2013

• Record 25 of
  
  Title:Design of Gires-Tournois mirrors used for the dispersion compensation in femtosecond lasers
  
  Author(s):Liao, Chun-Yan(1); Qin, Jun-Jun(2); Shao, Jian-Da(3); Cheng, Guang-Hua(2); Fan, Zheng-Xiu(3); Hu, Man-Li(1)
  Source: Guangzi Xuebao/Acta Photonica Sinica  Volume: 42  Issue: 8  DOI: 10.3788/gzxb20134208.0967  Published: August 2013  
  Abstract:Basic structure of Gires-Tournois mirror is described and the dispersion performance is calculated. The factors affecting the performance of the Gires-Tournois mirrors are discussed. The results show that the layer number of high reflector affects the reflectance of the Gires-Tournois mirrors but the thickness of the Gires-Tournois cavity and the layer number of the top reflector affect the dispersion performance of the Gires-Tournois mirrors; to achieve good design performance, the layer number of high reflector, the thickness of the Gires-Tournois cavity and the layer number of the top reflector are selected to be 40~60, λ/2 or λ and less than 5.
  Accession Number: 20134216860597
• Record 26 of
  
  Title:Electromagnetic resonance tunneling in a single-negative sandwich structure
  
  Author(s):Kang, Yongqiang(1,2,3); Zhang, Chunmin(1); Gao, Peng(1); Ren, Wenyi(1)
  Source: Journal of Modern Optics  Volume: 60  Issue: 13  DOI: 10.1080/09500340.2013.827251  Published: July 1, 2013  
  Abstract:The electromagnetic wave tunneling phenomenon in a sandwich structure consisting of epsilon-negative (ENG), mu-negative (MNG), and epsilon-negative (ENG) media was investigated. Merging of resonance tunneling modes is demonstrated when the conjugate matched trilayer condition is satisfied. The resonance frequency is found to be independent of the thickness ratio of the matched trilayer structure. The resonance tunneling possesses particular angular-dependent and polarization-free properties. The electric fields corresponding to the frequencies of the resonance modes are found to be strongly localized at just one interface with low transmittance. The possible influence on resonance tunneling due to the losses from the single-negative materials is also investigated. © 2013 Taylor and Francis.
  Accession Number: 20134216859892
• Record 27 of
  
  Title:Effective medium theory for two-dimensional random media composed of core-shell cylinders
  
  Author(s):Zhang, Hao(1,2); Shen, Yongqiang(1); Xu, Yuchen(1); Zhu, Heyuan(1); Lei, Ming(2); Zhang, Xiangchao(1); Xu, Min(1)
  Source: Optics Communications  Volume: 306  Issue:   DOI: 10.1016/j.optcom.2013.05.027  Published: 2013  
  Abstract:In this paper, based on the generalized coated coherent potential approximation method, we derive the mathematical formulae, for the extended effective medium theory, to investigate the optical properties of disordered media composed of core-shell cylinders. The effective indices of such media are obtained in the long-wavelength limit and in the Mie-scattering region. Moreover, we use this method to study optical properties of random media composed of core-shell cylinders with the core layer consisting of epsilon-less-than-one material. © 2013 Elsevier B.V. All rights reserved.
  Accession Number: 20132716458309
• Record 28 of
  
  Title:Object or background: Whose call is it in complicated scene classification?
  
  Author(s):Mou, Lichao(1,2); Lu, Xiaoqiang(1); Yuan, Yuan(1)
  Source: 2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings  Volume:   Issue:   DOI: 10.1109/ChinaSIP.2013.6625399  Published: 2013  
  Abstract:Scene semantic parsing is a challenging problem in the field of computer vision. Most approaches exploit low-level features to describe the whole scene. However, there is a large semantic gap between low-level features and high-level scene semantic. In this paper, a scene classification approach is proposed by exploiting semantic objects/materials of the background to reduce the semantic gap. The proposed approach can be divided three steps: First we construct two high-level semantic features (BCFs and BSLFs). Second, we design an approach to learn the prior probability of the Bayesian Networks from these two semantic features of training images. Finally, Bayesian Networks is used to achieve the goal of scene classification. Experimental results show that our approach achieves state-of-the-art performance on the task of scene classification compare with other approaches. © 2013 IEEE.
  Accession Number: 20135017076778
• Record 29 of
  
