2016

• Record 265 of
  
  Title:All-optical control of microfiber resonator by graphene's photothermal effect
  
  Author(s):Wang, Yadong(1); Gan, Xuetao(1); Zhao, Chenyang(1); Fang, Liang(1); Mao, Dong(1); Xu, Yiping(2); Zhang, Fanlu(1); Xi, Teli(1); Ren, Liyong(2); Zhao, Jianlin(1)
  Source: Applied Physics Letters  Volume: 108  Issue: 17  DOI: 10.1063/1.4947577  Published: April 25, 2016  
  Abstract:We demonstrate an efficient all-optical control of microfiber resonator assisted by graphene's photothermal effect. Wrapping graphene onto a microfiber resonator, the light-graphene interaction can be strongly enhanced via the resonantly circulating light, which enables a significant modulation of the resonance with a resonant wavelength shift rate of 71 pm/mW when pumped by a 1540 nm laser. The optically controlled resonator enables the implementation of low threshold optical bistability and switching with an extinction ratio exceeding 13 dB. The thin and compact structure promises a fast response speed of the control, with a rise (fall) time of 294.7 μs (212.2 μs) following the 10%-90% rule. The proposed device, with the advantages of compact structure, all-optical control, and low power acquirement, offers great potential in the miniaturization of active in-fiber photonic devices. © 2016 Author(s).
  Accession Number: 20162202429172
• Record 266 of
  
  Title:Measuring Collectiveness via Refined Topological Similarity
  
  Author(s):Li, Xuelong(1); Chen, Mulin(2); Wang, Qi(2)
  Source: ACM Transactions on Multimedia Computing, Communications and Applications  Volume: 12  Issue: 2  DOI: 10.1145/2854000  Published: March 2016  
  Abstract:Crowd system has motivated a surge of interests in many areas of multimedia, as it contains plenty of information about crowd scenes. In crowd systems, individuals tend to exhibit collective behaviors, and the motion of all those individuals is called collective motion. As a comprehensive descriptor of collective motion, collectiveness has been proposed to reflect the degree of individuals moving as an entirety. Nevertheless, existing works mostly have limitations to correctly find the individuals of a crowd system and precisely capture the various relationships between individuals, both of which are essential to measure collectiveness. In this article, we propose a collectiveness-measuring method that is capable of quantifying collectiveness accurately. Our main contributions are threefold: (1) we compute relatively accurate collectiveness bymaking the tracked feature points represent the individuals more precisely with a point selection strategy; (2) we jointly investigate the spatial-temporal information of individuals and utilize it to characterize the topological relationship between individuals by manifold learning; (3) we propose a stability descriptor to deal with the irregular individuals, which influence the calculation of collectiveness. Intensive experiments on the simulated and real world datasets demonstrate that the proposed method is able to compute relatively accurate collectiveness and keep high consistency with human perception. © 2016 Copyright held by the owner/author(s).
  Accession Number: 20162102408664
• Record 267 of
  
  Title:Ensemble Manifold Rank Preserving for Acceleration-Based Human Activity Recognition
  
  Author(s):Tao, Dapeng(1); Jin, Lianwen(1); Yuan, Yuan(2); Xue, Yang(1)
  Source: IEEE Transactions on Neural Networks and Learning Systems  Volume: 27  Issue: 6  DOI: 10.1109/TNNLS.2014.2357794  Published: June 2016  
  Abstract:With the rapid development of mobile devices and pervasive computing technologies, acceleration-based human activity recognition, a difficult yet essential problem in mobile apps, has received intensive attention recently. Different acceleration signals for representing different activities or even a same activity have different attributes, which causes troubles in normalizing the signals. We thus cannot directly compare these signals with each other, because they are embedded in a nonmetric space. Therefore, we present a nonmetric scheme that retains discriminative and robust frequency domain information by developing a novel ensemble manifold rank preserving (EMRP) algorithm. EMRP simultaneously considers three aspects: 1) it encodes the local geometry using the ranking order information of intraclass samples distributed on local patches; 2) it keeps the discriminative information by maximizing the margin between samples of different classes; and 3) it finds the optimal linear combination of the alignment matrices to approximate the intrinsic manifold lied in the data. Experiments are conducted on the South China University of Technology naturalistic 3-D acceleration-based activity dataset and the naturalistic mobile-devices based human activity dataset to demonstrate the robustness and effectiveness of the new nonmetric scheme for acceleration-based human activity recognition. © 2012 IEEE.
  Accession Number: 20144300129540
• Record 268 of
  
  Title:DISC: Deep Image Saliency Computing via Progressive Representation Learning
  
