2016

• Record 37 of
  
  Title:Congested scene classification via efficient unsupervised feature learning and density estimation
  
  Author(s):Yuan, Yuan(1); Wan, Jia(2); Wang, Qi(2)
  Source: Pattern Recognition  Volume: 56  Issue:   DOI: 10.1016/j.patcog.2016.03.020  Published: August 1, 2016  
  Abstract:An unsupervised learning algorithm with density information considered is proposed for congested scene classification. Though many works have been proposed to address general scene classification during the past years, congested scene classification is not adequately studied yet. In this paper, an efficient unsupervised feature learning approach with density information encoded is proposed to solve this problem. Based on spherical k-means, a feature selection process is proposed to eliminate the learned noisy features. Then, local density information which better reflects the crowdedness of a scene is encoded by a novel feature pooling strategy. The proposed method is evaluated on the assembled congested scene data set and UIUC-sports data set, and intensive comparative experiments justify the effectiveness of the proposed approach. © 2016 Elsevier Ltd
  Accession Number: 20162802589173
• Record 38 of
  
  Title:Incrementally Detecting Moving Objects in Video with Sparsity and Connectivity
  
  Author(s):Pan, Jing(1,2); Li, Xiaoli(1); Li, Xuelong(3); Pang, Yanwei(1)
  Source: Cognitive Computation  Volume: 8  Issue: 3  DOI: 10.1007/s12559-015-9373-5  Published: June 1, 2016  
  Abstract:Moving object detection is crucial for cognitive vision-based robot tasks. However, due to noise, dynamic background, variations in illumination, and high frame rate, it is a challenging task to robustly and efficiently detect moving objects in video using the clue of motion. State-of-the-art batch-based methods view a sequence of images as a whole and then model the background and foreground together with the constraints of foreground sparsity and connectivity (smoothness) in a unified framework. But the efficiency of the batch-based methods is very low. State-of-the-art incremental methods model the background by a subspace whose bases are updated frame by frame. However, such incremental methods do not make full use of the foreground sparsity and connectivity. In this paper, we develop an incremental method for detecting moving objects in video. Compared to existing methods, the proposed method not only incrementally models the subspace for background reconstruction but also takes into account the sparsity and connectivity of the foreground. The optimization of the model is very efficient. Experimental results on nine public videos demonstrate that the proposed method is much efficient than the state-of-the-art batch methods and has higher F1-score than the state-of-the-art incremental methods. © 2015, Springer Science+Business Media New York.
  Accession Number: 20155301738771
• Record 39 of
  
  Title:An improved acute lymphoblastic leukemia image segmentation scheme based on HSV color space
  
  Author(s):Li, Yan(1,2); Zhu, Rui(1); Mi, Lei(1); Cao, Yihui(1,2); Yao, Di(3)
  Source: ICIC Express Letters, Part B: Applications  Volume: 7  Issue: 9  DOI:   Published: September 1, 2016  
  Abstract:In this paper, an improved HSV color space based Acute Lymphoblastic Leukemia (ALL) image segmentation scheme is proposed to improve accuracy of segmentation on digital microscope color images, especially for those taken in non-uniform background illumination conditions and by a microscope with different magnifications. The proposed method has several steps including contrast stretching, color space transformation, threshold segmentation, morphological operations and median filtering. Adopting two thresholds is the innovation of this work. For performance evaluation, 260 ALL blood cell images from ALL IDB2-a public and free available blood sample dataset are used. The experimental results show the proposed method gets a higher accuracy in segmenting both high- and low- contrast blood cell images than the original HSV color space based single threshold method, showing a better prospect in subsequent automatic acute lymphoblastic leukemia feature extraction and classification. © 2016 ICIC International.
  Accession Number: 20163702788355
• Record 40 of
  