  Title:Mixture gradient detector for subpixel detection
  
  Author(s):Huang, Zihan(1,2); Yuan, Yuan(1); Lu, Xiaoqiang(1)
  Source: 2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings  Volume:   Issue:   DOI: 10.1109/ChinaSIP.2013.6625423  Published: 2013  
  Abstract:Subpixel detection is an important but difficult problem in hy-perspectral image. Due to the small size of the target, only spectral information can be used for detection. Many algorithms have been proposed to reduce this problem, and most of them assume that the distribution of hyperspectral image is multinormal. However, this assumption may not be an appropriate description of the distribution in hyperspectral image. After carefully study the distribution of hyperspectral image, it is concluded that the gradient of noise should also be considered. In this paper a new model is proposed, which assumes that gradient of the noise also follow Gaussian distribution. Based on the given model, two detectors, mixture gradient structured detector (MGSD) and mixture gradient unstructured detector (MGUD) are proposed. The proposed detectors take advantage of the new model, in which the distribution of noise is more accordant with the practical situation. Experiment results demonstrate that in general the proposed detectors perform better than state-of-the-art. © 2013 IEEE.
  Accession Number: 20135017076802
• Record 30 of
  
  Title:3D prostate MR image segmentation: A multi-task approach
  
  Author(s):Liu, Yin(1,2); Yuan, Yuan(1); Lu, Xiaoqiang(1)
  Source: 2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings  Volume:   Issue:   DOI: 10.1109/ChinaSIP.2013.6625326  Published: 2013  
  Abstract:Multi-atlas based approaches are effective for the medical image segmentation. The strategy of assigning weights for the atlases is critically important to the segmentation performance. Previous works either assign weights on the image level or assign weights of different regions independently, i.e., they can't employ the uniqueness of each region and the connectivity among different regions simultaneously. In this paper, a multi-task approach is proposed to reduce this drawback. To exploit the unique characteristic of each region, learning the segmentation result for each region is viewed as a single task. The weighted voting decision for each regions are made individually. To model the connectivity among different regions or tasks, a norm regularization term is introduced to refine the segmentation results made by each individual tasks. By this way, the proposed approach simultaneously exploits the unique character of each region and the connectivity among them. The proposed approach is tested on 60 3D prostate magnetic resonance (MR) images from 60 patients. Experiment results show that the proposed approach is comparative to or even superior to the state-of-the-art approaches for the prostate segmentation. © 2013 IEEE.
  Accession Number: 20135017076706
• Record 31 of
  
  Title:Prostate segmentation in MR images using discriminant boundary features
  
  Author(s):Yang, Meijuan(1); Li, Xuelong(1); Turkbey, Baris(2); Choyke, Peter L.(2); Yan, Pingkun(1)
  Source: IEEE Transactions on Biomedical Engineering  Volume: 60  Issue: 2  DOI: 10.1109/TBME.2012.2228644  Published: 2013  
  Abstract:Segmentation of the prostate in magnetic resonance image has become more in need for its assistance to diagnosis and surgical planning of prostate carcinoma. Due to the natural variability of anatomical structures, statistical shape model has been widely applied in medical image segmentation. Robust and distinctive local features are critical for statistical shape model to achieve accurate segmentation results. The scale invariant feature transformation (SIFT) has been employed to capture the information of the local patch surrounding the boundary. However, when SIFT feature being used for segmentation, the scale and variance are not specified with the location of the point of interest. To deal with it, the discriminant analysis in machine learning is introduced to measure the distinctiveness of the learned SIFT features for each landmark directly and to make the scale and variance adaptive to the locations. As the gray values and gradients vary significantly over the boundary of the prostate, separate appearance descriptors are built for each landmark and then optimized. After that, a two stage coarse-to-fine segmentation approach is carried out by incorporating the local shape variations. Finally, the experiments on prostate segmentation from MR image are conducted to verify the efficiency of the proposed algorithms. © 1964-2012 IEEE.
  Accession Number: 20130415939973
• Record 32 of
  