  Author(s):Chen, Tianshui(1); Lin, Liang(1); Liu, Lingbo(1); Luo, Xiaonan(1); Li, Xuelong(2)
  Source: IEEE Transactions on Neural Networks and Learning Systems  Volume: 27  Issue: 6  DOI: 10.1109/TNNLS.2015.2506664  Published: June 2016  
  Abstract:Salient object detection increasingly receives attention as an important component or step in several pattern recognition and image processing tasks. Although a variety of powerful saliency models have been intensively proposed, they usually involve heavy feature (or model) engineering based on priors (or assumptions) about the properties of objects and backgrounds. Inspired by the effectiveness of recently developed feature learning, we provide a novel deep image saliency computing (DISC) framework for fine-grained image saliency computing. In particular, we model the image saliency from both the coarse-and fine-level observations, and utilize the deep convolutional neural network (CNN) to learn the saliency representation in a progressive manner. In particular, our saliency model is built upon two stacked CNNs. The first CNN generates a coarse-level saliency map by taking the overall image as the input, roughly identifying saliency regions in the global context. Furthermore, we integrate superpixel-based local context information in the first CNN to refine the coarse-level saliency map. Guided by the coarse saliency map, the second CNN focuses on the local context to produce fine-grained and accurate saliency map while preserving object details. For a testing image, the two CNNs collaboratively conduct the saliency computing in one shot. Our DISC framework is capable of uniformly highlighting the objects of interest from complex background while preserving well object details. Extensive experiments on several standard benchmarks suggest that DISC outperforms other state-of-the-art methods and it also generalizes well across data sets without additional training. The executable version of DISC is available online: http://vision.sysu.edu.cn/projects/DISC. © 2015 IEEE.
  Accession Number: 20160201782781
• Record 269 of
  
  Title:Pedestrian Detection Inspired by Appearance Constancy and Shape Symmetry
  
  Author(s):Cao, Jiale(1); Pang, Yanwei(1); Li, Xuelong(2)
  Source: IEEE Transactions on Image Processing  Volume: 25  Issue: 12  DOI: 10.1109/TIP.2016.2609807  Published: October 2016  
  Abstract:Most state-of-the-art methods in pedestrian detection are unable to achieve a good trade-off between accuracy and efficiency. For example, ACF has a fast speed but a relatively low detection rate, while checkerboards have a high detection rate but a slow speed. Inspired by some simple inherent attributes of pedestrians (i.e., appearance constancy and shape symmetry), we propose two new types of non-neighboring features: side-inner difference features (SIDF) and symmetrical similarity features (SSFs). SIDF can characterize the difference between the background and pedestrian and the difference between the pedestrian contour and its inner part. SSF can capture the symmetrical similarity of pedestrian shape. However, it is difficult for neighboring features to have such above characterization abilities. Finally, we propose to combine both non-neighboring features and neighboring features for pedestrian detection. It is found that non-neighboring features can further decrease the log-average miss rate by 4.44%. The relationship between our proposed method and some state-of-the-art methods is also given. Experimental results on INRIA, Caltech, and KITTI data sets demonstrate the effectiveness and efficiency of the proposed method. Compared with the state-of-the-art methods without using CNN, our method achieves the best detection performance on Caltech, outperforming the second best method (i.e., checkerboards) by 2.27%. Using the new annotations of Caltech, it can achieve 11.87% miss rate, which outperforms other methods. © 2016 IEEE.
  Accession Number: 20164703035678
• Record 270 of
  
  Title:Influence of longitudinal argon flow on DC glow discharge at atmospheric pressure
  
  Author(s):Zhu, Sha(1); Jiang, Weiman(1); Tang, Jie(1); Xu, Yonggang(1,2); Wang, Yishan(1); Zhao, Wei(1); Duan, Yixiang(1,3)
  Source: Japanese Journal of Applied Physics  Volume: 55  Issue: 5  DOI: 10.7567/JJAP.55.056202  Published: May 2016  
  Abstract:A one-dimensional self-consistent fluid model was employed to investigate the influence of longitudinal argon flow on the DC glow discharge at atmospheric pressure. It is found that the charges exhibit distinct dynamic behaviors at different argon flow velocities, accompanied by a considerable change in the discharge structure. The positive argon flow allows for the reduction of charge densities in the positive column and negative glow regions, and even leads to the disappearance of negative glow. The negative argon flow gives rise to the enhancement of charge densities in the positive column and negative glow regions. These observations are attributed to the fact that the gas flow convection influences the transport of charges through different manners by comparing the argon flow velocity with the ion drift velocity. The findings are important for improving the chemical activity and work efficiency of the plasma source by controlling the gas flow in practical applications. © 2016 The Japan Society of Applied Physics.
  Accession Number: 20161902359183
• Record 271 of
  
