A variational formulation for physical noised image segmentation

Image segmentation is a hot topic in image science. In this paper we present a new variational segmentation model based on the theory of Mumford-Shah model. The aim of our model is to divide noised image, according to a certain criterion, into homogeneous and smooth regions that should correspond to structural units in the scene or objects of interest. The proposed region-based model uses total variation as a regularization term, and different fidelity term can be used for image segmentation in the cases of physical noise, such as Gaussian, Poisson and multiplicative speckle noise. Our model consists of five weighted terms, two of them are responsible for image denoising based on fidelity term and total variation term, the others assure that the three conditions of adherence to the data, smoothing, and discontinuity detection are met at once. We also develop a primal-dual hybrid gradient algorithm for our model. Numerical results on various synthetic and real images are provided to compare our method with others,these results show that our proposed model and algorithms are effective.     

Nonlinear PDEs with Modulated Dispersion I: Nonlinear Schrödinger Equations

We start a study of various nonlinear PDEs under the effect of a modulation in time of the dispersive term. In particular in this paper we consider the modulated non-linear Schrödinger equation (NLS) in dimension 1 and 2 and the derivative NLS in dimension 1. We introduce a deterministic notion of “irregularity” for the modulation and obtain local and global results similar to those valid without modulation. In some situations, we show how the irregularity of the modulation improves the well–posedness theory of the equations. We develop two different approaches to the analysis of the effects of the modulation. A first approach is based on novel estimates for the regularizing effect of the modulated dispersion on the non-linear term using the theory of controlled paths. A second approach is an extension of a Strichartz estimated first obtained by Debussche and Tsutsumi in the case of the Brownian modulation for the quintic NLS.     

The shot noise of the quantum dot weakly coupled to Luttinger liquid

We study the nonequilibrium transport through a single-level quantum dot weakly coupled to Luttinger liquid leads. A general shot noise expression is derived by using nonequilibrium Green function technique. We find that the differential shot noise and differential conductance demonstrate resonant-like behavior as a function of the bias voltage and the quantum dot's energy level for a weak or moderately strong interaction. In the limit of strong electron-electron interaction, the resonant behavior disappears and shows bias-voltage-dependent power law scalings. And the Fano factor also scales as a power law in high bias voltage region. In addition, the Fano factor is enhanced with the electron-electron interaction increased. It implies that the Fano factor can be controlled by tuning the electron-electron interaction in the leads.     

Inertial effects, mass separation and rectification power in Lévy ratchets

We study the dynamics of particles in an external anisotropic periodic potential under the influence of additive white Lévy noise, in a general not overdamped situation. Different quantities characterizing directionality, coherence and dispersion are analyzed as functions of the mass and other systems parameters. We show that, while the current decreases monotonously with the stability index of the Lévy noise, there exists a particular intermediate value of such parameter (slightly dependent on the mass) that minimizes the time required to form a coherent particle package advancing in the preferred direction. Moreover, we show the possibility of observing mass separation. This means that particles of different masses may advance in opposite directions when influenced by the same ratchet potential and the same Lévy noise. Finally, we show that the ratio of the advanced distance to the total distance travelled constitutes a relevant measure for the rectification power, useful not only for Lévy ratchets but also for general ratchets systems. In particular, we find that it behaves quite similar to the rectification efficiency for standard models of rocking and flashing ratchets found in the literature.     

Phase transitions induced by microscopic disorder: A study based on the order parameter expansion

Based on the order parameter expansion, we present an approximate method which allows us to reduce large systems of coupled differential equations with diverse parameters to three equations: one for the global, mean field, variable and two which describe the fluctuations around this mean value. The method is based on a systematic perturbation expansion and can be applied around the vicinity of the homogeneous state. With this tool we analyze phase transitions induced by microscopic disorder in three prototypical models of phase transitions which have been studied previously in the presence of thermal noise. We study how macroscopic order is induced or destroyed by time-independent local disorder and analyze the limits of the approximation by comparing the results with the numerical solutions of the self-consistency equation which arises from the property of self-averaging. Finally, we carry on a finite-size analysis of the numerical results and calculate the corresponding critical exponents.     

Flow-induced noise in a suction nozzle with a centrifugal fan of a vacuum cleaner and its reduction

Noise reduction in a vacuum cleaner with a brush nozzle for cleaning a bed blanket is investigated numerically in fluid dynamic aspects. Governing equations describing nonlinear flow fields in a suction nozzle are solved simultaneously. The components of a rotary fan, a brush drum, and a separation block are installed in the nozzle. First, flow patterns in the nozzle are analyzed and based on them, flow resistance is evaluated to find a primary noise source. Flow resistance induces the loss of a suction performance as well as noise generation. In the brush nozzle, the separation block and the rotary fan obstruct smooth air flow and result in high level of noise emission. The rotation of the fan itself affects little noise generation. From the numerical results, a method to reduce noise and maintain the suction performance is suggested. In this method, the suction performance is increased through the optimization of the separation block, which is attained by the modification of its shape. And then, the height of a fan blade is shortened, leading to the performance loss. At the cost of it, the sound power level of noise is reduced by 4-5 dB(A) and at the same time, the tonal noise and the sound quality are improved appreciably. The method has been verified by experimental tests. It is found that in the brush nozzle, flow resistance is critical in noise emission and accordingly, fluid dynamic approach to noise reduction is effective.     

Nuclear Spin‐Noise Spectra of Hyperpolarized Systems

Spin‐noise appeal : Detection of NMR spin‐noise is very appealing when dilute hyperpolarized species are considered. Continuous monitoring of the noise absorption at the Larmor frequency enables determination of T₁ and T₂*, independently of the static magnetic field. An inductively coupled microcoil located inside the NMR tube (see picture) allows acquisition of ¹²⁹Xe spin‐noise spectra without radio‐frequency excitation.