2016
2016
-
Record 409 of
Title:A parallel primal-dual splitting method for image restoration
Author(s):He, Chuan(1); Hu, Changhua(1); Li, Xuelong(2); Zhang, Wei(1)Source: Information Sciences Volume: 358-359 Issue: DOI: 10.1016/j.ins.2016.04.004 Published: September 1, 2016Abstract:We develop a parallel primal-dual splitting method to solve large-scale image restoration problems, which involve the sum of several linear-operator-coupled nonsmooth but proximable terms. With the proposed method, the objective function is decomposed into pieces that can be processed individually. No inverse operator is involved in our method and the highly parallel structure makes it preferable in distributed computation. The convergence is proven and the convergence rate is analyzed. Besides, its equivalence to the relaxed parallel linearized alternating direction method of multipliers (PLADMM) is addressed. Applications to image restoration problems with compound l1-regularizer and comparisons with state-of-the-art methods are detailed to show the superiority of the proposed method. ? 2016 Elsevier Inc. All rights reserved.Accession Number: 20161902367635 -
Record 410 of
Title:Angle Compensation and Asymmetry Effect of Light Diffracted by Millimeter Liquid Surface Slosh Wave
Author(s):Miao, Yang(1,2); Wu, Can(3); Wang, Ning(4); You, Jia-Qi(1)Source: Chinese Physics Letters Volume: 33 Issue: 7 DOI: 10.1088/0256-307X/33/7/074206 Published: July 2016Abstract:The angle compensation method is adopted to detect sloshing waves by laser diffraction, in the case that the wavelength of the sloshing waves is much greater than that of the incident light. The clear diffraction pattern is observed to be of asymmetry, involving orders, position and interval of the diffraction spots that are discovered during the light grazing incidence. It is found that the larger the angle of incidence is, the more obvious the asymmetry is. The higher the negative diffraction orders are, the smaller the intervals between spots are. On the contrary, in the positive region, the higher the diffraction orders are, the larger the spot intervals are. The positive interval is larger than that of the same negative diffraction order. If the incident angle reaches 1.558 rad in the experiment, all positive diffraction orders completely vanish. Based on the mechanism of phase modulation and with the Fourier transform method, the relations between the incident angle and position, interval spaces, and orders of diffraction spots are derived theoretically. The theoretical calculations are compared with the experimental data, and the comparison shows that the theoretical calculations are in good agreement with the experimental measurement. ? 2016 Chinese Physical Society and IOP Publishing Ltd.Accession Number: 20220711656847 -
Record 411 of
Title:Precision measuring system for tube support plate of steam generator
Author(s):Jiang, Xin(1); Li, Hua(1); Zhu, Hui(1); Liu, Jie(1)Source: Guangzi Xuebao/Acta Photonica Sinica Volume: 45 Issue: 3 DOI: 10.3788/gzxb20164503.0312002 Published: March 1, 2016Abstract:A precision measuring system for tube support plate of steam generator was built, and a parameter measurement method based on edge restriction condition was proposed. Firstly, two-dimensional discrete points were obtained by image processing technology, and the profile was pretreated by calculating the minimum enclosing rectangle of the points. Then, the profile was divided into a series of "extended neighborhood profiles", and an edge restriction algorithm based on curvature angle was built for precision demarcation. Finally, the profile parameters of part were measured precisely. The experiment results and error analysis show that the measurement precision can reach 0.02 mm, which has reference value for the precision measurement of components with complex profile. ? 2016, Science Press. All right reserved.Accession Number: 20161602269784 -
Record 412 of
Title:High-energy pulse fiber lasers
Author(s):Liu, Xueming(1)Source: Optics InfoBase Conference Papers Volume: Issue: DOI: 10.1364/ACPC.2016.AS1A.2 Published: July 21, 2014Abstract:nullAccession Number: 20171403520996 -
Record 413 of
Title:Study of a symmetrically structured SPR sensor and its dual-wavelength differential method
Author(s):Yang, Hai-Ma(1,2,3); Ma, Cai-Wen(1); Wang, Jian-Yu(2); Liu, Jin(4); Chen, Bao-Xue(3)Source: Journal of Modern Optics Volume: 63 Issue: 2 DOI: 10.1080/09500340.2015.1066455 Published: January 19, 2016Abstract:This study investigates a novel structure for surface plasmon resonance sensing and its dual-wavelength differential method. The surface plasmon wave is excited by a planar waveguide, which is prepared through the ion-exchange method. The distribution of refractive indices is fitted by the Fermi function. The sensing structure is based on a symmetric structure with a metal layer, a measured medium, and another metal layer. The condition for refractive index matching changes with the thicknesses of test samples, thus the test range can be adjusted using this structure. Given two appropriate wavelengths and for detection by the intensity method and an increase in the refractive index, the intensity variety at can be positive, whereas that at can be negative. When the refractive indices are determined based on differential values, sensitivity is improved. Solutions with refractive index values ranging from 1.33 to 1.428 are detected in the experiments using the single-wavelength method and the dual-wavelength differential method. Results show that the differential detection method enhances the adjustability and sensitivity of the SPR sensor in combination with a symmetric structure. ? 2015 Taylor & Francis.Accession Number: 20153101083519 -