  Title:Optimization of the electron collection efficiency of a large area MCP-PMT for the JUNO experiment
  
  Author(s):Chen, Lin(1,2,5); Tian, Jinshou(2); Liu, Chunliang(5); Wang, Yifang(3); Zhao, Tianchi(3); Liu, Hulin(2); Wei, Yonglin(2); Sai, Xiaofeng(2); Chen, Ping(1,2); Wang, Xing(2); Lu, Yu(2); Hui, Dandan(1,2); Guo, Lehui(1,2); Liu, Shulin(3); Qian, Sen(3); Xia, Jingkai(3); Yan, Baojun(3); Zhu, Na(3); Sun, Jianning(4); Si, Shuguang(4); Li, Dong(4); Wang, Xingchao(4); Huang, Guorui(4); Qi, Ming(6)
  Source: Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment  Volume: 827  Issue:   DOI: 10.1016/j.nima.2016.04.100  Published: August 11, 2016  
  Abstract:A novel large-area (20-inch) photomultiplier tube based on microchannel plate (MCP-PMTs) is proposed for the Jiangmen Underground Neutrino Observatory (JUNO) experiment. Its photoelectron collection efficiency Ce is limited by the MCP open area fraction (Aopen). This efficiency is studied as a function of the angular (θ), energy (E) distributions of electrons in the input charge cloud and the potential difference (U) between the PMT photocathode and the MCP input surface, considering secondary electron emission from the MCP input electrode. In CST Studio Suite, Finite Integral Technique and Monte Carlo method are combined to investigate the dependence of Ce on θ, E and U. Results predict that Ce can exceed Aopen, and are applied to optimize the structure and operational parameters of the 20-inch MCP-PMT prototype. Ce of the optimized MCP-PMT is expected to reach 81.2%. Finally, the reduction of the penetration depth of the MCP input electrode layer and the deposition of a high secondary electron yield material on the MCP are proposed to further optimize Ce. © 2016 Elsevier B.V. All rights reserved.
  Accession Number: 20162002384064
• Record 272 of
  
  Title:Deep representation for abnormal event detection in crowded scenes
  
  Author(s):Feng, Yachuang(1,2); Yuan, Yuan(1); Lu, Xiaoqiang(1)
  Source: MM 2016 - Proceedings of the 2016 ACM Multimedia Conference  Volume:   Issue:   DOI: 10.1145/2964284.2967290  Published: October 1, 2016  
  Abstract:Abnormal event detection is extremely important, especially for video surveillance. Nowadays, many detectors have been proposed based on hand-crafted features. However, it remains challenging to effectively distinguish abnormal events from normal ones. This paper proposes a deep representation based algorithm which extracts features in an unsupervised fashion. Specially, appearance, texture, and short-term motion features are automatically learned and fused with stacked denoising autoencoders. Subsequently, long-term temporal clues are modeled with a long short-term memory (LSTM) recurrent network, in order to discover meaningful regularities of video events. The abnormal events are identified as samples which disobey these regularities. Moreover, this paper proposes a spatial anomaly detection strategy via manifold ranking, aiming at excluding false alarms. Experiments and comparisons on real world datasets show that the proposed algorithm outper-forms state of the arts for the abnormal event detection problem in crowded scenes. © 2016 ACM.
  Accession Number: 20164603010560
• Record 273 of
  
  Title:Block-Row Sparse Multiview Multilabel Learning for Image Classification
  
  Author(s):Zhu, Xiaofeng(1,2); Li, Xuelong(3); Zhang, Shichao(4)
  Source: IEEE Transactions on Cybernetics  Volume: 46  Issue: 2  DOI: 10.1109/TCYB.2015.2403356  Published: February 2016  
  Abstract:In image analysis, the images are often represented by multiple visual features (also known as multiview features), that aim to better interpret them for achieving remarkable performance of the learning. Since the processes of feature extraction on each view are separated, the multiple visual features of images may include overlap, noise, and redundancy. Thus, learning with all the derived views of the data could decrease the effectiveness. To address this, this paper simultaneously conducts a hierarchical feature selection and a multiview multilabel (MVML) learning for multiview image classification, via embedding a proposed a new block-row regularizer into the MVML framework. The block-row regularizer concatenating a Frobenius norm (F-norm) regularizer and an 2,1-norm regularizer is designed to conduct a hierarchical feature selection, in which the F-norm regularizer is used to conduct a high-level feature selection for selecting the informative views (i.e., discarding the uninformative views) and the 2,1-norm regularizer is then used to conduct a low-level feature selection on the informative views. The rationale of the use of a block-row regularizer is to avoid the issue of the over-fitting (via the block-row regularizer), to remove redundant views and to preserve the natural group structures of data (via the F-norm regularizer), and to remove noisy features (the 2,1-norm regularizer), respectively. We further devise a computationally efficient algorithm to optimize the derived objective function and also theoretically prove the convergence of the proposed optimization method. Finally, the results on real image datasets show that the proposed method outperforms two baseline algorithms and three state-of-The-Art algorithms in terms of classification performance. © 2013 IEEE.
  Accession Number: 20150900590339
• Record 274 of
  
