AED-Net:An Abnormal Event Detection Network

It has long been a challenging task to detect an anomaly in a crowded scene.In this paper,a selfsupervised framework called the abnormal event detection network(AED-Net),which is composed of a principal component analysis network(PCAnet)and kernel principal component analysis(kPCA),is proposed to address this problem.Using surveillance video sequences of different scenes as raw data,the PCAnet is trained to extract high-level semantics of the crowd’s situation.Next,kPCA,a one-class classifier,is trained to identify anomalies within the scene.In contrast to some prevailing deep learning methods,this framework is completely self-supervised because it utilizes only video sequences of a normal situation.Experiments in global and local abnormal event detection are carried out on Monitoring Human Activity dataset from University of Minnesota(UMN dataset)and Anomaly Detection dataset from University of California,San Diego(UCSD dataset),and competitive results that yield a better equal error rate(EER)and area under curve(AUC)than other state-of-the-art methods are observed.Furthermore,by adding a local response normalization(LRN)layer,we propose an improvement to the original AED-Net.The results demonstrate that this proposed version performs better by promoting the framework’s generalization capacity.

基于MXene和多孔PVB的高灵敏压阻柔性传感器可用于健康监测

据世界卫生组织统计数据显示,心血管疾病已成为威胁人体健康的"第一杀手".全球5690万例死亡人群中,约有1790万人死于心血管疾病,占全球死亡总数的31%,其中,超过80%死于心脏病和中风.我国也是心血管疾病大国,心血管病患人数已达到2.9亿.面对老年心血管疾病的威胁,人们从最初的"治已病"转向聚焦"治未病".

Graphene and other two-dimensional materials

Over the last few years,research on graphene [1-3]has progressed significantly,and as a result,a number of reallife applications of graphene have been realized [4,5].This carbon allotrope (for a review on other forms of carbon, see [6])resulted in a number of novel physical phenomena already discovered (ranging from new types of quantum Hall effect to universal optical conductivity)and dramatically expanded the range of possible applications for such materials (from transparent conductive coating to ultrafast photodetectors).

A review of oxygen reduction mechanisms for metal-free carbon-based electrocatalysts

The sluggish kinetics of Oxygen Reduction Reaction(ORR)at the cathode in proton exchange membrane fuel cells or metal-air batteries requires highly effective and stable electrocatalysts to boost the reaction.The low abundance and high price of Pt-based electrocatalysts hamper the widespread application of proton exchange membrane fuel cells and metal-air batteries.As promising alternatives,metal-free carbon materials,especially upon doping heteroatoms or creating defects demonstrated excellent ORR activity,which is as efficient as or even superior to commercial platinum on carbon.Significant progress on the development of advanced carbon materials as highly stable and durable catalysts has been achieved,but the catalytic mechanisms of these materials still remain undistinguished.In present review,we summarized the up-to-date progress in the studies of carbon materials,and emphasized on the combination of experiment and theory to clarify the underlying mechanisms of these materials.At last,we proposed the perspectives on the proper strategies of elucidating the mechanisms of carbon materials as electrocatalysts towards ORR.

Two-dimensional materials： Emerging toolkit for construction of ultrathin high-efficiency microwave shield and absorber

Two-dimensional （2D） materials generally have unusual confined electro-strong interaction in a plane and can physical and chemical properties owing to the exhibit obvious anisotropy and a significant quantum-confinement effect, thus showing great promise in many fields. Some 2D materials, such as graphene and MXenes, have recently exhibited extraordinary electromagnetic-wave shielding and absorbing performance, which is attributed to their special electrical behavior, large specific surface area, and low mass density. Compared with traditional microwave attenuating materials, 2D materials have several obvious inherent advantages. First, similar to other nanomaterials, 2D materials have a very large specific surface area and can provide numerous interfaces for the enhanced interfacial polarization as well as the reflection and scattering of electromagnetic waves. Second, 2D materials have a particular 2D morphology with ultrasmall thickness, which is not only beneficial for the penetration and dissipation of electromagnetic waves through the 2D nanosheets, giving rise to multiple reflections and the dissipation of electromagnetic energy, but is also conducive to the design and fabrication of various well-defined structures, such as layer-by-layer assemblies, core-shell particles, and porous foam, for broadband attenuation of electromagnetic waves. Third, owing to their good processability, 2D materials can be integrated into various multifunctional composites for multimode attenuation of electromagnetic energy. In addition to behaving as microwave reflectors and absorbers, 2D materials can act as impedance regulators and provide structural support for good impedance matching and setup of the optimal structure. Numerous studies indicate that 2D materials are among the most promising microwave attenuation materials. In view of the rapid development and enormous advancement of 2D materials in shielding and absorbing electromagnetic wave, there is a strong need to summarize the recent research results in this field for presenting a comprehensive view and providing helpful suggestions for future development.

