Showing papers by "Nanjing University of Information Science and Technology published in 2017"

Cascaded Recurrent Neural Networks for Hyperspectral Image Classification

Abstract: In recent years, vector-based machine learning algorithms, such as random forests, support vector machines, and 1-D convolutional neural networks, have shown promising results in hyperspectral image classification. Such methodologies, nevertheless, can lead to information loss in representing hyperspectral pixels, which intrinsically have a sequence-based data structure. A recurrent neural network (RNN), an important branch of the deep learning family, is mainly designed to handle sequential data. Can sequence-based RNN be an effective method of hyperspectral image classification? In this paper, we propose a novel RNN model that can effectively analyze hyperspectral pixels as sequential data and then determine information categories via network reasoning. As far as we know, this is the first time that an RNN framework has been proposed for hyperspectral image classification. Specifically, our RNN makes use of a newly proposed activation function, parametric rectified tanh (PRetanh), for hyperspectral sequential data analysis instead of the popular tanh or rectified linear unit. The proposed activation function makes it possible to use fairly high learning rates without the risk of divergence during the training procedure. Moreover, a modified gated recurrent unit, which uses PRetanh for hidden representation, is adopted to construct the recurrent layer in our network to efficiently process hyperspectral data and reduce the total number of parameters. Experimental results on three airborne hyperspectral images suggest competitive performance in the proposed mode. In addition, the proposed network architecture opens a new window for future research, showcasing the huge potential of deep recurrent networks for hyperspectral data analysis.

Defect‐Engineered Ultrathin δ‐MnO2 Nanosheet Arrays as Bifunctional Electrodes for Efficient Overall Water Splitting

Abstract: Recently, defect engineering has been used to intruduce half-metallicity into selected semiconductors, thereby significantly enhancing their electrical conductivity and catalytic/electrocatalytic performance. Taking inspiration from this, we developed a novel bifunctional electrode consisting of two monolayer thick manganese dioxide (δ-MnO2) nanosheet arrays on a nickel foam, using a novel in-situ method. The bifunctional electrode exposes numerous active sites for electrocatalytic rections and displays excellent electrical conductivity, resulting in strong performance for both HER and OER. Based on detailed structure analysis and density functional theory (DFT) calculations, the remarkably OER and HER activity of the bifunctional electrode can be attributed to the ultrathin δ-MnO2 nanosheets containing abundant oxygen vacancies lead to the formation od Mn3+ active sites, which give rise to half-metallicity properties and strong H2O adsorption. This synthetic strategy introduced here represents a new method for the development of non-precious metal Mn-based electrocatalysts for eddicient energy conversion.

Aerosol and boundary-layer interactions and impact on air quality

Abstract: Air quality is concerned with pollutants in both the gas phase and solid or liquid phases. The latter are referred to as aerosols, which are multifaceted agents affecting air quality, weather and climate through many mechanisms. Unlike gas pollutants, aerosols interact strongly with meteorological variables with the strongest interactions taking place in the planetary boundary layer (PBL). The PBL hosting the bulk of aerosols in the lower atmosphere is affected by aerosol radiative effects. Both aerosol scattering and absorption reduce the amount of solar radiation reaching the ground and thus reduce the sensible heat fluxes that drive the diurnal evolution of the PBL. Moreover, aerosols can increase atmospheric stability by inducing a temperature inversion as a result of both scattering and absorption of solar radiation, which suppresses dispersion of pollutants and leads to further increases in aerosol concentration in the lower PBL. Such positive feedback is especially strong during severe pollution events. Knowledge of the PBL is thus crucial for understanding the interactions between air pollution and meteorology. A key question is how the diurnal evolution of the PBL interacts with aerosols, especially in vertical directions, and affects air quality. We review the major advances in aerosol measurements, PBL processes and their interactions with each other through complex feedback mechanisms, and highlight the priorities for future studies.