  Title:Hyperspectral anomaly detection by graph pixel selection
  
  Author(s):Yuan, Yuan(1); Ma, Dandan(1); Wang, Qi(2,3)
  Source: IEEE Transactions on Cybernetics  Volume: 46  Issue: 10  DOI: 10.1109/TCYB.2015.2497711  Published: November 20, 2015  
  Abstract:Hyperspectral anomaly detection (AD) is an important problem in remote sensing field. It can make full use of the spectral differences to discover certain potential interesting regions without any target priors. Traditional Mahalanobisdistancebased anomaly detectors assume the background spectrum distribution conforms to a Gaussian distribution. However, this and other similar distributions may not be satisfied for the real hyperspectral images. Moreover, the background statistics are susceptible to contamination of anomaly targets which will lead to a high false-positive rate. To address these intrinsic problems, this paper proposes a novel AD method based on the graph theory. We first construct a vertex- and edge-weighted graph and then utilize a pixel selection process to locate the anomaly targets. Two contributions are claimed in this paper: 1) no background distributions are required which makes the method more adaptive and 2) both the vertex and edge weights are considered which enables a more accurate detection performance and better robustness to noise. Intensive experiments on the simulated and real hyperspectral images demonstrate that the proposed method outperforms other benchmark competitors. In addition, the robustness of the proposed method has been validated by using various window sizes. This experimental result also demonstrates the valuable characteristic of less computational complexity and less parameter tuning for real applications. © 2015 IEEE.
  Accession Number: 20154801612558
• Record 275 of
  
  Title:Local structure learning in high resolution remote sensing image retrieval
  
  Author(s):Du, Zhongxiang(1,2); Li, Xuelong(1); Lu, Xiaoqiang(1)
  Source: Neurocomputing  Volume: 207  Issue:   DOI: 10.1016/j.neucom.2016.05.061  Published: 26 September 2016  
  Abstract:High resolution remote sensing image captured by the satellites or the aircraft is of great help for military and civilian applications. In recent years, with an increasing amount of high resolution remote sensing images, it becomes more and more urgent to find a way to retrieve them. In this case, a few methods based on the statistical information of the local features are proposed, which have achieved good performances. However, most of the methods do not take the topological structure of the features into account. In this paper, we propose a new method to represent these images, by taking the structural information into consideration. The main contributions of this paper include: (1) mapping the features into a manifold space by a Lipschitz smooth function to enhance the representation ability of the features; (2) training an anchor set with several regularization constrains to get the intrinsic manifold structure. In the experiments, the method is applied to two challenging remote sensing image datasets: UC Merced land use dataset and Sydney dataset. Compared to the state-of-the-art approaches, the proposed method can achieve a more robust and commendable performance. © 2016 Elsevier B.V.
  Accession Number: 20162802588788
• Record 276 of
  
  Title:Pixel-to-Model Distance for Robust Background Reconstruction
  
  Author(s):Yang, Lu(1); Cheng, Hong(1); Su, Jianan(1); Li, Xuelong(2)
  Source: IEEE Transactions on Circuits and Systems for Video Technology  Volume: 26  Issue: 5  DOI: 10.1109/TCSVT.2015.2424052  Published: May 2016  
  Abstract:Background information is crucial for many video surveillance applications such as object detection and scene understanding. In this paper, we present a novel pixel-to-model (P2M) paradigm for background modeling and restoration in surveillance scenes. In particular, the proposed approach models the background with a set of context features for each pixel, which are compressively sensed from local patches. We determine whether a pixel belongs to the background according to the minimum P2M distance, which measures the similarity between the pixel and its background model in the space of compressive local descriptors. The pixel feature descriptors of the background model are properly updated with respect to the minimum P2M distance. Meanwhile, the neighboring background model will be renewed according to the maximum P2M distance to handle ghost holes. The P2M distance plays an important role of background reliability in the 3-D spatial-temporal domain of surveillance videos, leading to the robust background model and recovered background videos. We applied the proposed P2M distance for foreground detection and background restoration on synthetic and real-world surveillance videos. Experimental results show that the proposed P2M approach outperforms the state-of-the-art approaches both in indoor and outdoor surveillance scenes. © 2015 IEEE.
  Accession Number: 20162202437322