2015

• Record 349 of
  
  Title:Efficient Video Stitching Based on Fast Structure Deformation
  
  Author(s):Li, Jing(1); Xu, Wei(1); Zhang, Jianguo(2); Zhang, Maojun(1); Wang, Zhengming(3); Li, Xuelong(4)
  Source: IEEE Transactions on Cybernetics  Volume: 45  Issue: 12  DOI: 10.1109/TCYB.2014.2381774  Published: December 2015  
  Abstract:In computer vision, video stitching is a very challenging problem. In this paper, we proposed an efficient and effective wide-view video stitching method based on fast structure deformation that is capable of simultaneously achieving quality stitching and computational efficiency. For a group of synchronized frames, firstly, an effective double-seam selection scheme is designed to search two distinct but structurally corresponding seams in the two original images. The seam location of the previous frame is further considered to preserve the interframe consistency. Secondly, along the double seams, 1-D feature detection and matching is performed to capture the structural relationship between the two adjacent views. Thirdly, after feature matching, we propose an efficient algorithm to linearly propagate the deformation vectors to eliminate structure misalignment. At last, image intensity misalignment is corrected by rapid gradient fusion based on the successive over relaxation iteration (SORI) solver. A principled solution to the initialization of the SORI significantly reduced the number of iterations required. We have compared favorably our method with seven state-of-the-art image and video stitching algorithms as well as traditional ones. Experimental results show that our method outperforms the existing ones compared in terms of overall stitching quality and computational efficiency. © 2013 IEEE.
  Accession Number: 20150200417444
• Record 350 of
  
  Title:An improved image sharpness assessment method based on contrast sensitivity
  
  Author(s):Zhang, Li(1,2); Tian, Yan(1); Yin, Yili(1,2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9675  Issue:   DOI: 10.1117/12.2203096  Published: 2015  
  Abstract:An image sharpness assessment method based on the property of Contrast Sensitivity Function (CSF) was proposed to realize the sharpness assessment of unfocused image. Firstly, image was performed the two-dimensional Discrete Fourier Transform (DFT), and intermediate frequency coefficients and high frequency coefficients are divided into two parts respectively. Secondly the four parts were performed the inverse Discrete Fourier Transform (IDFT) to obtain subimages. Thirdly, using Range Function evaluates the four sub-image sharpness value. Finally, the image sharpness is obtained through the weighted sum of the sub-image sharpness value. In order to comply with the CSF characteristics, weighting factor is setting based on the Contrast Sensitivity Function. The new algorithm and four typical evaluation algorithm: Fourier, Range, Variance and Wavelet are evaluated based on the six quantitative evaluation index, which include the width of steep part of focusing curve, the ration of sharpness, the steepness, the variance of float part of focusing curve, the factor of local extreme and the sensitivity. On the other hand, the effect of noise, and image content on algorithm is analyzed in this paper. The experiment results show that the new algorithm has better performance of sensitivity, anti-nose than the four typical evaluation algorithms. The evaluation results are consistent with human visual characteristics. © Copyright 2015 SPIE.
  Accession Number: 20161602266736
• Record 351 of
  
  Title:Efficient object detection by prediction in 3D space
  
  Author(s):Pang, Yanwei(1); Jiang, Xiaoheng(1); Li, Xuelong(2); Pan, Jing(3)
  Source: Signal Processing  Volume:   Issue:   DOI: 10.1016/j.sigpro.2014.08.039  Published: March 11, 2014  
  Abstract:Because the scale, horizontal and vertical coordinates of an object in an image are arbitrary, so object detection can be viewed as a process of searching the object in the 3D space spanned by the scale, horizontal, and vertical factors. Traditional sliding window based method has to exhaustively search and check the 3D space, resulting in prohibitive computation cost. To deal with this problem, in this paper, we propose to explore both the scaling capacity and translation capacity of object detector to accelerate detection speed, without loss of detection accuracy. In our paradigm, scaling capacity can relieve the use of all possible sizes of templates at the first stage, i.e., only a few number of templates that can cover a large range of target object size are used to coarsely find the targets. Similarly, translation capacity can avoid dense grid sampling at the very beginning. After initial estimation, further evaluations with templates of finer scales are carried out around the candidates to verify the existence of target objects. Moreover, different from traditional uniform grid scanning, we present an interlaced scanning method called diamond grid scanning which can reduce redundant evaluation. Experimental results on face detection demonstrate the advantage of our method. © 2014 Elsevier B.V.
  Accession Number: 20150900572893
• Record 352 of
  