Record 414 of
Title:Adaptive retinex algorithm based on genetic algorithm and human visual system
Author(s):Song, Xiaodong(1); Zhou, Zuofeng(1); Guo, Huinan(1); Zhao, Xiaodong(1); Zhang, Hui(1)Source: Proceedings - 2016 8th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2016 Volume: 1 Issue: DOI: 10.1109/IHMSC.2016.27 Published: December 13, 2016Abstract:Multi-scale retinex is a luminance perceptual algorithm. It is widely used in image enhancement. To get best effect, the parameters need be adjusted according to the scene manually. In order to handle this issue, a novel retinex algorithm based on real-coded genetic algorithm and human visual system was proposed in this paper. The result was quantified and selected by a novel human visual system based comprehensive fitness function. The employed fitness function operates adaptively and tends to provide outstanding comprehensive results in both image clarity and color quality. Images captured in low-light conditions were processed by proposed method. The result indicates that the new algorithm can improve image brightness and avoid color distortion at the same time. ? 2016 IEEE.Accession Number: 20170403288781 -
Record 415 of
Title:Multiple background sampling adaptive non-uniform correction algorithm
Author(s):Duan, Chengpeng(1); Liu, Wei(1); Chen, Yaohong(1); Xie, Qingsheng(1); Yi, Bo(1); Zhou, Zuofeng(1)Source: Guangxue Xuebao/Acta Optica Sinica Volume: 36 Issue: 10 DOI: 10.3788/AOS201636.1020001 Published: October 10, 2016Abstract:To improve the non-uniformity of infrared output images, the two-point correction and neural network algorithms are commonly used. However, the two-point correction algorithm cannot overcome the influence of environmental temperature drift effectively. Due to the slow convergence speed of the neural network algorithm, the still images using the neural network algorithm gradually integrate to the background, and the moving target appears an artifact. So a multiple background sampling adaptive non-uniform correction algorithm is proposed, and multiple groups of high- and low-temperature backgrounds are collected at different temperature points. The relationship between the achieved non-uniform correction coefficient and the environmental temperature is fitted by means of the least square method, and the adaptive non-uniform correction is implemented based the change of the environmental temperature. Test results show that this method is simple and feasible, and it can effectively overcome the influence of environmental temperature drift. ? 2016, Chinese Lasers Press. All right reserved.Accession Number: 20164502988969 -
Record 416 of
Title:Salient object segmentation with a shape-constrained level set
Author(s):Wang, Bin(1); Gao, Xinbo(1); Tao, Dacheng(2); Li, Xuelong(3); Balla-Arabé, Souleymane(4)Source: Biomedical Image Segmentation: Advances and Trends Volume: Issue: DOI: 10.1201/9781315372273 Published: November 17, 2016Abstract:A novel salient object segmentation based on the level set method with shape constraint is proposed. Given an image, this method first initializes the curve according to the saliency map which is subsequently combined into evolution in a multi-channel manner, and then rebuilds an energy functional with shape constraint. In such a way, the salient objects with a specified shape can be de-parted out automatically. The proposed one has some advantages as follows. First, the saliency map is taken as a considerable clue into LSMs to eliminate manual interaction and realize an automatic segmentation. Second, the saliency map is involved into curve's evolution which contributes to accelerating the evolution. Finally, shape priors are utilized to constrain the shape of curve which endow LSMs with a capability of handing broken or occluded objects. We respectively applied the proposed method to synthetic, medical, and natural images and obtained promising results. ? 2017 by Taylor & Francis Group, LLC.Accession Number: 20184906183019 -
Record 417 of
Title:The estimation method on diffusion spot energy concentration of the detection system