Layered‐Double‐Hydroxide Nanosheets as Efficient Visible‐Light‐Driven Photocatalysts for Dinitrogen Fixation

Abstract: Semiconductor photocatalysis attracts widespread interest in water splitting, CO2 reduction, and N2 fixation. N2 reduction to NH3 is essential to the chemical industry and to the Earth's nitrogen cycle. Industrially, NH3 is synthesized by the Haber-Bosch process under extreme conditions (400-500 °C, 200-250 bar), stimulating research into the development of sustainable technologies for NH3 production. Herein, this study demonstrates that ultrathin layered-double-hydroxide (LDH) photocatalysts, in particular CuCr-LDH nanosheets, possess remarkable photocatalytic activity for the photoreduction of N2 to NH3 in water at 25 °C under visible-light irradiation. The excellent activity can be attributed to the severely distorted structure and compressive strain in the LDH nanosheets, which significantly enhances N2 chemisorption and thereby promotes NH3 formation.

Weather conditions conducive to Beijing severe haze more frequent under climate change

Abstract: Severe winter air pollution events, attributed to emissions from development, have increased in Beijing in recent decades. This study looks at how atmospheric conditions contribute and projects climate change will increase conditions favourable to such events. The frequency of Beijing winter severe haze episodes has increased substantially over the past decades1,2,3,4, and is commonly attributed to increased pollutant emissions from China’s rapid economic development5,6. During such episodes, levels of fine particulate matter are harmful to human health and the environment, and cause massive disruption to economic activities3,4,7,8,9,10,11,12,13,14,15,16, as occurred in January 201317,18,19,20,21. Conducive weather conditions are an important ingredient of severe haze episodes3,21, and include reduced surface winter northerlies3,21, weakened northwesterlies in the midtroposphere, and enhanced thermal stability of the lower atmosphere1,3,16,21. How such weather conditions may respond to climate change is not clear. Here we project a 50% increase in the frequency and an 80% increase in the persistence of conducive weather conditions similar to those in January 2013, in response to climate change. The frequency and persistence between the historical (1950–1999) and future (2050–2099) climate were compared in 15 models under Representative Concentration Pathway 8.5 (RCP8.5)22. The increased frequency is consistent with large-scale circulation changes, including an Arctic Oscillation upward trend23,24, weakening East Asian winter monsoon25,26, and faster warming in the lower troposphere27,28. Thus, circulation changes induced by global greenhouse gas emissions can contribute to the increased Beijing severe haze frequency.

A Robust Regularization Path Algorithm for $\nu $ -Support Vector Classification

Abstract: The $ u $ -support vector classification has the advantage of using a regularization parameter $ u $ to control the number of support vectors and margin errors. Recently, a regularization path algorithm for $ u $ -support vector classification ( $ u $ -SvcPath) suffers exceptions and singularities in some special cases. In this brief, we first present a new equivalent dual formulation for $ u $ -SVC and, then, propose a robust $ u $ -SvcPath, based on lower upper decomposition with partial pivoting. Theoretical analysis and experimental results verify that our proposed robust regularization path algorithm can avoid the exceptions completely, handle the singularities in the key matrix, and fit the entire solution path in a finite number of steps. Experimental results also show that our proposed algorithm fits the entire solution path with fewer steps and less running time than original one does.

A novel collaborative optimization algorithm in solving complex optimization problems

Abstract: To overcome the deficiencies of weak local search ability in genetic algorithms (GA) and slow global convergence speed in ant colony optimization (ACO) algorithm in solving complex optimization problems, the chaotic optimization method, multi-population collaborative strategy and adaptive control parameters are introduced into the GA and ACO algorithm to propose a genetic and ant colony adaptive collaborative optimization (MGACACO) algorithm for solving complex optimization problems. The proposed MGACACO algorithm makes use of the exploration capability of GA and stochastic capability of ACO algorithm. In the proposed MGACACO algorithm, the multi-population strategy is used to realize the information exchange and cooperation among the various populations. The chaotic optimization method is used to overcome long search time, avoid falling into the local extremum and improve the search accuracy. The adaptive control parameters is used to make relatively uniform pheromone distribution, effectively solve the contradiction between expanding search and finding optimal solution. The collaborative strategy is used to dynamically balance the global ability and local search ability, and improve the convergence speed. Finally, various scale TSP are selected to verify the effectiveness of the proposed MGACACO algorithm. The experiment results show that the proposed MGACACO algorithm can avoid falling into the local extremum, and takes on better search precision and faster convergence speed.

