2016

• Record 49 of
  
  Title:Wireless and sensorless 3D ultrasound imaging
  
  Author(s):Gao, Haitao(1,2); Huang, Qinghua(1,2); Xu, Xiangmin(1); Li, Xuelong(3)
  Source: Neurocomputing  Volume: 195  Issue:   DOI: 10.1016/j.neucom.2015.08.109  Published: June 26, 2016  
  Abstract:The past decade has witnessed great advances in three-dimensional (3-D) medical ultrasound (US) imaging instrumentation. An increasing demand for portable 3-D US equipment is one of the main trends upcoming in the market. In this study, we developed a low cost, portable, sensorless and wireless 3-D US imaging system. A laptop US scanner with a conventional linear probe and a convex probe was used to acquire 2-D US B-scans. A client program was developed and run on the US scanner for capturing the pictures of screen during a freehand scanning without a positional sensor, and then the JPEG compression was applied to the pictures for reducing the image data size. The image data was sent to a remote workstation in real-time through Wi-Fi connection. A neural network model was used to recognize the characters (e.g. imaging depth and probe model information) displayed on the screen of the US scanner. The server on the remote workstation communicated with the US scanner, received raw image data, and finally reconstructed 3-D US images. The positions of the B-scans were obtained by estimating the spacings of B-scan image sequence, which was learned by measuring adaptive speckle decorrelation curves in mechanically collected B-scan frames. The performance of the proposed system has been demonstrated through experiments conducted on a US resolution phantom in vitro as well as human tissues in vivo. © 2016 Elsevier B.V.
  Accession Number: 20160701945500
• Record 50 of
  
  Title:Discriminative transfer subspace learning via low-rank and sparse representation
  
  Author(s):Xu, Yong(1,2); Fang, Xiaozhao(1); Wu, Jian(1); Li, Xuelong(3); Zhang, David(4)
  Source: IEEE Transactions on Image Processing  Volume: 25  Issue: 2  DOI: 10.1109/TIP.2015.2510498  Published: February 2016  
  Abstract:In this paper, we address the problem of unsupervised domain transfer learning in which no labels are available in the target domain. We use a transformation matrix to transfer both the source and target data to a common subspace, where each target sample can be represented by a combination of source samples such that the samples from different domains can be well interlaced. In this way, the discrepancy of the source and target domains is reduced. By imposing joint low-rank and sparse constraints on the reconstruction coefficient matrix, the global and local structures of data can be preserved. To enlarge the margins between different classes as much as possible and provide more freedom to diminish the discrepancy, a flexible linear classifier (projection) is obtained by learning a non-negative label relaxation matrix that allows the strict binary label matrix to relax into a slack variable matrix. Our method can avoid a potentially negative transfer by using a sparse matrix to model the noise and, thus, is more robust to different types of noise. We formulate our problem as a constrained low-rankness and sparsity minimization problem and solve it by the inexact augmented Lagrange multiplier method. Extensive experiments on various visual domain adaptation tasks show the superiority of the proposed method over the state-of-the art methods. The MATLAB code of our method will be publicly available at http://www.yongxu.org/lunwen.html. © 2015 IEEE.
  Accession Number: 20170303260633
• Record 51 of
  
  Title:Toward solving the Steiner travelling salesman problem on urban road maps using the branch decomposition of graphs
  
  Author(s):Xia, Yingjie(1); Zhu, Mingzhe(2); Gu, Qianping(2); Zhang, Luming(3); Li, Xuelong(4)
  Source: Information Sciences  Volume: 374  Issue:   DOI: 10.1016/j.ins.2016.09.043  Published: December 20, 2016  
  Abstract:The Steiner travelling salesman problem (STSP) is an important issue in intelligent transportation systems and has various practical applications, such as travelling and parcel delivery. In this study, we consider the STSP in real-world road maps, i.e., given a road map with a set of service points or intersections, selection of the representative vertices of a graph G to determine a closed route so a vehicle can visit each service point at least once with the minimum cost. It has been theoretically proved that the STSP and its ancestor, travelling salesman problem, are NP-hard. However, it has been empirically demonstrated that they can be solved efficiently on planar graphs with small branchwidths. Therefore, we propose a branch decomposition-based extraction algorithm for the STSP in planar graphs. Our algorithm can solve the STSP in a planar graph G with a time complexity of 2O(bw(G))nO(1), where bw(G) is the branchwidth of G. In the case of planar graphs with a small branchwidth, our algorithm performs efficiently. We evaluated our algorithm by applying it to real-world urban road maps, which demonstrated the suitability of our method for real-world applications. © 2016 Elsevier Inc.
  Accession Number: 20163902852169
• Record 52 of
  
