In image processing, the Rudin–Osher–Fatemi (ROF) model [L. Rudin, S. Osher, and E. Fatemi, Phys. D, 60 (1992), pp. 259–268] based on total variation (TV) minimization has proven to be very useful. So far many researchers have contributed designing fast numerical schemes overcoming nondifferentiability of model. Methods considered particularly efficient for ROF include Chan–Golub–Mulet (CGM) primal-dual method [T.F. Chan, G.H. Golub, P. Mulet, SIAM J. Sci. Comput., 20 (1999), 1964–1977],...

Recently augmented Lagrangian method has been successfully appliedto image restoration. We extend the to total variation (TV)restoration models with non-quadratic fidelities. will firstintroduce and present an iterative algorithm for TVrestoration a quite general fidelity. In each iteration, threesub-problems need be solved, two of which can very efficientlysolved via Fast Fourier Transform (FFT) implementation or closedform solution. third sub-problem iterativesolvers. then apply our TV...

With the rapid growth in availability of data and computational technologies, multiple machine learning frameworks have been proposed for forecasting air pollution. However, feasibility these complex approaches has seldom verified developing countries, which generally suffer from heavy To forecast PM2.5 concentrations over different time intervals, we implemented three approaches: additive regression trees (MART), a deep feedforward neural network (DFNN) new hybrid model based on long...

Mesh surface denoising is a fundamental problem in geometry processing. The main challenge to remove noise while preserving sharp features (such as edges and corners) preventing generating false edges. We propose this paper combine total variation (TV) piecewise constant function space for variational mesh denoising. first give definitions of spaces associated operators. A method will then be presented by combining TV space. It proved that, the solution (the key part method) some sense...

The problem of decomposing a 3D mesh into meaningful segments (or parts) is great practical importance in computer graphics. This article presents variational decomposition algorithm that can efficiently partition prescribed number segments. extends the Mumford-Shah model to meshes contains data term measuring variation within segment using eigenvectors dual Laplacian matrix whose weights are related dihedral angle between adjacent triangles and regularization length boundary Such...

We propose a new variant of the Mumford-Shah model for simultaneous bias correction and segmentation images with intensity inhomogeneity. First, based on inhomogeneity, we introduce an L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> gradient regularizer to true smooth field. In addition, derive data fidelity using local properties allow field be influenced by its neighborhood. Second, use two-stage method, where fast alternating...

We consider to design a new efficient and easy-to-implement algorithm solve general group sparse optimization model with class of non-convex non-Lipschitz regularizations, named as fast iterative thresholding support-and-scale shrinking (FITS3). In this paper we focus on the case least-squares fidelity. FITS3 is designed from lower bound theory such models by integrating operation, linearization extrapolation techniques. The has two advantages. Firstly, it quite especially suitable for...

To analyze the results and current status of hospital infection monitoring quality control work at a tertiary specialized cancer hospital, providing references for further improvement HAI quality. Combined information-based with manual measures surveillance, collecting data such as report cards missed reports to monitoring. Starting from January 2023, comprehensive were taken, before after implementation compared evaluate effectiveness work. After management measures, incidence rate in...

Curvature flow (planar geometric heat flow) has been extensively applied to image processing, computer vision, and material science. To extend the numerical schemes algorithms of this on surfaces is very significant for corresponding motions curves images defined surfaces. In work, we are interested in geodesic curvature over triangulated using a level set formulation. First, present equation general smooth manifolds based an energy minimization curves. The then discretized by semi-implicit...

Brain extraction is an important preprocessing step for further analysis of brain MR images. Significant intensity inhomogeneity can be observed in rodent images due to the high-field MRI technique. Unlike most existing methods that require bias corrected MRI, we present a high-order and L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> regularized variational model correction extraction. The composed data fitting term, piecewise...

Many applications show that nonconvex nonsmooth regularization has advantages for restoring images with neat edges. This phenomenon been provided as a mathematical explanation the anisotropic model through establishing uniform lower bound nonzero gradients of recovered images. For and using potential functions composed $\ell_2$ norm image gradient, i.e., isotropic model, there lack reasonable theoretical results in literature. Based on some properties gradient fields not considered...

Impulse noise is a special type of which corrupts portion the image pixels while keeping others unaffected. L$_1$TV, L$_1$Nonconvex, and NonconvexTV are three common variational models for impulse removal. However, non-Lipschitz cases were ruled out L$_1$Nonconvex in former works. In this paper, we consider L$_1$non-Lipschitz non-LipschitzTV. Uniform lower bound theories these two obtained, yield models' characteristics. Based on theories, propose new algorithms models. Instead introducing...

Abstract Peripherally inserted central venous catheter (PICC) is the main access for cancer patients when they receive chemotherapy and nutritional support, but PICC-related thrombosis has become one of most common serious complications. It very important to further explore relationship among these features, so that prevent treat thrombosis. To investigate clinical features related factors upper extremity asymptomatic in patients, provide theoretical basis prevention A total 127 tumor with...

Abstract This paper considers the problem of interactively finding cutting contour to extract components from a given mesh. Some existing methods support cuts arbitrary shape but require careful and tedious input user. Others need little user however they are sensitive postprocessing step smooth generated jaggy contours. The popular geometric snake can be used optimize contour, it cannot deal with topology change. In this paper, we propose geodesic curvature flow based framework overcome all...

Abstract. This paper presents a framework of introducing GIS technology to record and analyse cultural heritages in continuous spatial scales. The research team is developing systematic approach support heritage conservation practice on historical buildings, courtyards, towns, archaeological sites ad landscapes. These studies are conducted not only from the property or site scales, but also investigated their contexts setting as well regional From these continues authenticity integrity can...

The Mumford-Shah model is an important tool for image labeling and segmentation, which pursues a piecewise smooth approximation of the original boundaries with shortest length. In contrast to previous efforts, use total variation regularization measure length boundaries, we build up novel by utilizing non-convex ℓ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">p</sup> regularity term p ∈ (0,1), can well preserve sharp edges eliminate geometric...