  Title:Sparse kernel entropy component analysis for dimensionality reduction of biomedical data
  
  Author(s):Shi, Jun(1); Jiang, Qikun(1); Zhang, Qi(1); Huang, Qinghua(2); Li, Xuelong(3)
  Source: Neurocomputing  Volume: 168  Issue:   DOI: 10.1016/j.neucom.2015.05.032  Published: November 30, 2015  
  Abstract:Dimensionality reduction is ubiquitous in biomedical applications. A newly proposed spectral dimensionality reduction method, named kernel entropy component analysis (KECA), can reveal the structure related to Renyi entropy of an input space data set. However, each principal component in the Hilbert space depends on all training samples in KECA, causing degraded performance. To overcome this drawback, a sparse KECA (SKECA) algorithm based on a recursive divide-and-conquer (DC) method is proposed in this work. The original large and complex problem of KECA is decomposed into a series of small and simple sub-problems, and then they are solved recursively. The performance of SKECA is evaluated on four biomedical datasets, and compared with KECA, principal component analysis (PCA), kernel PCA (KPCA), sparse PCA and sparse KPCA. Experimental results indicate that the SKECA outperforms conventional dimensionality reduction algorithms, even for high order dimensional features. It suggests that SKECA is potentially applicable to biomedical data processing. © 2015 Elsevier B.V.
  Accession Number: 20152400938851
• Record 353 of
  
  Title:Visual Tracking Using Strong Classifier and Structural Local Sparse Descriptors
  
  Author(s):Ma, Bo(1); Shen, Jianbing(1); Liu, Yangbiao(1); Hu, Hongwei(1); Shao, Ling(2); Li, Xuelong(3)
  Source: IEEE Transactions on Multimedia  Volume: 17  Issue: 10  DOI: 10.1109/TMM.2015.2463221  Published: October 2015  
  Abstract:Sparse coding methods have achieved great success in visual tracking, and we present a strong classifier and structural local sparse descriptors for robust visual tracking. Since the summary features considering the sparse codes are sensitive to occlusion and other interfering factors, we extract local sparse descriptors from a fraction of all patches by performing a pooling operation. The collection of local sparse descriptors is combined into a boosting-based strong classifier for robust visual tracking using a discriminative appearance model. Furthermore, a structural reconstruction error based weight computation method is proposed to adjust the classification score of each candidate for more precise tracking results. To handle appearance changes during tracking, we present an occlusion-aware template update scheme. Comprehensive experimental comparisons with the state-of-the-art algorithms demonstrated the better performance of the proposed method. © 2015 IEEE.
  Accession Number: 20161102100473
• Record 354 of
  
  Title:Optical system design of multispectral space camera
  
  Author(s):Li, Xu-Yang(1); Yi, Hong-Wei(1); Qi, Hao-Cheng(2)
  Source: Guangzi Xuebao/Acta Photonica Sinica  Volume: 44  Issue: 3  DOI: 10.3788/gzxb20154403.0311002  Published: March 1, 2015  
  Abstract:The reflective optical system which is characterized by the spectrum band 450~900 nm, focal length of 5 000 mm and F number of 10 was designed. The result indicates that the field angle reaches 1.6°, optical system distortion is less than 0.5%. Modulation Transfer Function (MTF) reaches 0.65 at Nyquist frequency(25 lp/mm) with 6% central obscure, the quality of image reaches the diffraction limited. At the same time, the paper studies the static MTF of all spectrums based on using time delayed integration CCD. So this kind of optical system can meet the demand for the use of multispectral space cameras. ©, 2015, Chinese Optical Society. All right reserved.
  Accession Number: 20151600765018
• Record 355 of
  
  Title:Event-Based Media Enrichment Using an Adaptive Probabilistic Hypergraph Model
  