Structural Minimax Probability Machine

Abstract: Minimax probability machine (MPM) is an interesting discriminative classifier based on generative prior knowledge. It can directly estimate the probabilistic accuracy bound by minimizing the maximum probability of misclassification. The structural information of data is an effective way to represent prior knowledge, and has been found to be vital for designing classifiers in real-world problems. However, MPM only considers the prior probability distribution of each class with a given mean and covariance matrix, which does not efficiently exploit the structural information of data. In this paper, we use two finite mixture models to capture the structural information of the data from binary classification. For each subdistribution in a finite mixture model, only its mean and covariance matrix are assumed to be known. Based on the finite mixture models, we propose a structural MPM (SMPM). SMPM can be solved effectively by a sequence of the second-order cone programming problems. Moreover, we extend a linear model of SMPM to a nonlinear model by exploiting kernelization techniques. We also show that the SMPM can be interpreted as a large margin classifier and can be transformed to support vector machine and maxi–min margin machine under certain special conditions. Experimental results on both synthetic and real-world data sets demonstrate the effectiveness of SMPM.

Effective and Efficient Global Context Verification for Image Copy Detection

Abstract: To detect illegal copies of copyrighted images, recent copy detection methods mostly rely on the bag-of-visual-words (BOW) model, in which local features are quantized into visual words for image matching. However, both the limited discriminability of local features and the BOW quantization errors will lead to many false local matches, which make it hard to distinguish similar images from copies. Geometric consistency verification is a popular technology for reducing the false matches, but it neglects global context information of local features and thus cannot solve this problem well. To address this problem, this paper proposes a global context verification scheme to filter false matches for copy detection. More specifically, after obtaining initial scale invariant feature transform (SIFT) matches between images based on the BOW quantization, the overlapping region-based global context descriptor (OR-GCD) is proposed for the verification of these matches to filter false matches. The OR-GCD not only encodes relatively rich global context information of SIFT features but also has good robustness and efficiency. Thus, it allows an effective and efficient verification. Furthermore, a fast image similarity measurement based on random verification is proposed to efficiently implement copy detection. In addition, we also extend the proposed method for partial-duplicate image detection. Extensive experiments demonstrate that our method achieves higher accuracy than the state-of-the-art methods, and has comparable efficiency to the baseline method based on the BOW quantization.

Study on an improved adaptive PSO algorithm for solving multi-objective gate assignment

Abstract: Display Omitted An improved adaptive PSO based on Alpha-stable distribution and dynamic fractional calculus is studied.A new multi-objective optimization model of gate assignment problem is proposed.The actual data are used to demonstrate the effectiveness of the proposed method. Gate is a key resource in the airport, which can realize rapid and safe docking, ensure the effective connection between flights and improve the capacity and service efficiency of airport. The minimum walking distances of passengers, the minimum idle time variance of each gate, the minimum number of flights at parking apron and the most reasonable utilization of large gates are selected as the optimization objectives, then an efficient multi-objective optimization model of gate assignment problem is proposed in this paper. Then an improved adaptive particle swarm optimization(DOADAPO) algorithm based on making full use of the advantages of Alpha-stable distribution and dynamic fractional calculus is deeply studied. The dynamic fractional calculus with memory characteristic is used to reflect the trajectory information of particle updating in order to improve the convergence speed. The Alpha-stable distribution theory is used to replace the uniform distribution in order to escape from the local minima in a certain probability and improve the global search ability. Next, the DOADAPO algorithm is used to solve the constructed multi-objective optimization model of gate assignment in order to fast and effectively assign the gates to different flights in different time. Finally, the actual flight data in one domestic airport is used to verify the effectiveness of the proposed method. The experiment results show that the DOADAPO algorithm can improve the convergence speed and enhance the local search ability and global search ability, and the multi-objective optimization model of gate assignment can improve the comprehensive service of gate assignment. It can effectively provide a valuable reference for assigning the gates in hub airport.