  Title:Data-dependent semi-supervised hyperspectral image classification
  
  Author(s):Lv, Haobo(1,2); Lu, Xiaoqiang(1); Yuan, Yuan(1)
  Source: 2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings  Volume:   Issue:   DOI: 10.1109/ChinaSIP.2013.6625425  Published: 2013  
  Abstract:Hyperspectral imagery provides more powerful information than multispectral remote sensing data. However, when hyperspectral data is used for classification task, the highdimension features often lead to ill-conditioned problems, such as the Hughes phenomenon. To tackle this problem, various supervised dimensional reduction methods are proposed. However, these methods only exploit the labeled training data and ignore the huge unlabelled data. To utilize the unlabelled data space structure information in dimension reduction, a method is proposed as Data-dependent semi-supervised (DDSS). The proposed method exploits the space structure of labeled data and unlabelled data jointly to reduce the dimensionality of the image cures. Experimental results show that this method significantly outperforms the state-of-the-art dimension reduction methods for classification and denoising. © 2013 IEEE.
  Accession Number: 20135017076804
• Record 33 of
  
  Title:Opto-digital image encryption by using Baker mapping and 1-D fractional Fourier transform
  
  Author(s):Liu, Zhengjun(1,2); Li, She(3); Liu, Wei(3); Liu, Shutian(3)
  Source: Optics and Lasers in Engineering  Volume: 51  Issue: 3  DOI: 10.1016/j.optlaseng.2012.10.008  Published: March 2013  
  Abstract:We present an optical encryption method based on the Baker mapping in one-dimensional fractional Fourier transform (1D FrFT) domains. A thin cylinder lens is controlled by computer for implementing 1D FrFT at horizontal direction or vertical direction. The Baker mapping is introduced to scramble the amplitude distribution of complex function. The amplitude and phase of the output of encryption system are regarded as encrypted image and key. Numerical simulation has been performed for testing the validity of this encryption scheme. © 2012 Elsevier Ltd.
  Accession Number: 20125015777294
• Record 34 of
  
  Title:Topographic NMF for data representation
  
  Author(s):Xiao, Yanhui(1,2); Zhu, Zhenfeng(1,2); Zhao, Yao(3); Wei, Yunchao(1,2); Wei, Shikui(1,2); Li, Xuelong(4)
  Source: IEEE Transactions on Cybernetics  Volume: 44  Issue: 10  DOI: 10.1109/TCYB.2013.2294215  Published: October 1, 2014  
  Abstract:Nonnegative matrix factorization (NMF) is a useful technique to explore a parts-based representation by decomposing the original data matrix into a few parts-based basis vectors and encodings with nonnegative constraints. It has been widely used in image processing and pattern recognition tasks due to its psychological and physiological interpretation of natural data whose representation may be parts-based in human brain. However, the nonnegative constraint for matrix factorization is generally not sufficient to produce representations that are robust to local transformations. To overcome this problem, in this paper, we proposed a topographic NMF (TNMF), which imposes a topographic constraint on the encoding factor as a regularizer during matrix factorization. In essence, the topographic constraint is a two-layered network, which contains the square nonlinearity in the first layer and the square-root nonlinearity in the second layer. By pooling together the structure-correlated features belonging to the same hidden topic, the TNMF will force the encodings to be organized in a topographical map. Thus, the feature invariance can be promoted. Some experiments carried out on three standard datasets validate the effectiveness of our method in comparison to the state-of-the-art approaches. © 2013 IEEE.
  Accession Number: 20143900073586
• Record 35 of
  