  Author(s):Liu, Xueliang(1); Wang, Meng(1); Yin, Bao-Cai(2); Huet, Benoit(3); Li, Xuelong(4)
  Source: IEEE Transactions on Cybernetics  Volume: 45  Issue: 11  DOI: 10.1109/TCYB.2014.2374755  Published: November 2015  
  Abstract:Nowadays, with the continual development of digital capture technologies and social media services, a vast number of media documents are captured and shared online to help attendees record their experience during events. In this paper, we present a method combining semantic inference and multimodal analysis for automatically finding media content to illustrate events using an adaptive probabilistic hypergraph model. In this model, media items are taken as vertices in the weighted hypergraph and the task of enriching media to illustrate events is formulated as a ranking problem. In our method, each hyperedge is constructed using the K-nearest neighbors of a given media document. We also employ a probabilistic representation, which assigns each vertex to a hyperedge in a probabilistic way, to further exploit the correlation among media data. Furthermore, we optimize the hypergraph weights in a regularization framework, which is solved as a second-order cone problem. The approach is initiated by seed media and then used to rank the media documents using a transductive inference process. The results obtained from validating the approach on an event dataset collected from EventMedia demonstrate the effectiveness of the proposed approach. © 2015 IEEE.
  Accession Number: 20161102104089
• Record 356 of
  
  Title:A fine-grained image categorization system by cellet-encoded spatial pyramid modeling
  
  Author(s):Zhang, Luming(1); Gao, Yue(2); Xia, Yingjie(3); Dai, Qionghai(2); Li, Xuelong(4)
  Source: IEEE Transactions on Industrial Electronics  Volume: 62  Issue: 1  DOI: 10.1109/TIE.2014.2327558  Published: January 1, 2015  
  Abstract:In this paper, a new fine-grained image categorization system that improves spatial pyramid matching is developed. In this method, multiple cells are combined into cellets in the proposed categorization model, which are highly responsive to an object's fine categories. The object components are represented by cellets that can connect spatially adjacent cells within the same pyramid level. Here, image categorization can be formulated as the matching between the cellets of corresponding images. Toward an effective matching process, an active learning algorithm that can effectively select a few representative cells for constructing the cellets is adopted. A linear-discriminant-analysis-like scheme is employed to select discriminative cellets. Then, fine-grained image categorization can be conducted with a trained linear support vector machine. Experimental results on three real-world data sets demonstrate that our proposed system outperforms the state of the art. © 1982-2012 IEEE.
  Accession Number: 20150100399393
• Record 357 of
  
  Title:Advances in bonding technology for high power diode laser bars
  
  Author(s):Wang, Jingwei(1); Li, Xiaoning(1,2,3); Hou, Dong(1); Feng, Feifei(1); Liu, Yalong(1); Liu, Xingsheng(1,2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9346  Issue:   DOI: 10.1117/12.2079619  Published: 2015  
  Abstract:Due to their high electrical-optical conversion efficiency, compact size and long lifetime, high power diode lasers have found increased applications in many fields. As the improvement of device technology, high power diode laser bars with output power of tens or hundreds watts have been commercially available. With the increase of high current and output power, the reliability and lifetime of high power diode laser bars becomes a challenge, especially under harsh working conditions and hard-pulse operations. The bonding technology is still one of the bottlenecks of the advancement of high power diode laser bars. Currently, materials used in bonding high power diode laser bars are commonly indium and goldtin solders. Experimental and field application results indicates that the lifetime and reliability of high power diode laser bars bonded by gold-tin solder is much better than that bonded by indium solder which is prone to thermal fatigue, electro-migration and oxidization. In this paper, we review the bonding technologies for high power diode laser bars and present the advances in bonding technology for single bars, horizontal bar arrays and vertical bar stacks. We will also present the challenges and issues in bonding technology for high power diode laser bars and discuss some approaches and strategies in addressing the challenges and issues. © 2015 SPIE.
  Accession Number: 20152600969174
• Record 358 of
  
  Title:Effects of packaging on the performances of high brightness 9xx nm CW mini-bar diode lasers
  