  Title:The effect analysis of conic coefficient error based on data measured from Talysurf and simulation of Zernike coefficients
  
  Author(s):Kai, Jiang(1); Liu, Kai(1); Song, Chong(1); Peng, Qiu(1); Peng, Wang(1); Li, Gang(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9684  Issue:   DOI: 10.1117/12.2242968  Published: 2016  
  Abstract:Derivation of the conic coefficient error of practical aspheric optic surface is quite significant to aspheric machining accuracy, optical system imaging quality analysis and decomposition analysis of optical lenses. The primary mirror of R-c telescope system was tested by Taylor Hobson Talysurf. The practical surface was fitted using Zernike polynomials based on the date measured from Talysurf. Though taking the Zernike coefficients into the optical system, the effect of the aberration which was brought by optical machining to the optical system imaging quality was obtained. The analysis shows that the spherical aberration was brought into the optical system because of the figure error of the primary mirror. And the value of the spherical aberration was same to the practical alignment result. Then the conicoid aspherical degree of the primary mirror was tested by the Talysurf. The machining deviation of the conic coefficient was gotten though comparing the conicoid aspherical degree of the practical primary mirror with that of the perfect primary mirror. The practical conic coefficient was calculated by the deviation. Taking the practical conic coefficient into the R-c telescope system, the degradation of the optical system imaging quality was known. Also the spherical aberration was brought into the optical system. Experimental results show that the value of the spherical aberration analyzed by the two methods is same and consist with the practical alignment result. That is to say that the conic coefficient changed due to machining error of the conicoid aspherical degree. Because of the change the spherical aberration was attached to primary mirror. And which caused the optical system imaging quality declined. Finally, corrector was designed to balance the spherical aberration of the primary mirror. Ensure that the optical system imaging quality meet the requirement. © 2016 SPIE.
  Accession Number: 20165003113966
• Record 53 of
  
  Title:Planar waveguides in neodymium-doped calcium niobium gallium garnet crystals produced by proton implantation
  
  Author(s):Liu, Chun-Xiao(1); Chen, Meng(1); Fu, Li-Li(2); Zheng, Rui-Lin(1); Guo, Hai-Tao(3); Zhou, Zhi-Guang(3); Li, Wei-Nan(3); Lin, She-Bao(4); Wei, Wei(1)
  Source: Chinese Physics B  Volume: 25  Issue: 4  DOI: 10.1088/1674-1056/25/4/044211  Published: April 2016  
  Abstract:In this work, the fabrication and optical properties of a planar waveguide in a neodymium-doped calcium niobium gallium garnet (Nd:CNGG) crystal are reported. The waveguide is produced by proton (H+) implantation at 480 keV and a fluence of 1.01017 ions/cm2. The prism-coupling measurement is performed to obtain the dark mode of the waveguide at a wavelength of 632.8 nm. The reflectivity calculation method (RCM) is used to reconstruct the refractive index profile. The finite-difference beam propagation method (FD-BPM) is employed to calculate the guided mode profile of the waveguide. The stopping and range of ions in matter 2010 (SRIM 2010) code is used to simulate the damage profile induced by the ion implantation. The experimental and theoretical results indicate that the waveguide can confine the light propagation. © 2016 Chinese Physical Society and IOP Publishing Ltd.
  Accession Number: 20161702289073
• Record 54 of
  
  Title:Research on photodiode detector-based spatial transient light detection and processing system
  
  Author(s):Liu, Meiying(1); Wang, Hu(1); Liu, Yang(1); Zhao, Hui(1); Nan, Meng(2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 10156  Issue:   DOI: 10.1117/12.2246843  Published: 2016  
  Abstract:In order to realize real-Time signal identification and processing of spatial transient light, the features and the energy of the captured target light signal are first described and quantitatively calculated. Considering that the transient light signal has random occurrence, a short duration and an evident beginning and ending, a photodiode detector based spatial transient light detection and processing system is proposed and designed in this paper. This system has a large field of view and is used to realize non-imaging energy detection of random, transient and weak point target under complex background of spatial environment. Weak signal extraction under strong background is difficult. In this paper, considering that the background signal changes slowly and the target signal changes quickly, filter is adopted for signal's background subtraction. A variable speed sampling is realized by the way of sampling data points with a gradually increased interval. The two dilemmas that real-Time processing of large amount of data and power consumption required by the large amount of data needed to be stored are solved. The test results with self-made simulative signal demonstrate the effectiveness of the design scheme. The practical system could be operated reliably. The detection and processing of the target signal under the strong sunlight background was realized. The results indicate that the system can realize real-Time detection of target signal's characteristic waveform and monitor the system working parameters. The prototype design could be used in a variety of engineering applications. © 2016 SPIE.
  Accession Number: 20165103154182
• Record 55 of
  
