M3Net: A Multi-scale Multi-view Framework for Multi-phase Pancreas Segmentation Based on Cross-phase Non-local Attention

Recently, deep learning-based techniques have shown great power in image inpainting especially dealing with squared holes. However, they fail to generate plausible results inside the missing regions for irregular and large holes as there is a lack of understanding between missing regions and existing counterparts. To overcome this limitation, we combine two non-local mechanisms including a contextual attention module (CAM) and an implicit diversified Markov random fields (ID-MRF) loss with a multi-scale architecture which uses several dense fusion blocks (DFB) based on the dense combination of dilated convolution to guide the generative network to restore discontinuous and continuous large masked areas. To prevent color discrepancies and grid-like artifacts, we apply the ID-MRF loss to improve the visual appearance by comparing similarities of long-distance feature patches. To further capture the long-term relationship of different regions in large missing regions, we introduce the CAM. Although CAM has the ability to create plausible results via reconstructing refined features, it depends on initial predicted results. Hence, we employ the DFB to obtain larger and more effective receptive fields, which benefits to predict more precise and fine-grained information for CAM. Extensive experiments on two widely-used datasets demonstrate that our proposed framework significantly outperforms the state-of-the-art approaches both in quantity and quality.

Download Full-text

Single image super resolution based on multi-scale structure and non-local smoothing

EURASIP Journal on Image and Video Processing ◽

10.1186/s13640-021-00552-8 ◽

2021 ◽

Vol 2021 (1) ◽

Author(s):

Wenyi Wang ◽

Jun Hu ◽

Xiaohong Liu ◽

Jiying Zhao ◽

Jianwen Chen

Keyword(s):

Super Resolution ◽

Principal Component ◽

Feature Mapping ◽

Multi Scale ◽

Non Local ◽

Image Super Resolution ◽

Gaussian Blur ◽

Single Image Super Resolution ◽

Global And Local ◽

Bicubic Interpolation

AbstractIn this paper, we propose a hybrid super-resolution method by combining global and local dictionary training in the sparse domain. In order to present and differentiate the feature mapping in different scales, a global dictionary set is trained in multiple structure scales, and a non-linear function is used to choose the appropriate dictionary to initially reconstruct the HR image. In addition, we introduce the Gaussian blur to the LR images to eliminate a widely used but inappropriate assumption that the low resolution (LR) images are generated by bicubic interpolation from high-resolution (HR) images. In order to deal with Gaussian blur, a local dictionary is generated and iteratively updated by K-means principal component analysis (K-PCA) and gradient decent (GD) to model the blur effect during the down-sampling. Compared with the state-of-the-art SR algorithms, the experimental results reveal that the proposed method can produce sharper boundaries and suppress undesired artifacts with the present of Gaussian blur. It implies that our method could be more effect in real applications and that the HR-LR mapping relation is more complicated than bicubic interpolation.

Download Full-text

Temperature Resistance of Mo3Si: Phase Stability, Microhardness, and Creep Properties

Metals ◽

10.3390/met11040564 ◽

2021 ◽

Vol 11 (4) ◽

pp. 564 ◽

Cited By ~ 1

Author(s):

Olha Kauss ◽

Susanne Obert ◽

Iurii Bogomol ◽

Thomas Wablat ◽

Nils Siemensmeyer ◽

...

Keyword(s):

Power Law ◽

Phase Stability ◽

Gas Turbines ◽

Constitutive Models ◽

Creep Properties ◽

Multi Scale ◽

Scale Modeling ◽

Significant Difference ◽

Before And After ◽

Multi Phase

Mo-Si-B alloys are one of the most promising candidates to substitute Ni based superalloys in gas turbines. The optimization of their composition can be achieved more effectively using multi-scale modeling of materials behavior and structural analysis of components for understanding, predicting, and screening properties of new alloys. Nevertheless, this approach is dependent on data on the properties of the single phases in these alloys. The focus of this investigation is Mo3Si, one of the phases in typical Mo-Si-B alloys. The effect of 100 h annealing at 1600 °C on phase stability and microhardness of its three near-stoichiometric compositions—Mo-23Si, Mo-24Si and Mo-25Si (at %)—is reported. While Mo-23Si specimen consist only of Mo3Si before and after annealing, Mo-24Si and Mo-25Si comprise Mo5Si3 and Mo3Si before annealing. The latter is then increased by the annealing. No significant difference in microhardness was detected between the different compositions as well as after annealing. The creep properties of Mo3Si are described at 1093 °C and 1300 °C at varying stress levels as well as at 300 MPa and temperatures between 1050 °C and 1350 °C. Three constitutive models were used for regression of experimental results—(i) power law with a constant creep exponent, (ii) stress range dependent law, and (iii) power law with a temperature-dependent creep exponent. It is confirmed that Mo3Si has a higher creep resistance than Moss and multi-phase Mo-Si-B alloys, but a lower creep strength as compared to Mo5SiB2.

Download Full-text

Introduction to special issue on advances in multi-phase fluid and multi-scale particulate reactive flow modeling and simulation for petroleum engineering applications

SIMULATION ◽

10.1177/0037549720970383 ◽

2020 ◽

pp. 003754972097038

Author(s):

Bin Yuan

Keyword(s):

Modeling And Simulation ◽

Petroleum Engineering ◽

Reactive Flow ◽

Flow Modeling ◽

Special Issue ◽

Engineering Applications ◽

Multi Scale ◽

Multi Phase ◽

Phase Fluid

Download Full-text

Multi-scale non-local denoising method in neuroimaging

Journal of X-Ray Science and Technology ◽

10.3233/xst-160564 ◽

2016 ◽

Vol 24 (3) ◽

pp. 477-487

Author(s):

Yiping Chen ◽

Cheng Wang ◽

Liansheng Wang

Keyword(s):

Denoising Method ◽

Multi Scale ◽

Non Local

Download Full-text

The obstructed jet in Mrk 231

Monthly Notices of the Royal Astronomical Society ◽

10.1093/mnras/stab587 ◽

2021 ◽

Author(s):

Ailing Wang ◽

Tao An ◽

Sumit Jaiswal ◽

Prashanth Mohan ◽

Yuchan Wang ◽

...

Keyword(s):

Large Scale ◽

Position Angle ◽

Opening Angle ◽

Luminous Infrared Galaxies ◽

Multi Scale ◽

Long Baseline ◽

Relativistic Jet ◽

Multi Phase ◽

Jet Structure ◽

Projected Distance

Abstract Mrk 231 is the closest radio-quiet quasar known and one of the most luminous infrared galaxies in the local Universe. It is characterised by the co-existence of a radio jet and powerful multi-phase multi-scale outflows, making it an ideal laboratory to study active galactic nucleus (AGN) feedback. We analyse the multi-epoch very long baseline interferometry data of Mrk 231 and estimate the jet head advance speed to be ≲ 0.013 c, suggesting a sub-relativistic jet flow. The jet position angle changes from −113○ in the inner parsec to −172○ at a projected distance of 25 parsec. The jet structure change might result from either a jet bending following the rotation of the circum-nuclear disc or the projection of a helical jet on the plane of the sky. In the large opening angle (∼60○) cone, the curved jet interacts with the interstellar medium and creates wide-aperture-angle shocks which subsequently dissipate a large portion of the jet power through radiation and contribute to powering the large-scale outflows. The low power and bent structure of the Mrk 231 jet, as well as extensive radiation dissipation, are consistent with the obstruction of the short-length jet by the host galaxy’s environment.

Download Full-text