Performance Comparison of PS Star-16QAM and PS Square-Shaped 16QAM (Square-16QAM)

Abstract: We investigate and compare the performance of star-shaped 16-ary quadrature amplitude modulation (Star-16QAM) and square-shaped 16QAM (Square-16QAM) in the probabilistic shaping (PS) and uniform schemes with coherent detection. With the help of PS technology, the bit error ratio (BER) improvement achieved in the PS-Star-16QAM scheme is greater than that of the PS-Square-16QAM when compared with the uniform schemes in our numerical simulation and experiment. Therefore, the PS-Star-16QAM shows superiority over the PS-Square-16QAM in terms of the BER improvement.

An Efficient Public Auditing Protocol With Novel Dynamic Structure for Cloud Data

Abstract: With the rapid development of cloud computing, cloud storage has been accepted by an increasing number of organizations and individuals, therein serving as a convenient and on-demand outsourcing application However, upon losing local control of data, it becomes an urgent need for users to verify whether cloud service providers have stored their data securely Hence, many researchers have devoted themselves to the design of auditing protocols directed at outsourced data In this paper, we propose an efficient public auditing protocol with global and sampling blockless verification as well as batch auditing, where data dynamics are substantially more efficiently supported than is the case with the state of the art Note that, the novel dynamic structure in our protocol consists of a doubly linked info table and a location array Moreover, with such a structure, computational and communication overheads can be reduced substantially Security analysis indicates that our protocol can achieve the desired properties Moreover, numerical analysis and real-world experimental results demonstrate that the proposed protocol achieves a given efficiency in practice

Black-carbon absorption enhancement in the atmosphere determined by particle mixing state

Abstract: Mixing with non-black carbon can enhance the radiative effect of black-carbon aerosols. Lab and field measurements of aerosol properties reveal that the mass ratio of black to non-black carbon determines the amount of enhancement.

Compact and uniform TiO2@g-C3N4 core-shell quantum heterojunction for photocatalytic degradation of tetracycline antibiotics

Abstract: Optimizing the heterojunction structure of semiconductor photocatalysts is significant for taking full advantage of their abilities for organic molecules degradation. Here, we demonstrate a feasible strategy of polymerizing the quantum-thick graphitic carbon nitride (g-C3N4) on to the surface of anatase titanium dioxide (TiO2) nanosheets with exposed {001} facets to form the TiO2@g-C3N4 (TCN) core-shell quantum heterojunction for improving photocatalytic tetracycline degradation activity. 100 mg of TCN photocatalyst shows the highest tetracycline degradation rate of 2.2 mg/min, which is 36% higher than that of the TiO2/g-C3N4 random mixture (TCN(mix)), 2 times higher than that of TiO2, and 2.3 times higher than that of bulk g-C3N4. Results also indicate that h+ and ·O2− are the main oxidant species for the efficient photocatalytic reaction. The decisive factors in improving the photocatalytic activity of TCN is the unique structural advantages of quantum-thick g-C3N4 shell, compact and uniform contact interface, richly available reaction sites, more surface adsorbed hydroxyl (OH) groups. Efficient electron transfer between TiO2 and g-C3N4 is also demonstrated by the significant enhancement of photocurrent response of TCN electrodes and decrement of fluorescence emission spectra. This work demonstrates new sights for synthesizing high-efficient and environment-stable photocatalysts by engineering the surface heterojunction.

Recent advances in understanding of the mechanism and control of Li2O2 formation in aprotic Li–O2 batteries

Abstract: Aprotic Li–O2 batteries represent promising alternative devices for electrical energy storage owing to their extremely high energy densities. Upon discharge, insulating solid Li2O2 forms on cathode surfaces, which is usually governed by two growth models, namely the solution model and the surface model. These Li2O2 growth models can largely determine the battery performances such as the discharge capacity, round-trip efficiency and cycling stability. Understanding the Li2O2 formation mechanism and controlling its growth are essential to fully realize the technological potential of Li–O2 batteries. In this review, we overview the recent advances in understanding the electrochemical and chemical processes that occur during the Li2O2 formation. In the beginning, the oxygen reduction mechanisms, the identification of O2−/LiO2 intermediates, and their influence on the Li2O2 morphology have been discussed. The effects of the discharge current density and potential on the Li2O2 growth model have been subsequently reviewed. Special focus is then given to the prominent strategies, including the electrolyte-mediated strategy and the cathode-catalyst-tailoring strategy, for controlling the Li2O2 growth pathways. Finally, we conclude by discussing the profound implications of controlling Li2O2 formation for further development in Li–O2 batteries.