  Title:Resolution allocation and budget of the Lyman-α ultraviolet telescope
  
  Author(s):Kai, Jiang(1); Jiang, Bo(1); Liu, Kai(1); Yan, Peipei(1); Duan, Jing(1); Shan, Qiusha(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 10154  Issue:   DOI: 10.1117/12.2246786  Published: 2016  
  Abstract:As a high-resolution imaging instrument, angular resolution is the most important index of Lyman-α ultraviolet telescope. In this paper a new allocation and budget method is introduced. An resolution error allocation of surface roughness, figure error and alignment error was developed early in the program. And the allocation was used to guide the design. Though testing the surface roughness and figure error in visible light, the variation of diffraction encircled energy can be obtained by non-sequence model and Zernike coefficients brought into optical design software. The numerical results show that the effective RMS surface roughness of primary and secondary mirrors are 0.49nm and 0.40nm in the spatial frequency from 1/D (D is the diameter of the mirror) to 1/λ (λ is an incident wavelength). And the effects of the surface roughness are both less than 0.1″. The figure error of the primary and secondary mirrors are 0.009λ and 0.007λ (Λ=632.8nm). The resolution errors which were brought by the figure error are 0.33″ and 0.16″. Then the effect of alignment error on angular resolution was gotten by testing visual resolution. Finally the angular resolution in ultraviolet band can be calculated. The focal length of Lyman-α ultraviolet telescope is 2000mm and the pixel size of detector is 14μm. So the pixel resolution is 1.4″. Experimental results show that the angular resolution of Lyman-α ultraviolet telescope is 0.59″, which is approached to the estimate and meet the requirement. © 2016 SPIE.
  Accession Number: 20170503310094
• Record 56 of
  
  Title:Joint content replication and request routing for social video distribution over cloud CDN: A community clustering method
  
  Author(s):Hu, Han(1); Wen, Yonggang(1); Chua, Tat-Seng(2); Huang, Jian(3); Zhu, Wenwu(4); Li, Xuelong(5)
  Source: IEEE Transactions on Circuits and Systems for Video Technology  Volume: 26  Issue: 7  DOI: 10.1109/TCSVT.2015.2455712  Published: July 2016  
  Abstract:The increasing popularity of online social networks (OSNs) has been transforming the dissemination pattern of social video contents. We can utilize the social information propagation pattern to improve the efficiency of social video distribution. In this paper, motivated by the social community classification, we present a social video replication and user request dispatching mechanism in the cloud content delivery network architecture to reduce the system operational cost, while guaranteeing the averaged service latency. Specifically, we first present a community classification method that clusters social users with social relationships, close geolocations, and similar video watching interests into various communities. Then, we conduct a large-scale measurement on a real OSN system to study the diversities of social video propagation and the effectiveness of our communities on smoothing the diversity. Finally, we propose the community-based video replication and request dispatching strategy and formulate it as a constrained optimization problem. Based on a stochastic optimization framework, we derive an online solution and rigorously prove the optimality. We evaluate our algorithm on a real trace under realistic settings and demonstrate that our algorithm can reduce the monetary cost by 30% against traditional approaches with the same service latency. © 2015 IEEE.
  Accession Number: 20162902598036
• Record 57 of
  