  Title:Fuzzy bag of words for social image description
  
  Author(s):Li, Yanshan(1,2,3); Liu, Weiming(2); Huang, Qinghua(1); Li, Xuelong(4)
  Source: Multimedia Tools and Applications  Volume: 75  Issue: 3  DOI: 10.1007/s11042-014-2138-4  Published: February 1, 2016  
  Abstract:Rapid growth of social media resources brings huge challenges and opportunities for image description technologies. The performance of image description method directly affects the accuracy of image retrieval, image annotation and image recognition. Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between the words in the codebook and the features extracted from the images are actually ambiguous. As the Fuzzy Sets Theory (FST) is a powerful means for dealing with uncertainty efficiently, we utilize the FST to solve the problem caused by the ambiguity between the features and words. Accordingly, we propose a new type of BoW named as FBoW to describe images based on FST. Firstly, the features are extracted from the images. Secondly, k-means is utilized to learn the codebook. Thirdly, a fuzzy membership function is designed to measure the similarity between the features and words. The optimal parameters of the fuzzy membership function are obtained by using a Genetic Algorithm (GA). The histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed FBoW outperforms traditional BoW for social image description. © 2014, Springer Science+Business Media New York.
  Accession Number: 20143600025676
• Record 41 of
  
  Title:Carbon-implanted monomode waveguides in magneto-optical glasses for waveguide isolators
  
  Author(s):Liu, Chun-Xiao(1); Fu, Li-Li(2); Zhang, Liao-Lin(1); Guo, Hai-Tao(3); Li, Wei-Nan(3); Lin, She-Bao(4); Wei, Wei(1)
  Source: Applied Physics A: Materials Science and Processing  Volume: 122  Issue: 2  DOI: 10.1007/s00339-016-9631-8  Published: February 1, 2016  
  Abstract:Tb3+-doped aluminum borosilicate glasses are important magneto-optical materials for optical isolators. Optical waveguides are basic components in integrated photonics. By using the ion implantation technique, optical guiding structures can be produced in Tb3+-doped aluminum borosilicate glasses, and miniaturized waveguide isolators can be realized. In this paper, planar waveguides have been fabricated in Tb3+-doped aluminum borosilicate glasses by (6.0 + 5.5) MeV carbon ion implantation at doses of (8.0 + 4.0) × 1013 ions/cm2. The optical properties of optical waveguides are measured by equipments of prism coupling and end-face coupling systems. They are also analyzed by simulation programs of intensity calculation method and beam propagation method. The waveguides with good optical performances suggest potential applications on fabrication of waveguide isolators in Tb3+-doped aluminum borosilicate glasses. © 2016, Springer-Verlag Berlin Heidelberg.
  Accession Number: 20160501873410
• Record 42 of
  
  Title:Two side liquid-cooled and passively Q-switched disk oscillator with nanosheets in flowing CCl4
  
  Author(s):Nie, Rongzhi(1,2); She, Jiangbo(1); Li, Dongdong(4); Li, Fuli(2); Peng, Bo(1,3)
  Source: Applied Physics B: Lasers and Optics  Volume: 122  Issue: 9  DOI: 10.1007/s00340-016-6519-4  Published: September 1, 2016  
  Abstract:A passively Q-switched and two side liquid-cooled Nd:YAG disk oscillator is demonstrated, which is operated at a pump pulse width of 300 μs and a pump repetition rate of 10 Hz. The coolant flows over the two large surfaces of the disk and will be passed through by laser beam, so it can also serve as a saturable absorber. For the unmodulated laser, the pure CCl4 was employed as coolant and a plane output mirror of 15 % transmission was employed. The maximum output energy of 795 mJ is realized corresponding to the optical–optical efficiency of 27.4 % and the slope efficiency of 30 %; for the graphene Q-switched laser, the CCl4 with graphene nanosheets was employed as coolant and a plane output mirror of 40 % transmission was employed. The maximum output energy of 376 mJ is realized corresponding to the optical–optical efficiency of 13 % and the slope efficiency of 18 %. The maximum average Q-switching repetition rate is 385 kHz, and the minimum average pulse width is 116 ns; for the MoS2 Q-switched laser, the CCl4 with MoS2 nanosheets was employed as coolant and a plane output mirror of 30 % transmission was employed. The maximum output energy of 486 mJ is realized corresponding to the optical–optical efficiency of 17 % and the slope efficiency of 22 %.The maximum average Q-switching repetition rate is 470 kHz, and the minimum average pulse width is 137 ns. © 2016, Springer-Verlag Berlin Heidelberg.
  Accession Number: 20163602781445
• Record 43 of
  