Revisiting the contribution of transpiration to global terrestrial evapotranspiration

Abstract: Even though knowing the contributions of transpiration (T), soil and open water evaporation (E), and interception (I) to terrestrial evapotranspiration (ET = T + E + I) is crucial for understanding the hydrological cycle and its connection to ecological processes, the fraction of T is unattainable by traditional measurement techniques over large scales. Previously reported global mean T/(E + T + I) from multiple independent sources, including satellite-based estimations, reanalysis, land surface models, and isotopic measurements, varies substantially from 24% to 90%. Here we develop a new ET partitioning algorithm, which combines global evapotranspiration estimates and relationships between leaf area index (LAI) and T/(E + T) for different vegetation types, to upscale a wide range of published site-scale measurements. We show that transpiration accounts for about 57.2% (with standard deviation ± 6.8%) of global terrestrial ET. Our approach bridges the scale gap between site measurements and global model simulations,and can be simply implemented into current global climate models to improve biological CO2 flux simulations.

Stacked Hourglass Network for Robust Facial Landmark Localisation

Abstract: With the increasing number of public available training data for face alignment, the regression-based methods attracted much attention and have become the dominant methods to solve this problem. There are two main factors, the variance of the regression target and the capacity of the regression model, affecting the performance of the regression task. In this paper, we present a Stacked Hourglass Network for robust facial landmark localisation. We first adopt a supervised face transformation to remove the translation, scale and rotation variation of each face, in order to reduce the variance of the regression target. Then we employ a deep convolutional neural network named Stacked Hourglass Network to increase the capacity of the regression model. To better evaluate the proposed method, we reimplement two popular cascade shape regression models, SDM and LBF, for comparison. Extensive experiments on four challenging datasets prove the effectiveness of the proposed method.

CoupleNet: Coupling Global Structure with Local Parts for Object Detection

Abstract: The region-based Convolutional Neural Network (CNN) detectors such as Faster R-CNN or R-FCN have already shown promising results for object detection by combining the region proposal subnetwork and the classification subnetwork together. Although R-FCN has achieved higher detection speed while keeping the detection performance, the global structure information is ignored by the position-sensitive score maps. To fully explore the local and global properties, in this paper, we propose a novel fully convolutional network, named as CoupleNet, to couple the global structure with local parts for object detection. Specifically, the object proposals obtained by the Region Proposal Network (RPN) are fed into the the coupling module which consists of two branches. One branch adopts the position-sensitive RoI (PSRoI) pooling to capture the local part information of the object, while the other employs the RoI pooling to encode the global and context information. Next, we design different coupling strategies and normalization ways to make full use of the complementary advantages between the global and local branches. Extensive experiments demonstrate the effectiveness of our approach. We achieve state-of-the-art results on all three challenging datasets, i.e. a mAP of 82.7% on VOC07, 80.4% on VOC12, and 34.4% on COCO. Codes will be made publicly available1.

The peak structure and future changes of the relationships between extreme precipitation and temperature

Abstract: Theoretical models predict that, in the absence of moisture limitation, extreme precipitation intensity could exponentially increase with temperatures at a rate determined by the Clausius-Clapeyron (C-C) relationship. Climate models project a continuous increase of precipitation extremes for the twenty-first century over most of the globe. However, some station observations suggest a negative scaling of extreme precipitation with very high temperatures, raising doubts about future increase of precipitation extremes. Here we show for the present-day climate over most of the globe,the curve relating daily precipitation extremes with local temperatures has a peak structure, increasing as expected at the low medium range of temperature variations but decreasing at high temperatures. However, this peak-shaped relationship does not imply a potential upper limit for future precipitation extremes. Climate models project both the peak of extreme precipitation and the temperature at which it peaks (T(sub peak)) will increase with warming; the two increases generally conform to the C-C scaling rate in mid- and high-latitudes,and to a super C-C scaling in most of the tropics. Because projected increases of local mean temperature (T(sub mean)) far exceed projected increases of T(sub peak) over land, the conventional approach of relating extreme precipitation to T(sub mean) produces a misleading sub-C-C scaling rate.

