Wuhan University

About: Wuhan University is a education organization based out in Wuhan, China. It is known for research contribution in the topics: Computer science & Population. The organization has 92849 authors who have published 92882 publications receiving 1691049 citations. The organization is also known as: WHU & Wuhan College.

DeepCrack: Learning Hierarchical Convolutional Features for Crack Detection

Abstract: Cracks are typical line structures that are of interest in many computer-vision applications. In practice, many cracks, e.g., pavement cracks, show poor continuity and low contrast, which bring great challenges to image-based crack detection by using low-level features. In this paper, we propose DeepCrack-an end-to-end trainable deep convolutional neural network for automatic crack detection by learning high-level features for crack representation. In this method, multi-scale deep convolutional features learned at hierarchical convolutional stages are fused together to capture the line structures. More detailed representations are made in larger scale feature maps and more holistic representations are made in smaller scale feature maps. We build DeepCrack net on the encoder–decoder architecture of SegNet and pairwisely fuse the convolutional features generated in the encoder network and in the decoder network at the same scale. We train DeepCrack net on one crack dataset and evaluate it on three others. The experimental results demonstrate that DeepCrack achieves $F$ -measure over 0.87 on the three challenging datasets in average and outperforms the current state-of-the-art methods.

High‐Performance Alkaline Polymer Electrolyte for Fuel Cell Applications

Abstract: Although the proton exchange membrane fuel cell (PEMFC) has made great progress in recent decades, its commercialization has been hindered by a number of factors, among which is the total dependence on Pt-based catalysts. Alkaline polymer electrolyte fuel cells (APEFCs) have been increasingly recognized as a solution to overcome the dependence on noble metal catalysts. In principle, APEFCs combine the advantages of and alkaline fuel cell (AFC) and a PEMFC: there is no need for noble metal catalysts and they are free of carbonate precipitates that would break the waterproofing in the AFC cathode. However, the performance of most alkaline polyelectrolytes can still not fulfill the requirement offuel cell operations. In the present work, detailed information about the synthesis and physicochemical properties of the quaternary ammonia polysulfone (QAPS), a high-performance alkaline polymer electrolyte that has been successfully applied in the authors' previous work to demonstrate an APEFC completely free from noble metal catalysts (S. Lu, J. Pan, A. Huang, L. Zhuang, J. Lu, Proc. Natl. Acad. Sci. USA 2008, 105, 20611), is reported. Monitored by NMR analysis, the synthetic process of QAPS is seen to be simple and efficient. The chemical and thermal stability, as well as the mechanical strength of the synthetic QAPS membrane, are outstanding in comparison to commercial anion-exchange membranes. The ionic conductivity of QAPS at room temperature is measured to be on the order of 10 -2 S cm -1 Such good mechanical and conducting performances can be attributed to the superior microstructure of the polyelectrolyte, which features interconnected ionic channels in tens of nanometers diameter, as revealed by HRTEM observations. The electrochemical behavior at the Pt/ QAPS interface reveals the strong alkaline nature of this polyelectrolyte, and the preliminary fuel cell test verifies the feasibility of QAPS for fuel cell applications.

Collagen as a double-edged sword in tumor progression

Abstract: It has been recognized that cancer is not merely a disease of tumor cells, but a disease of imbalance, in which stromal cells and tumor microenvironment play crucial roles. Extracellular matrix (ECM) as the most abundant component in tumor microenvironment can regulate tumor cell behaviors and tissue tension homeostasis. Collagen constitutes the scaffold of tumor microenvironment and affects tumor microenvironment such that it regulates ECM remodeling by collagen degradation and re-deposition, and promotes tumor infiltration, angiogenesis, invasion and migration. While collagen was traditionally regarded as a passive barrier to resist tumor cells, it is now evident that collagen is also actively involved in promoting tumor progression. Collagen changes in tumor microenvironment release biomechanical signals, which are sensed by both tumor cells and stromal cells, trigger a cascade of biological events. In this work, we discuss how collagen can be a double-edged sword in tumor progression, both inhibiting and promoting tumor progression at different stages of cancer development.

New insight into the enhanced photocatalytic activity of N-, C- and S-doped ZnO photocatalysts

Abstract: In general, N-, C- and S-doped ZnO exhibit much higher phototcatalytic activity than the pure ZnO. However, the essential factors and underlying mechanism regarding the enhancement of photocatalytic activity are still unclear. In this work, the electronic structures, optical properties and effective masses of charge carriers are investigated by first-principle density functional theory calculation. Due to the nature of p-type doping, N and C doping can generate vacant states above the Fermi level and shift the conduction band into lower energy region, resulting in narrowing of band gap. Thus, N- and C-doped ZnO demonstrate much stronger light absorption in both visible and ultraviolet region. In contrast, because of the absence of vacant states, only limited enhancement of light absorption is observed for S-doped ZnO whose improved photocatalytic performance can only be attributed to the direct reduction of band gap. The calculation of the effective masses show that ZnO typically possess light electrons and heavy holes, confirming its intrinsic character of n-type semiconductor, while N, C and S doping can generally render electrons lighter and holes heavier, resulting in slower recombination rate of photogenerated electron–hole pairs. Noticeably, C doping can discourage such recombination to the greatest extent and separate electron–hole pairs most efficiently compared with N and S doping, serving as a potentially promising pathway to increase the quantum efficiency of ZnO-based photocatalysts. This work will provide some new insights into the understanding of doping effect over the enhancement of photocatalytic activity of N-, C- and S-doped ZnO.

DDcGAN: A Dual-Discriminator Conditional Generative Adversarial Network for Multi-Resolution Image Fusion

Abstract: In this paper, we proposed a new end-to-end model, termed as dual-discriminator conditional generative adversarial network (DDcGAN), for fusing infrared and visible images of different resolutions. Our method establishes an adversarial game between a generator and two discriminators. The generator aims to generate a real-like fused image based on a specifically designed content loss to fool the two discriminators, while the two discriminators aim to distinguish the structure differences between the fused image and two source images, respectively, in addition to the content loss. Consequently, the fused image is forced to simultaneously keep the thermal radiation in the infrared image and the texture details in the visible image. Moreover, to fuse source images of different resolutions, e.g. , a low-resolution infrared image and a high-resolution visible image, our DDcGAN constrains the downsampled fused image to have similar property with the infrared image. This can avoid causing thermal radiation information blurring or visible texture detail loss, which typically happens in traditional methods. In addition, we also apply our DDcGAN to fusing multi-modality medical images of different resolutions, e.g. , a low-resolution positron emission tomography image and a high-resolution magnetic resonance image. The qualitative and quantitative experiments on publicly available datasets demonstrate the superiority of our DDcGAN over the state-of-the-art, in terms of both visual effect and quantitative metrics. Our code is publicly available at https://github.com/jiayi-ma/DDcGAN .