constructing the echo signal feature functions and the conditional random field (CRF) model based on the full-waveform decomposition. The processing result for actual measurement data verified that the CRF strategy can effectively reduce the uncertainty of target surface detection. Compared with the single-beam echo sounder, the root mean square errors of the elevation deviation underwater were reduced by 3.2 cm and 4.9 cm respectively in the two different experimental areas.

Микромеханизм многомасштабного динамического разрушения

The shock propagation in non-uniform medium is studied theoretically and experimentally. A set of experiments on shock loading of high-strength low-alloyed AB2 steel is performed. The "trigger" mechanism for switching the dynamic fracture from one scale to another is shown to be the resonance interaction of plastic flow oscillations and mesoscopic structural elements. There is a threshold strain rate at which the defect (decrease) of mass velocity at the plateau of compressive pulse increases abruptly. Under strain rates higher threshold value, a multitude of short transcrystalline cracks is formed, which results in decay of shock wave,

Numerical approaches and analysis of optical measurements of laser radar cross-sections affected by aero-optical transmission

One of the concerns in the process of measuring laser radar cross-section (LRCS) using optical measurement methods is the aero-optical effect around a high-speed flow field. In this paper, a computational model for LRCS considering supersonic flow in a non-homogeneous medium is presented. First, by taking a typical 15° half-cone blunt cone as the research object, based on CFD calculation, the flow field distribution of the non-uniform medium around the high-speed vehicle is obtained. The ray-tracing method is used to calculate the light transmission in an aero-optical environment. An LRCS approach for arbitrarily complex targets in the aero-optical environment is then proposed, and the method is verified by analytical and numerical methods. Finally, the effects of incoming density, incoming Mach number, and material bidirectional reflectance distribution function (BRDF) parameters on LRCS measurements are investigated, and the errors caused by the steady aero-optical environment are analyzed. The results show that the maximum relative error of the measurement caused by the aero-optical environment in the direction of the shock-wave layer observation is significant when the Fresnel function approximation exponential function parameter is too large or when the root mean square (RMS) of the surface slope is too small in the BRDF parameters of the material. The influence of the aero-optical environmental flow field on the LRCS is characterized by a directional and regional distribution, determined by the incoming Mach number and shock-wave angle, with the influence of the incident flow density on the regional distribution being weakly correlated. The proposed method and calculation results are useful for the simulation, analysis, and calibration of LRCS measurements in complex aero-optical environments.

Learning-Based Motion Planning of a 14-DOF Biped Robot on 3D Uneven Terrain Containing a Ditch

In this paper, a new method is proposed that integrates the 3D terrain information, ditch geometry, and biped dynamics for motion planning of a 14-DOF biped robot on 3D terrain containing a 3D ditch. The path planning is modeled as a wavefront propagation in Non-uniform medium represented by the Eikonal equation. A speed function expresses the inhomogeneity of uneven terrain. The Eikonal equation is solved by the Fast marching method (FMM) to obtain the global path. Ten Footstep variables (FSVs) characterize one step of walk on a 3D uneven terrain surface. The hip and foot trajectory parameters (HFTPs) are used to construct cubic spline-based hip and foot trajectories of the biped robot. The optimal value of HFTPs is obtained by a Genetic algorithm by minimization of energy and satisfying the constraint of dynamic balance. A walk-dataset is created that contains the optimal trajectories for different FSVs. The generated walk-dataset was used to train the biped to walk on rough terrain using a feed-forward Neural network for making a real-time estimate of optimal HFTPs. Simulation results validate the effectiveness of the proposed method of ditch crossing on different uneven terrains.

Performance analysis of a sub-Rayleigh resolution distance spaced MIMO for 5G/WLAN applications

In this article, a two element mutiple-input multiple-output (MIMO) antenna is designed to operate in 4.98 GHz to 6.04 GHz. The antenna has inter-element edge to edge spacing of 0.4 mm (≈0.013λg/0.00664 λ0, where λg and λ0 are the guided and free space wavelength respectively at the lowest resonant frequency 4.98 GHz) which is extremely smaller than Rayleigh resolution distance. In spite of this, an acceptable port isolation is achieved by employing a stub in the common ground. Each radiating element comprises of etched rectangular patch having largest dimension ≈0.6 λg fed with 50 Ω microstrip line. The proposed MIMO has overall size of 40 × 26 × 1.67 (≈1.34 λg × 0.87 λg × 0.056 λg) mm3. Furthermore, the diversity parameters such as envelope correlation coefficient, diversity gain and mean effective gain have been studied in non-uniform medium. The total active reflection coefficient (TARC) and loss in channel capacity are found to be remain within the acceptable limit to exhibit the desired performances. The proposed antenna can be applied in 5G (below 6 GHz), WLAN and Wi-Fi bands.

Nonlinear dynamics and phase space transport by chorus emission

Chorus emission in planetary magnetospheres is taken as working paradigm to motivate a short tutorial trip through theoretical plasma physics methods and their applications. Starting from basic linear theory, readers are first made comfortable with whistler wave packets and their propagation in slowly varying weakly non-uniform media, such as the Earth’s magnetosphere, where they can be amplified by a population of supra-thermal electrons. The nonlinear dynamic description of energetic electrons in the phase space in the presence of self-consistently evolving whistler fluctuation spectrum is progressively introduced by addressing renormalization of the electron response and spectrum evolution equations. Analytical and numerical results on chorus frequency chirping are obtained and compared with existing observations and particle-in-cell simulations. Finally, the general theoretical framework constructed during this short trip through chorus physics is used to draw analogies with condensed matter and laser physics as well as magnetic confinement fusion research. Discussing these analogies ultimately presents plasma physics as an exciting cross-disciplinary field to study.

The nanofluid flows in the channel with linearly varying wall temperature

In many applications of heat exchanger, due to the heat convection, the wall temperature of the channel usually varies along the streamwise direction. The present paper reports the analysis of nanofluid flows in the channel with linearly varying wall temperature. The non-uniform equilibrium fluid medium model proposed by Buongiorno is applied, where the Brownian diffusion and thermophoresis of nanoparticles are considered. The numerical solutions of laminar flows are determined by the iteration method in conjunction with the Chebyshev collocation method. The results show that: (1) there is a Gaussian distribution of nanoparticle concentration in the wall-normal direction for convective cooling, while its standard deviation depends on the Peclet number and the ratio of Brownian diffusivity to thermophoretic diffusivity; (2) there are obvious stratifications for the viscosity and thermal conductivity in the flow; (3) the velocity decreases significantly with the increase of nanoparticle concentration for convective heating, while there is only a little decrease for the velocity in the cold region for convective cooling.

Resonance and mode locking in gaseous detonation propagation in a periodically nonuniform medium

Resonance and mode locking in gaseous detonation propagation in a periodically nonuniform medium

Improvement of Image Classification by Multiple Optical Scattering

Multiple optical scattering occurs when light propagates in a non-uniform medium. During the multiple scattering, images were distorted and the spatial information they carried became scrambled. However, the image information is not lost but presents in the form of speckle patterns (SPs). In this study, we built up an optical random scattering system based on an liquid crystal display (LCD) and an RGB laser source. We found that the image classification can be improved by the help of random scattering which is considered as a feedforward neural network to extracts features from image. Along with the ridge classification deployed on computer, we achieved excellent classification accuracy higher than 94%, for a variety of data sets covering medical, agricultural, environmental protection and other fields. In addition, the proposed optical scattering system has the advantages of high speed, low power consumption, and miniaturization, which is suitable for deploying in edge computing applications.