Distributed Aggregate Privacy-Preserving Authentication in VANETs

Abstract: Existing secure and privacy-preserving vehicular communication protocols in vehicular ad hoc networks face the challenges of being fast and not depending on ideal tamper-proof devices (TPDs) embedded in vehicles. To address these challenges, we propose a vehicular authentication protocol referred to as distributedaggregate privacy-preserving authentication. The proposed protocol is based on our new multiple trusted authority one-time identity-based aggregate signature technique. With this technique a vehicle can verify many messages simultaneously and their signatures can be compressed into a single one that greatly reduces the storage space needed by a vehicle or a data collector (e.g., the traffic management authority). Instead of ideal TPDs, our protocol only requires realistic TPDs and hence is more practical.

Bioinspired Polydopamine Sheathed Nanofibers Containing Carboxylate Graphene Oxide Nanosheet for High-Efficient Dyes Scavenger

Abstract: New hierarchical bioinspired nanocomposite materials of poly(vinyl alcohol)/poly(acrylic acid)/carboxylate graphene oxide nanosheet@polydopamine (PVA/PAA/GO-COOH@PDA) were successfully prepared by electrospinning technique, thermal treatment, and polydopamine modification. The obtained composite membranes are composed of polymeric nanofibers with carboxylate graphene oxide nanosheets, which are anchored on the fibers by heat-induced cross-linking reaction. The preparation process demonstrate eco-friendly and controllable manner. These as-formed nanocomposites were characterized by various morphological methods and spectral techniques. Due to the unique polydopamine and graphene oxide containing structures in composites, the as-obtained composite demonstrate well efficient adsorption capacity toward dye removal, which is primarily due to the specific surface area of electrospun membranes and the active polydopamine/graphene oxide components. In addition, the composite membranes reported here are easy to re...

Theories on formation of an anomalous anticyclone in western North Pacific during El Niño: A review

Abstract: The western North Pacific anomalous anticyclone (WNPAC) is an important atmospheric circulation system that conveys El Nino impact on East Asian climate. In this review paper, various theories on the formation and maintenance of the WNPAC, including warm pool atmosphere–ocean interaction, Indian Ocean capacitor, a combination mode that emphasizes nonlinear interaction between ENSO and annual cycle, moist enthalpy advection/Rossby wave modulation, and central Pacific SST forcing, are discussed. It is concluded that local atmosphere–ocean interaction and moist enthalpy advection/Rossby wave modulation mechanisms are essential for the initial development and maintenance of the WNPAC during El Nino mature winter and subsequent spring. The Indian Ocean capacitor mechanism does not contribute to the earlier development but helps maintain the WNPAC in El Nino decaying summer. The cold SST anomaly in the western North Pacific, although damped in the summer, also plays a role. An inter-basin atmosphere–ocean interaction across the Indo-Pacific warm pool emerges as a new mechanism in summer. In addition, the central Pacific cold SST anomaly may induce the WNPAC during rapid El Nino decaying/La Nina developing or La Nina persisting summer. The near-annual periods predicted by the combination mode theory are hardly detected from observations and thus do not contribute to the formation of the WNPAC. The tropical Atlantic may have a capacitor effect similar to the tropical Indian Ocean.

Dye aggregation in dye-sensitized solar cells

Abstract: Dye aggregation dictates structural and optoelectronic properties of photoelectrodes in dye-sensitized solar cells (DSSCs), thereby playing an essential role in their photovoltaic performance. It is therefore important to understand this molecular phenomenon so that dye aggregation can be suitably controlled in DSSC devices. Accordingly, this review presents the molecular origins of dye aggregation, manifestations of hypsochromic (H) and bathochromic (J) dye aggregation in DSSCs, a classification of the molecular factors that cause this aggregation, and ways by which dye aggregation can be suppressed in cases where it is undesirable. To this end, a classification of molecular engineering methods that are being used in order to better understand and control dye aggregation is described. The review concludes with a broader outlook on how molecular aggregation in chromophores can similarly effect a wider range of optoelectronic phenomena.