  Author(s):Li, Xiaoning(1,2,3); Wang, Jingwei(2); Feng, Feifei(1); Liu, Yalong(2); Yu, Dongshan(2); Zhang, Pu(1); Liu, Xingsheng(1,2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9346  Issue:   DOI: 10.1117/12.2080480  Published: 2015  
  Abstract:9xx nm CW mini-bar diode lasers and stacks with high brightness and reliability are desired for pumping fiber lasers and direct fiber coupling applications. For the traditional cm-bar with 1mm-2mm cavity, it can provide CW output power up to 80W-100W and high reliability, whereas the brightness is relatively low. In comparison, mini-bar based diode lasers with 4mm cavity offer a superior performance balance between power, brightness, and reliability. However, the long cavity and large footprint of mini-bar diode laser renders its sensitivity towards thermal stress formed in packaging process, which directly affects the performances of high bright mini-bar diode lasers. In this work, the thermal stress correlating with package structure and packaging process are compared and analyzed. Based on the experiment and analysis results, an optimized package structure of CW 60W 976 nm mini-bar diode lasers is designed and developed which relieves thermal stress. © 2015 SPIE.
  Accession Number: 20152600969151
• Record 359 of
  
  Title:Effect of interface layer on the performance of high power diode laser arrays
  
  Author(s):Zhang, Pu(1); Wang, Jingwei(2); Xiong, Lingling(1); Li, Xiaoning(1,3); Hou, Dong(2); Liu, Xingsheng(1,2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9346  Issue:   DOI: 10.1117/12.2077029  Published: 2015  
  Abstract:Packaging is an important part of high power diode laser (HPLD) development and has become one of the key factors affecting the performance of high power diode lasers. In the package structure of HPLD, the interface layer of die bonding has significant effects on the thermal behavior of high power diode laser packages and most degradations and failures in high power diode laser packages are directly related to the interface layer. In this work, the effects of interface layer on the performance of high power diode laser array were studied numerically by modeling and experimentally. Firstly, numerical simulations using finite element method (FEM) were conducted to analyze the effects of voids in the interface layer on the temperature rise in active region of diode laser array. The correlation between junction temperature rise and voids was analyzed. According to the numerical simulation results, it was found that the local temperature rise of active region originated from the voids in the solder layer will lead to wavelength shift of some emitters. Secondly, the effects of solder interface layer on the spectrum properties of high power diode laser array were studied. It showed that the spectrum shape of diode laser array appeared "right shoulder" or "multi-peaks", which were related to the voids in the solder interface layer. Finally, "void-free" techniques were developed to minimize the voids in the solder interface layer and achieve high power diode lasers with better optical-electrical performances. © 2015 SPIE.
  Accession Number: 20152600969176
• Record 360 of
  
  Title:Packaging of complete indium-free high reliable and high power diode laser array
  
  Author(s):Wang, Jingwei(1); Li, Xiaoning(1,2,3); Feng, Feifei(1); Liu, Yalong(1); Hou, Dong(1); Liu, Xingsheng(1,2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 9255  Issue:   DOI: 10.1117/12.2072357  Published: 2015  
  Abstract:High power diode lasers have been widely used in many fields. For many applications, a diode laser needs to be robust under on-off power-cycling as well as environmental thermal cycling conditions. To meet the requirements, the conduction cooled single bar CS-packaged diode laser arrays must have high durability to withstand thermal fatigue and long lifetime. In this paper, a complete indium-free bonding technology is presented for packaging high power diode laser arrays. Numerical simulations on the thermal behavior of CS-packaged diode laser array with different packaging structure were conducted and analyzed. Based on the simulation results, the device structure and packaging process of complete indium-free CS-packaged diode laser array were optimized. A series of high power hard solder CS (HCS) diode laser arrays were fabricated and characterized. Under the harsh working condition of 90s on and 30s off, good lifetime was demonstrated on 825nm 60W single bar CS-packaged diode laser with a lifetime test of more than 6100hours achieved so far with less 5% power degradation and less 1.5nm wavelength shift. Additionally, the measurement results indicated that the lower smile of complete indium-free CS-packaged diode laser arrays were achieved by advanced packaging process. © 2015 SPIE.
  Accession Number: 20151000608744