Machine learning prediction of magnetic properties of Fe-based metallic glasses considering glass forming ability

Fe-based metallic glasses(MGs)have shown great commercial values due to their excellent soft magnetic properties.Magnetism prediction with consideration of glass forming ability(GFA)is of great signifi-cance for developing novel functional Fe-based MGs.However,theories or models established based on condensed matter physics exhibit limited accuracy and some exceptions.In this work,based on 618 Fe-based MGs samples collected from published works,machine learning(ML)models were well trained to predict saturated magnetization(B_(s))of Fe-based MGs.GFA was treated as a feature using the experimental data of the supercooled liquid region(△T_(x)).Three ML algorithms,namely eXtreme gradient boosting(XGBoost),artificial neural networks(ANN)and random forest(RF),were studied.Through feature selection and hyperparameter tuning,XGBoost showed the best predictive performance on the randomly split test dataset with determination coefficient(R^(2))of 0.942,mean absolute percent error(MAPE)of 5.563%,and root mean squared error(RMSE)of 0.078 T.A variety of feature importance rankings derived by XGBoost models showed that T_(x) played an important role in the predictive performance of the models.This work showed the proposed ML method can simultaneously aggregate GFA and other features in ther-modynamics,kinetics and structures to predict the magnetic properties of Fe-based MGs with excellent accuracy.

A new strategy for the fabrication of a flexible and highly sensitive capacitive pressure sensor

The development of flexible capacitive pressure sensors has wide application prospects in the fields of electronic skin and intelligent wearable electronic devices,but it is still a great challenge to fabricate capacitive sensors with high sensitivity.Few reports have considered the use of interdigital electrode structures to improve the sensitivity of capacitive pressure sensors.In this work,a new strategy for the fabrication of a high-performance capacitive flexible pressure sensor based on MXene/polyvinylpyrrolidone(PVP)by an interdigital electrode is reported.By increasing the number of interdigital electrodes and selecting the appropriate dielectric layer,the sensitivity of the capacitive sensor can be improved.The capacitive sensor based on MXene/PVP here has a high sensitivity(~1.25 kPa^(−1)),low detection limit(~0.6 Pa),wide sensing range(up to 294 kPa),fast response and recovery times(~30/15 ms)and mechanical stability of 10000 cycles.The presented sensor here can be used for various pressure detection applications,such as finger pressing,wrist pulse measuring,breathing,swallowing and speech recognition.This work provides a new method of using interdigital electrodes to fabricate a highly sensitive capacitive sensor with very promising application prospects in flexible sensors and wearable electronics.

Abnormal event detection via the analysis of multi-frame optical flow information

Security surveillance of public scene is closely relevant to routine safety of individual.Under the stimulus of this concern,abnormal event detection is becoming one of the most important tasks in computer vision and video processing.In this paper,we propose a new algorithm to address the visual abnormal detection problem.Our algorithm decouples the problem into a feature descriptor extraction process,followed by an AutoEncoder based network called cascade deep AutoEncoder(CDA).The movement information is represented by a novel descriptor capturing the multi-frame optical flow information.And then,the feature descriptor of the normal samples is fed into the CDA network for training.Finally,the abnormal samples are distinguished by the reconstruction error of the CDA in the testing procedure.We validate the proposed method on several video surveillance datasets.

New topological semimetai Candidate of nonsymmorphic PdSb2 with unique six-fold degenerate point

Since the discovery of Majorana fermions in condensed matter systems, new quasiparticle predictions of novel fermions have been predicted in solid state systems which exhibit three, six or eight fold degenerate band crossings protected by crystal symmetty in presence of spin orbit coupling and time reversal symmetry [1]. The nontrivial topology in condensed matter systems results from the crossings of conduction and valence bands. And the cases of g = 3, 6, and 8 are of particularly interesting as they can only be found in condensed matter systems, having no high energy analogues as constrained by Poincare symmetry [2].

