BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Guangcun Shan

Guangcun Shan contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computer Visioncond-mat.mtrl-sciphysics.app-pheess.IVphysics.ins-detcond-mat.dis-nncond-mat.mes-hallcond-mat.str-elDistributed, Parallel, and Cluster ComputingMachine LearningNeural and Evolutionary Computingphysics.chem-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
10works
0followers
12topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

10 published item(s)

preprint2022arXiv

An Efficient Target Detection and Recognition Method in Aerial Remote-sensing Images Based on Multiangle Regions-of-Interest

Recently, deep learning technology have been extensively used in the field of image recognition. However, its main application is the recognition and detection of ordinary pictures and common scenes. It is challenging to effectively and expediently analyze remote-sensing images obtained by the image acquisition systems on unmanned aerial vehicles (UAVs), which includes the identification of the target and calculation of its position. Aerial remote sensing images have different shooting angles and methods compared with ordinary pictures or images, which makes remote-sensing images play an irreplaceable role in some areas. In this study, a new target detection and recognition method in remote-sensing images is proposed based on deep convolution neural network (CNN) for the provision of multilevel information of images in combination with a region proposal network used to generate multiangle regions-of-interest. The proposed method generated results that were much more accurate and precise than those obtained with traditional ways. This demonstrated that the model proposed herein displays tremendous applicability potential in remote-sensing image recognition.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Prediction of novel two-dimensional rare-earth material with room-temperature ferromagnetism and large perpendicular magnetic anisotropy

Novel 2D ferromagnets with high Curie temperature and large perpendicular magnetic anisotropy are especially attractive owing to the future promising application in modern spintronics, but meanwhile the 2D ferromagnetic materials with high Curie temperature and large perpendicular magnetic anisotropy are rarely reported. Based on density functional theory (DFT) calculations, we predict a new kind of 2D ferromagnetic materials - GdB2N2, which possesses large magnetic moment, high Curie temperature (335 K) and large perpendicular magnetic anisotropy (10.38 meV/f.u.). Biaxial strain ranging from -0.5% to 5% and different concentrations of charge-carrier doping are applied to reveal the influence on the Curie temperature and magnetic anisotropy energy (MAE). Besides, magnetic coupling process within GdB2N2 is found to be via a Ruderman-Kittel-Kasuya-Yosida (RKKY) mechanism. In summary, our work here predicts a novel 2D rare-earth material GdB2N2, which not only enriches the category of 2D room-temperature ferromagnets, but also proposes a new possibility of combining traditional 2D materials and rare-earth materials to achieve more intriguing magnetic properties, finally it carves out the path for the next-generation spintronic devices and sensors.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Machine learning prediction of magnetic properties of Fe-based metallic glasses considering local structures

Magnetism prediction is of great significance for Fe-based metallic glasses (FeMGs), which have shown great commercial value. Theories or models established based on condensed matter physics exhibit several exceptions and limited accuracy. In this work, machine learning (ML) models learned from a large amount of experimental data were trained based on eXtreme gradient boosting (XGBoost), artificial neural networks (ANN), and random forest to predict the magnetic properties of FeMGs. The XGBoost and ANN models exhibited comparably excellent predictive performance, with R^2 >= 0.903, mean absolute percentage error (MAPE) <= 6.17, and root mean squared error (RMSE) <= 0.098. The trained ML models aggregate the influence of 13 factors, which is difficult to achieve in traditional physical models. The influence of local structure, which was represented by the experimental parameter of the supercooled liquid region, presented a significant impact on the predictive performance of ML models. The developed ML-based method here can predict the magnetic properties of FeMGs by considering multiple factors simultaneously, including complex local structures.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A highly sensitive piezoresistive sensor based on MXene and polyvinyl butyral with a wide detection limit and low power consumption

As a new class of two-dimensional transition-metal carbide and carbonitride, MXene have been widely used in the energy storage, sensor, catalysis, electromagnetic interference shielding and other field. It is a challenge to simultaneously realize a sensor of extremely high sensitivity, wide detection limits, low power consumption and good mechanical stability. In this work, taking advantage of high conductivity of MXene and porous structure of polyvinyl butyral, a highly sensitive piezoresistive sensor was fabricated. The fabricated MXene/PVB-based sensor exhibits highly sensitive reliably with a factor of ~11.9 kPa^-1, ~1.15 kPa^-1 and ~0.20 kPa^-1 in the ranges of 31.2 Pa-312 Pa, 312 Pa- 62.4 kPa and 62.4 kPa-1248.4 kPa, respectively. The sensor has a wide detection range (~31.2 Pa to ~2.205 MPa), low detection limit (6.8 Pa), low detection voltage (0.1 mV), low power consumption (~3.6 * 10^-10 W), fast response time ( ~110 ms), as well as good mechanical stability (over 10,000 maximum-pressure cycles). Moreover, it is demonstrated that the sensor can detect subtle bending and release activities of human, including arterial pulses and voice signal, which is potentially suitable as a wide detection range, highly sensitive and low power consumption piezoresistive sensor. This work provides a new avenue to expand the application of MXene-based flexible pressure sensor in the field of wide sensing range and ultra-low power consumption.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Accelerating temporal action proposal generation via high performance computing