Author(s):Gao, Wei(1); Song, Zongxi(1); Liu, Feng(1); Dan, Lijun(1); Sun, Zhonghan(1); Du, Yunfei(1)Source: Proceedings of SPIE - The International Society for Optical Engineering Volume: 9684 Issue: DOI: 10.1117/12.2242811 Published: 2016Abstract:We propose a method to estimate the diffusion spot energy of the detection system. We do outdoor observation experiments in Xinglong Observatory, by using a detection system which diffusion spot energy concentration is estimated (the correlation coefficient is approximate 0.9926).The aperture of system is 300mm and limiting magnitude of system is 14.15Mv. Observation experiments show that the highest detecting magnitude of estimated system is 13.96Mv, and the average detecting magnitude of estimated system is about 13.5Mv. The results indicate that this method can be used to evaluate the energy diffusion spot concentration level of detection system efficiently. ? 2016 SPIE.Accession Number: 20165003113860 -
Record 418 of
Title:Stepless digital zoom for high definition camera
Author(s):Guo, Huinan(1); Fang, Yao(1); Liu, Qing(1); Zhang, Hui(1); Ma, Nan(2); Duan, Baosong(3)Source: Proceedings of SPIE - The International Society for Optical Engineering Volume: 10033 Issue: DOI: 10.1117/12.2245155 Published: 2016Abstract:Real-time full frame view of surveillance camera provides integral visual information of scene. High definition imaging and zooming are widely adopted when observers want to capture more detail information of targets. The optical zooming can provide high zoom ratio detail images of target as well as maintain imaging definition, however, it fails to obtain full view of scene when zoomed, and the field of observation decreases to region of target. Digital zooming is an effective approach to keep the balance between imaging field and visual detail information. This paper presents a method of stepless zooming for digital video camera which could be widely used in autonomous pan, tilt and zoom surveillance system. In view of hardware resources and algorithm realizability, an optimized zooming processing structure is proposed. According to input zoom ratio parameter, it can extract pixels of region of interest adaptively and display with original imaging size by mapping and interpolation algorithms. Experimental results indicate that the stepless zooming method can be capable of achieving 1080p high definition imaging and 30 frame/s video capture. ? 2016 SPIE.Accession Number: 20164903101516 -
Record 419 of
Title:Multimodal learning via exploring deep semantic similarity
Author(s):Hu, Di(1); Lu, Xiaoqiang(2); Li, Xuelong(2)Source: MM 2016 - Proceedings of the 2016 ACM Multimedia Conference Volume: Issue: DOI: 10.1145/2964284.2967239 Published: October 1, 2016Abstract:Deep learning is skilled at learning representation from raw data, which are embedded in the semantic space. Traditional multimodal networks take advantage of this, and maximize the joint distribution over the representations of different modalities. However, the similarity among the representations are not emphasized, which is an important property for multimodal data. In this paper, we will introduce a novel learning method for multimodal networks, named as Semantic Similarity Learning (SSL), which aims at training the model via enhancing the similarity between the highlevel features of different modalities. Sets of experiments are conducted for evaluating the method on different multimodal networks and multiple tasks. The experimental results demonstrate the effectiveness of SSL in keeping the shared information and improving the discrimination. Particularly, SSL shows its ability in encouraging each modality to learn transferred knowledge from the other one when faced with missing data. ? 2016 ACM.Accession Number: 20164603010662 -
Record 420 of
Title:Dehazing method for hyperspectral remote sensing imagery with hyperspectral linear unmixing
Author(s):Gan, Yuquan(1); Hu, Bingliang(1); Wen, Desheng(1); Wang, Shuang(1)Source: Proceedings of SPIE - The International Society for Optical Engineering Volume: 10156 Issue: DOI: 10.1117/12.2246656 Published: 2016Abstract:Haze always exists in hyper-spectral remote sensing imagery, and it is a key reason that influences the effective information extraction of hyper-spectral images. Specially, when the faint haze covers part of the target in remote sensing images, the target still can be detected but not clear. So, how to remove the influence of the haze and improve the applicable efficiency of hyper-spectral images is a popular research point. This paper proposes a dehazing method for hyper-spectral images based on linear unmixing. First, a popular hyper-spectral unmixing method called FUN is used to get the signature of all the end-members and their corresponding abundance. And then, the abundance of the haze end-member is removed and the abundances of the rest end-members are adjusted to satisfy the sum-To-one and non-negative constraint. Lastly, the new abundance and the signature of the end-members are linearly mixed to get the dehazed hyper-spectral images. The experiment result shows that the dehazed hyper-spectral images exhibit better target information and details. The method is effective and available. ? 2016 SPIE.Accession Number: 20165103154176