  Title:Subspace clustering by capped l1 norm
  
  Author(s):Lu, Quanmao(1,2); Li, Xuelong(1); Dong, Yongsheng(1); Tao, Dacheng(3)
  Source: Communications in Computer and Information Science  Volume: 662  Issue:   DOI: 10.1007/978-981-10-3002-4_54  Published: 2016  
  Abstract:Subspace clustering, as an important clustering problem, has drawn much attention in recent years. State-of-the-art methods generally try to design an efficient model to regularize the coefficient matrix while ignore the influence of the noise model on subspace clustering. However, the real data are always contaminated by the noise and the corresponding subspace structures are likely to be corrupted. In order to solve this problem, we propose a novel subspace clustering algorithm by employing capped l1norm to deal with the noise. Consequently, the noise term with large error can be penalized by the proposed method. So it is more robust to the noise. Furthermore, the grouping effect of our method is theoretically proved, which means highly correlated points can be grouped together. Finally, the experimental results on two real databases show that our method outperforms state-of-the-art methods. © Springer Nature Singapore Pte Ltd. 2016.
  Accession Number: 20164603024510
• Record 44 of
  
  Title:Experimental verification of the HERD prototype at CERN SPS
  
  Author(s):Dong, Yongwei(1,3); Quan, Zheng(3,4); Wang, Junjing(3,4); Xu, Ming(1,3); Albergo, Sebastiano(15); Ambroglini, Filippo(5); Ambrosi, Giovanni(5); Azzarello, Philipp(6); Bai, Yonglin(7); Bao, Tianwei(1,3); Baldini, Luca(16); Battiston, Roberto(13); Bernardini, Paolo(8); Chen, Zhen(7); D'alessandro, Raffaello(14); Duranti, Matteo(5); D'urso, Domenico(5); Fusco, Piergiorgio(9); Gao, Jiarui(7); Gao, Xiaohui(7); Gargano, Fabio(9); Giglietto, Nicola(9); Hu, Bingliang(7); Li, Ran(7); Li, Yong(7); Liu, Xin(1,3); Loparco, Francesco(9); Lu, Junguang(2,3); Marsella, Giovanni(8); Mazziotta, Mario N.(9); De Mitri, Ivan(8); Mori, Nicola(10); Orsi, Silvio(17); Oscar, Adriani(10); Pearce, Mark(18); Pohl, Martin(6); Ryde, Felix(18); Shi, Dalian(7); Spillantini, Piero(10); Su, Meng(11,12); Sun, Xin(7); Surdo, Antonio(8); Walter, Roland(17); Wang, Bo(7); Wang, Le(7); Wang, Ruijie(1,3); Wang, Zhigang(2,3); Wu, Bobing(1,3); Wu, Xin(6); Yan, Peng(7); Zhang, Li(1,3); Zhang, Shuangnan(1,3)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9905  Issue:   DOI: 10.1117/12.2231804  Published: 2016  
  Abstract:The High Energy cosmic-Radiation Detection (HERD) facility is one of several space astronomy payloads of the cosmic light house program onboard China's Space Station, which is planned for operation starting around 2020 for about 10 years. Beam test with a HERD prototype, to verify the HERD specifications and the reading out method of wavelength shifting fiber and image intensified CCD, was taken at CERN SPS in November, 2015. The prototype is composed of an array of 5∗5∗10 LYSO crystals, which is 1/40th of the scale of HERD calorimeter. Experimental results on the performances of the calorimeter are discussed. © 2016 SPIE.
  Accession Number: 20165003127633
• Record 45 of
  
  Title:Phase-sensitive amplification with net gain in low-loss integrated waveguides
  