Temporal action recognition always depends on temporal action proposal generation to hypothesize actions and algorithms usually need to process very long video sequences and output the starting and ending times of each potential action in each video suffering from high computation cost. To address this, based on boundary sensitive network we propose a new temporal convolution network called Multipath Temporal ConvNet (MTN), which consists of two parts i.e. Multipath DenseNet and SE-ConvNet. In this work, one novel high performance ring parallel architecture based on Message Passing Interface (MPI) is further introduced into temporal action proposal generation, which is a reliable communication protocol, in order to respond to the requirements of large memory occupation and a large number of videos. Remarkably, the total data transmission is reduced by adding a connection between multiple computing load in the newly developed architecture. It is found that, compared to the traditional Parameter Server architecture, our parallel architecture has higher efficiency on temporal action detection task with multiple GPUs, which is suitable for dealing with the tasks of temporal action proposal generation, especially for large datasets of millions of videos. We conduct experiments on ActivityNet-1.3 and THUMOS14, where our method outperforms other state-of-art temporal action detection methods with high recall and high temporal precision. In addition, a time metric is further proposed here to evaluate the speed performance in the distributed training process.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Hydrogen Bond Interaction Promotes Flash Energy Transport at MXene-water Interface

There are emerging applications for photothermal conversion utilizing MXene, but the mechanism under these applications related interfacial energy migration from MXene to the attached surface layer is still unknown. In this paper, with comprehensive ultrafast studies, we reported the energy migration pathway from MXene (Ti3C2Tx) to local environment under plasmonic excitation. Our data found that in water, energy dissipation is divided into fast hydrogen bond mediated channel and slow lattice motion mediated channel.The experimental results suggest that in water, nearly 80% energy in MXene that gained from the photoexcitation quickly dissipates into surrounding water molecules within 7 ps as a hydrogen bond mediated fast channel, and the remaining energy vanishes with time constant ~100 ps as a lattice motion mediated slow channel. The fast energy migration would result in the prominent interfacial energy conductance 150-300 MW *m{-2}* K{-1} for MXene-water interface. Tuning the solvent into ethanol could both narrow the energy dissipation to 35% through the fast channel and slow down the thermal channel (400 ps). To gain the molecular insight, molecular dynamic results presented different solvents had significantly different H bond forming ability on MXene surface. Our results suggested that interfacial interaction is crucial for effective hydrogen bonds on MXene surface to channel the excitation dissipation, providing important insights into the photothermal applications with MXene.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Recognizing Handwritten Mathematical Expressions as LaTex Sequences Using a Multiscale Robust Neural Network

In this paper, a robust multiscale neural network is proposed to recognize handwritten mathematical expressions and output LaTeX sequences, which can effectively and correctly focus on where each step of output should be concerned and has a positive effect on analyzing the two-dimensional structure of handwritten mathematical expressions and identifying different mathematical symbols in a long expression. With the addition of visualization, the model&#39;s recognition process is shown in detail. In addition, our model achieved 49.459% and 46.062% ExpRate on the public CROHME 2014 and CROHME 2016 datasets. The present model results suggest that the state-of-the-art model has better robustness, fewer errors, and higher accuracy.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Robust Encoder-Decoder Learning Framework towards Offline Handwritten Mathematical Expression Recognition Based on Multi-Scale Deep Neural Network

Offline handwritten mathematical expression recognition is a challenging task, because handwritten mathematical expressions mainly have two problems in the process of recognition. On one hand, it is how to correctly recognize different mathematical symbols. On the other hand, it is how to correctly recognize the two-dimensional structure existing in mathematical expressions. Inspired by recent work in deep learning, a new neural network model that combines a Multi-Scale convolutional neural network (CNN) with an Attention recurrent neural network (RNN) is proposed to identify two-dimensional handwritten mathematical expressions as one-dimensional LaTeX sequences. As a result, the model proposed in the present work has achieved a WER error of 25.715% and ExpRate of 28.216%.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Understanding high ordering temperature in Gd$_6$FeBi$_2$ magnet: critical behavior, electronic structure and crystal-field analysis

Gd6FeBi2 is reported as the only one room-temperature magnet with a Curie temperature (Tc) of ca. 350 K among more than hundreds of compounds with its structural type, which makes it more attractive in potential applications. To reveal the origin of such high ordering temperature, critical behaviors, electronic structure and crystal-field effects of Gd6FeBi2 are investigated in this work. The short-range Gd-Fe ferrimagnetic interaction is supported by the non-Curie-Weiss paramagnetic behavior, crystal and electronic structure analyses, in agreement with previous DFT calculations. Unlike the strong TM-TM exchange interactions, the Gd-Fe exchange interaction shows limited influence on the critical exponents determined by long-range exchange interactions, which seems a common feature in RE-TM based alloys without TM-TM exchange interactions. However, the strong Gd-Fe hybridization reduces the influence of vibronic couplings on the short-range exchange interaction and thus allows a high Tc. The broadening or splitting mechanism of Gd 4f-electron bands is addressed based on crystal-field analysis and likely another factor for elevated Tc in Gd6FeBi2 and Gd-based compounds with non-magnetic elements. Different magnetic behaviors among isostructural compounds, and the relationship between the band splitting and crystal-field effects is also discussed.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Two-Dimensional Scanning Phased Array Based on Pattern Reconfigurable Antenna

This paper presents a novel planar phased array that can work in Ka band designed for two-dimensional scanning. A four-modes pattern reconfigurable element based on microstrip Yagi antenna is proposed, and then developed into the 4X4 phased array, in which two feed points are introduced to reduce sidelobe level (SLL). The simulated results show that our proposed phased array is able to cover the scanning range from -60° to +60°in H-plane(yoz-plane) and -47.2°to +47.2°in E-plane(xoz-plane) with the gain reduction less than 3dB, both the SLLs of which are 10dB less than main beam. The two-dimensional scanning phased array based on pattern reconfigurable antenna (PRA) presented here is suitable for the satellite communication.

0 citations0 reviewsNo code linkedOpen access