Bifunctional oxygen electrocatalysts for rechargeable zinc-air battery based on MXene and beyond

Oxygen electrocatalysts are of great importance for the air electrode in zinc–air batteries(ZABs).Owing to large surface area,high electrical conductivity and ease of modification,two-dimensional(2D)materials have been widely studied as oxygen electrocatalysts for the rechargable ZABs.The elaborately modified 2D materials-based electrocatalysts,usually exhibit excellent performance toward the oxygen reduction reaction(ORR)and oxygen evolution reaction(OER),which have attracted extensive interests of worldwide researchers.Given the rapid development of bifunctional electrocatalysts toward ORR and OER,the latest progress of non-noble electrocatalysts based on layered double hydroxides(LDHs),graphene,and MXenes are intensively reviewed.The discussion ranges from fundamental structure,synthesis,electrocatalytic performance of these catalysts,as well as their applications in the rechargeable ZABs.Finally,the challenges and outlook are provided for further advancing the commercialization of rechargeable ZABs.

Bioinspired mineral MXene hydrogels for tensile strain sensing and radionuclide adsorption applications

MXene-based hydrogels have drawn considerable attention as flexible and wearable sensors.However,the application of MXene-based hydrogels after sensing failure has rarely been investigated,which is of great significance for expanding their engineering application.In this work,multifunctional mineral MXene hydrogels(MMHs)were synthesized via a simple method inspired by biomineralization.The prepared MMHs were stretchable,self-healable and conductive,and can be used to fabricate wearable tensile strain sensors showing a super-wide sensing range with excellent sensitivity.MMHs-based strain sensors were designed to be directly attached to the skin surface to detect tiny and large human motions.In addition,with the advantages of a large specific area,excellent hydrophilicity and abundant active adsorption sites for MXene nanosheets and hydrogels,dehydrated MMHs were used as highly efficient adsorbents for the removal of strontium ions from aqueous solutions.This work shows the great potential of MXene in promoting the development of nextgeneration functional materials.

Two-dimensional MXenes and their applications

Two-dimensional(2D)transition metal carbides,nitrides or carbonitrides(MXene)have attracted tremendous attention as a potentially the largest family of inorganic materials.They have a general formula Mn+1XnTx,where M is an early transition metal(2 or more can be present in a random mix or a specific order),X is C and/or N elements(oxygen can partially substitute),and T_(x)(x is variable)represent surface terminations,which can include reactive end groups(such as–OH and=O),halogens and chalcogens,and n can vary from 1 to 4(fractions are possible).All MXenes can either be produced as multilayer particles(ml-MXene)or delaminated(d-MXene)single-layer flakes.

Flexoelectricity Driven Fano Resonance in Slotted Carbon Nanotubes for Decoupled Multifunctional Sensing

Multifunctionality,interference-free signal readout,and quantum effect are important considerations for flexible sensors equipped within a single unit towards further miniaturization.To address these criteria,we present the slotted carbon nanotube(CNT)junction features tunable Fano resonance driven by flexoelectricity,which could serve as an ideal multimodal sensory receptor.Based on extensive ab initio calculations,we find that the effective Fano factor can be used as a temperature-insensitive extrinsic variable for sensing the bending strain,and the Seebeck coefficient can be used as a strain-insensitive intrinsic variable for detecting temperature.Thus,this dual-parameter permits simultaneous sensing of temperature and strain without signal interference.We further demonstrate the applicability of this slotted junction to ultrasensitive chemical sensing which enables precise determination of donor-type,acceptor-type,and inert molecules.This is due to the enhancement or counterbalance between flexoelectric and chemical gating.Flexoelectric gating would preserve the electron–hole symmetry of the slotted junction whereas chemical gating would break it.As a proof-of-concept demonstration,the slotted CNT junction provides an excellent quantum platform for the development of multistimuli sensation in artificial intelligence at the molecular scale.

Nickel-metal-organic framework nanobelt based composite membranes for efficient Sr^(2+) removal from aqueous solution

The sorption removal of radionuclides Sr^(2+) using a freestanding functional membrane is an interesting and significant research area in the remediation of radioactive wastes.Herein,a novel self-assembled membrane consisting of metaleorganic framework(MOF)nanobelts and graphene oxides(GOs)are synthesized through a simple and facile filtration method.The membrane possesses a unique interwove morphology as evidenced from SEM images.Batch experiments suggest that the GO/Ni-MOF composite membrane could remove Sr^(2+) ions from aqueous solutions and the Sr^(2+) adsorption capacity and efficiency of the GO/Ni-MOF composite membrane is relevant to the MOF content in the composite.Thus,the dominant interaction mechanism was interface or surface complexation,electrostatic interaction as well as ion substitution.The maximum effective sorption of Sr^(2+) over GO/Ni-MOF membrane is 32.99% with 2 mg composite membrane containing a high content of Ni-MOF at 299 K in 100 mg/L Sr^(2+) aqueous solution.The FT-IR and XPS results suggest that the synergistic effect between GO and Ni-MOF is determinant in the sorption Sr^(2+) process.The GO/Ni-MOF composite membrane is demonstrated to have the advantages of efficient removal of Sr^(2+),low cost and simple synthesis route,which is promising in the elimination of radionuclide contamination.