  Title:Image Categorization by Learning a Propagated Graphlet Path
  
  Author(s):Zhang, Luming(1); Hong, Richang(1); Gao, Yue(2); Ji, Rongrong(3); Dai, Qionghai(2); Li, Xuelong(4)
  Source: IEEE Transactions on Neural Networks and Learning Systems  Volume: 27  Issue: 3  DOI: 10.1109/TNNLS.2015.2444417  Published: March 2016  
  Abstract:Spatial pyramid matching is a standard architecture for categorical image retrieval. However, its performance is largely limited by the prespecified rectangular spatial regions when pooling local descriptors. In this paper, we propose to learn object-shaped and directional receptive fields for image categorization. In particular, different objects in an image are seamlessly constructed by superpixels, while the direction captures human gaze shifting path. By generating a number of superpixels in each image, we construct graphlets to describe different objects. They function as the object-shaped receptive fields for image comparison. Due to the huge number of graphlets in an image, a saliency-guided graphlet selection algorithm is proposed. A manifold embedding algorithm encodes graphlets with the semantics of training image tags. Then, we derive a manifold propagation to calculate the postembedding graphlets by leveraging visual saliency maps. The sequentially propagated graphlets constitute a path that mimics human gaze shifting. Finally, we use the learned graphlet path as receptive fields for local image descriptor pooling. The local descriptors from similar receptive fields of pairwise images more significantly contribute to the final image kernel. Thorough experiments demonstrate the advantage of our approach. © 2012 IEEE.
  Accession Number: 20155101697007
• Record 58 of
  
  Title:Study on the deep neural network of intelligent image detection and the improvement of elastic momentum on image recognition
  
  Author(s):Yue, Qi(1,2); Ma, Caiwen(1)
  Source: Journal of Computational and Theoretical Nanoscience  Volume: 13  Issue: 5  DOI: 10.1166/jctn.2016.4994  Published: May 2016  
  Abstract:Aiming at deep neural network of intelligent image detection, this paper proposes parameter learning method of exponential elastic momentum, applies this method in the pedestrian detection, the detection accuracy is 97.8%. In addition, this algorithm is compared with adaptive momentum methods, standard momentum gradient method and elastic momentum, and the accuracy obtained through this method increases by 2.6% and 6.5% on average respectively. Copyright © 2016 American Scientific Publishers All rights reserved.
  Accession Number: 20163202702043
• Record 59 of
  
  Title:Shape-Constrained Sparse and Low-Rank Decomposition for Auroral Substorm Detection
  
  Author(s):Yang, Xi(1); Gao, Xinbo(1); Tao, Dacheng(2); Li, Xuelong(3); Han, Bing(4); Li, Jie(4)
  Source: IEEE Transactions on Neural Networks and Learning Systems  Volume: 27  Issue: 1  DOI: 10.1109/TNNLS.2015.2411613  Published: January 2016  
  Abstract:An auroral substorm is an important geophysical phenomenon that reflects the interaction between the solar wind and the Earth's magnetosphere. Detecting substorms is of practical significance in order to prevent disruption to communication and global positioning systems. However, existing detection methods can be inaccurate or require time-consuming manual analysis and are therefore impractical for large-scale data sets. In this paper, we propose an automatic auroral substorm detection method based on a shape-constrained sparse and low-rank decomposition (SCSLD) framework. Our method automatically detects real substorm onsets in large-scale aurora sequences, which overcomes the limitations of manual detection. To reduce noise interference inherent in current SLD methods, we introduce a shape constraint to force the noise to be assigned to the low-rank part (stationary background), thus ensuring the accuracy of the sparse part (moving object) and improving the performance. Experiments conducted on aurora sequences in solar cycle 23 (1996-2008) show that the proposed SCSLD method achieves good performance for motion analysis of aurora sequences. Moreover, the obtained results are highly consistent with manual analysis, suggesting that the proposed automatic method is useful and effective in practice. © 2015 IEEE.
  Accession Number: 20151400711792
• Record 60 of
  
  Title:Optimization design of C-T imaging spectrometer based on the tilt field len
  
  Author(s):Hao, Ai-Hua(1); Hu, Bing-Liang(2); Li, Li-Bo(2); Li, Yun(2)
  Source: Guangzi Xuebao/Acta Photonica Sinica  Volume: 45  Issue: 4  DOI: 10.3788/gzxb20164504.0412002  Published: April 1, 2016  
  Abstract:Residual smile limits the application of Czerny-Turner plane grating spectrometer in the imaging spectrometer. In this paper, different from the traditional method of smile correction based on prism, a method is proposed, which is based on the tilt field mirror. The field curve is corrected, at the same time, the different wavelength slit images in different regions of the field lens are corrected, and the other optical properties of the system are not changed. The Czerny Turner imaging spectrometer slit is 7.8 millimeter length and 0.016 millimeter wide, spectral range from 0.31 to 0.5 microns, spectral resolution 0.4 nanometers, the object focal length 70 millimeters, 1:1 magnification. Optimization design results, MTF over 0.8, RMS spot radius less than 9 microns, the relative smile less than 0.2%, meet the design requirements. It shows that the method can be used in the systems with weak signal and low transmittance optical glass. © 2016, Chinese Optical Society. All right reserved.
  Accession Number: 20161902355993