  Author(s):Zhang, Y.(1); Wu, J.(1); Reimer, C.(1); Roztocki, P.(1); Wetzel, B.(1,2); Little, Brent E.(3); Chu, S.T.(4); Moss, D.J.(5); Kues, M.(1); Morandotti, R.(1,6)
  Source: Optics InfoBase Conference Papers  Volume:   Issue:   DOI: 10.1364/FIO.2016.JW4A.152  Published: 2016  
  Abstract:We demonstrate phase sensitive amplification based on pump-degenerate four-wave mixing in highly nonlinear low-loss waveguides. A net gain of 6 dB (including propagation losses) is achieved, with an extinction ratio of 16 dB. © OSA 2016.
  Accession Number: 20171503547866
• Record 46 of
  
  Title:Research on broadband spectral imaging spectrometer based on CDP
  
  Author(s):Zhang, Zhou-Feng(1,2,3); Hu, Bing-Liang(1); Yin, Qin-Ye(2); Gao, Xiao-Hui(1)
  Source: Guang Pu Xue Yu Guang Pu Fen Xi/Spectroscopy and Spectral Analysis  Volume: 36  Issue: 7  DOI: 10.3964/j.issn.1000-0593(2016)07-2284-03  Published: July 1, 2016  
  Abstract:In order to satisfy the application requirements of real-time spectral imaging for moving targets, we design a static, snapshot imaging spectrometer based on a CDP(crossed dispersion prism). The spectral imaging principle is studied, and an optical system of broadband spectral imaging spectrometer is designed according to this principle. Imaging spectrometer consists of a CDP, an imaging lens and a detector, with ±2°field of view, spectral coverage from 0.6 to 5.0 μm, The results show that this instrument has a better ability to detect spectral from 0.6 to 5.0 μm while the average spectral resolution is 20 nm. The technology for dynamic target real-time spectral imaging provides a new technical way. It has a great potential for detecting, locating and identifying unknown energetic events in real-time. © 2016, Peking University Press. All right reserved.
  Accession Number: 20163002637923
• Record 47 of
  
  Title:The lightweight structure design of a CFRP mirror
  
  Author(s):Ding, Jiaoteng(1); Xu, Liang(1); Ma, Zhen(1); Xie, Yongjie(1); Luo, Yao(2); Wang, Yongjie(1); Pang, Zhihai(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9683  Issue:   DOI: 10.1117/12.2243440  Published: 2016  
  Abstract:The advantage of Carbon Fiber Reinforced Polymer (CFRP) is obvious as a common space material for low density, low thermal expansion coefficient and high specific stiffness characteristics, it is the ideal material choice for space optical reflector. Mirror structure with honeycomb can achieve high rates of lightweight, as well as high specific stiffness. For Φ300mm CFRP mirror, accounting of the actual process properties of CFRP, mirror panels laminated based on thermal stability design, honeycomb fabricated using one innovative inlaying-grafting design method. Finally, lightweight structure design of the CFRP primary mirror completed, the thermal stability result of the Φ300mm CFRP mirror achieved is 10nm°C. © 2016 SPIE.
  Accession Number: 20164903096458
• Record 48 of
  
  Title:Image fusion based on group sparse representation
  
  Author(s):Yin, Fei(1,2); Gao, Wei(1); Song, Zongxi(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 10033  Issue:   DOI: 10.1117/12.2244879  Published: 2016  
  Abstract:Sparse representation based image fusion has been widely studied recently. However, it's not popular in some fields for the high time complexity. In this paper, a new image fusion method based on group sparse representation is proposed to overcome this problem. The K-SVD method is utilized to get the sparse representation of the source images. Therefore, it is necessary to find the best size of the group according to its property about time consuming. And there is no need to sparse all the patches once but to sparse some groups simultaneously. Because every group image vectors sparse representation is unique from the others, using the parallel-processing strategy can reduce the time badly. Besides, all dictionaries are learned from local source image vectors, so the quality of the results fused by the group sparse representation method will be better than those fused by the normal sparse representation methods. Compared with four types of state-of-the-art algorithms, the proposed method has the excellent fusion performance in experiments. © 2016 SPIE.
  Accession Number: 20164903101655