AI-Enabled Wearable and Flexible Electronics for Assessing Full Personal Exposures

Research on the exposome has been extended to personal exposures,and full assessment of personal exposures is of great significance for personal health monitoring and epidemiological studies.Compared with static measurement instruments,wearable sensors are more suitable for dynamic personal exposures assessment.The development of flexible wearable sensors with the features of being physically comfortable and easy to use can be a promising solution for the measurement of personal exposures.With the support of big data and AI,large-scale personal exposures assessment could foster the transition from population-based to individual-based epidemiological studies and upgrade the intelligence level of medical services.

Accelerating temporal action proposal generation via high performance computing

Temporal action proposal generation aims to output the starting and ending times of each potential action for long videos and often suffers from high computation cost.To address the issue,we propose a new temporal convolution network called Multipath Temporal ConvNet(MTCN).In our work,one novel high performance ring parallel architecture based is further introduced into temporal action proposal generation 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.Compared to the traditional Parameter Server architecture,our parallel architecture has higher efficiency on temporal action detection tasks with multiple GPUs.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 performancein the distributed training process.

Guiding topological semimetals towards water oxidation in a Kagomé crystal lattice

Photocatalytic splitting of relatively plentiful water into O2 and H2 supplies a possible solution for storing solar energy to meet energy demands and environmental requirements[1].The water oxidation half reaction to form dioxygen known as the oxygen evolution reaction(OER,2H2O→O2+4H++4e-)is attracting more attention because there are already known materials for efficient mediation of the reduction step(4H++4e-→2H2)[2].Current research focuses on understanding the mechanisms of OER and development of new OER catalysts.The widely studied OER systems include platinum surface[1],transition metal oxides[2],transition metal complexes[3],3d transition metal spinels and perovskites[4],and metal-organic framework(MOF)materials based on3d transition metals[5].In these materials,transition metals in the surface generally serve as active centers for the OER,while the surrounding atomic environments determine their performance and stability[6].However,preparation of the surface has a large effect on reactivity(defects,kinks,low coordinate sites)therefore it is difficult to predict an OER material’s properties by its bulk structure.

Vertical-external-cavity surface-emitting lasers and quantum dot lasers

The use of cavity to manipulate photon emission of quantum dots （QDs） has been opening unprecedented opportunities for realizing quantum functional nanophotonic devices and quantum information devices. In particular, in the field of semiconductor lasers, QDs were introduced as a superior alternative to quantum wells （QWs） to suppress the temperature dependence of the threshold current in vertical-external-cavity surfaceemitting lasers （VECSELs）. In this work, a review of properties and development of semiconductor VECSEL devices and QD laser devices is given. Based on the features of VECSEL devices, the main emphasis is put on the recent development of technological approach on semiconductor QD VECSELs. Then, from the viewpoint of both single QD nanolaser and cavity quantum electro- dynamics （QED）, a single-QD-cavity system resulting from the strong coupling of QD cavity is presented. In this review, we will cover both fundamental aspects and technological approaches of QD VECSEL devices. Lastly, the presented review here has provided deep insight into useful guideline for the development of QD VECSEL technology, future quantum functional nanophotonic devices and monolithic photonic integrated circuits （MPhlCs）.

Flexoelectricity Driven Fano Resonance in Slotted Carbon Nanotubes for Decoupled Multifunctional Sensing

Mulifunctionality,interference fre signal readout,and quantum eft are important considerations for flexible sensors equipped within a single unit towards further miniaturization.To address these criteria,we present the slotted carbon nanotube(CNT)junction features lunable Fano resonance driven by flexoelectricity,which could serve as an ideal multimodal sensory receptor.Based on extensive ab initio calculations,we find that the efective Fano factor can be used as a temperature insensitive extrinsic variable for sensing the bending strain,and the Seebeck cefficient can be used as a strain-insensitive intrinsic variable for detecting temperalure.Thus,this dual parameter permits simultaneous sensing of temperature and strain without signal interference.We further demonstrale the applcability of this slotted junction to ultrasensitive chemical sensing which enables precise determination of donor type.acceptor type.and inert molecules.This is due to the enhancement or counterbalance between flexoelectric and chemical gating.Flexoelectric gating would preserve the electron-hole symmetry of the slotted junction whereas chemical gating would break it.As a proof-of-concept demonstration,the slotted CNT junction provides an excellent quantum platform for the development of mulistimuli sensation in artificial itelligence at the molecular scale.