  Title:Global structure constrained local shape prior estimation for medical image segmentation
  
  Author(s):Yan, Pingkun(1); Zhang, Wuxia(1); Turkbey, Baris(2); Choyke, Peter L.(2); Li, Xuelong(1)
  Source: Computer Vision and Image Understanding  Volume: 117  Issue: 9  DOI: 10.1016/j.cviu.2013.03.006  Published: 2013  
  Abstract:Organ shape plays an important role in clinical diagnosis, surgical planning and treatment evaluation. Shape modeling is a critical factor affecting the performance of deformable model based segmentation methods for organ shape extraction. In most existing works, shape modeling is completed in the original shape space, with the presence of outliers. In addition, the specificity of the patient was not taken into account. This paper proposes a novel target-oriented shape prior model to deal with these two problems in a unified framework. The proposed method measures the intrinsic similarity between the target shape and the training shapes on an embedded manifold by manifold learning techniques. With this approach, shapes in the training set can be selected according to their intrinsic similarity to the target image. With more accurate shape guidance, an optimized search is performed by a deformable model to minimize an energy functional for image segmentation, which is efficiently achieved by using dynamic programming. Our method has been validated on 2D prostate localization and 3D prostate segmentation in MRI scans. Compared to other existing methods, our proposed method exhibits better performance in both studies. © 2013 Elsevier Inc. All rights reserved.
  Accession Number: 20134216859393
• Record 36 of
  
  Title:Universal blind image quality assessment metrics via natural scene statistics and multiple kernel learning
  
  Author(s):Gao, Xinbo(1); Gao, Fei(1); Tao, Dacheng(2); Li, Xuelong(3)
  Source: IEEE Transactions on Neural Networks and Learning Systems  Volume: 24  Issue: 12  DOI: 10.1109/TNNLS.2013.2271356  Published: 2013  
  Abstract:Universal blind image quality assessment (IQA) metrics that can work for various distortions are of great importance for image processing systems, because neither ground truths are available nor the distortion types are aware all the time in practice. Existing state-of-the-art universal blind IQA algorithms are developed based on natural scene statistics (NSS). Although NSS-based metrics obtained promising performance, they have some limitations: 1) they use either the Gaussian scale mixture model or generalized Gaussian density to predict the nonGaussian marginal distribution of wavelet, Gabor, or discrete cosine transform coefficients. The prediction error makes the extracted features unable to reflect the change in nonGaussianity (NG) accurately. The existing algorithms use the joint statistical model and structural similarity to model the local dependency (LD). Although this LD essentially encodes the information redundancy in natural images, these models do not use information divergence to measure the LD. Although the exponential decay characteristic (EDC) represents the property of natural images that large/small wavelet coefficient magnitudes tend to be persistent across scales, which is highly correlated with image degradations, it has not been applied to the universal blind IQA metrics; and 2) all the universal blind IQA metrics use the same similarity measure for different features for learning the universal blind IQA metrics, though these features have different properties. To address the aforementioned problems, we propose to construct new universal blind quality indicators using all the three types of NSS, i.e., the NG, LD, and EDC, and incorporating the heterogeneous property of multiple kernel learning (MKL). By analyzing how different distortions affect these statistical properties, we present two universal blind quality assessment models, NSS global scheme and NSS two-step scheme. In the proposed metrics: 1) we exploit the NG of natural images using the original marginal distribution of wavelet coefficients; 2) we measure correlations between wavelet coefficients using mutual information defined in information theory; 3) we use features of EDC in universal blind image quality prediction directly; and 4) we introduce MKL to measure the similarity of different features using different kernels. Thorough experimental results on the Laboratory for Image and Video Engineering database II and the Tampere Image Database2008 demonstrate that both metrics are in remarkably high consistency with the human perception, and overwhelm representative universal blind algorithms as well as some standard full reference quality indexes for various types of distortions. © 2012 IEEE.
  Accession Number: 20134817019583