2019

2019

  • Record 85 of

    Title:Vision-to-Language Tasks Based on Attributes and Attention Mechanism
    Author(s):Li, Xuelong(1); Yuan, Aihong(2,3); Lu, Xiaoqiang(2)
    Source: arXiv  Volume:   Issue:   DOI: null  Published: May 29, 2019  
    Abstract:Vision-to-Language tasks aim to integrate computer vision and natural language processing together, which has attracted the attention of many researchers. For typical approaches, they encode image into feature representations and decode it into natural language sentences. While they neglect high-level semantic concepts and subtle relationships between image regions and natural language elements. To make full use of these information, this paper attempt to exploit the text-guided attention and semantic-guided attention to find the more correlated spatial information and reduce the semantic gap between vision and language. Our method includes two level attention networks. One is the text-guided attention network which is used to select the text-related regions. The other is semantic-guided attention network which is used to highlight the concept-related regions and the region-related concepts. At last, all these information are incorporated to generate captions or answers. Practically, image captioning and visual question answering experiments have been carried out, and the experimental results have shown the excellent performance of the proposed approach. Copyright © 2019, The Authors. All rights reserved.
    Accession Number: 20200337417
  • Record 86 of

    Title:Modified phase diversity technique to eliminate Poisson noise for reconstructing high-resolution images
    Author(s):Zhang, Ling(1,2); Zhao, Hui(1); Yi, Hongwei(1); Fan, Xuewu(1); Xie, Xiaopeng(1,2)
    Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 10838  Issue:   DOI: 10.1117/12.2505019  Published: 2019  
    Abstract:Phase diversity technique (PD) is a widely known method to estimate wave-front aberration of optical imaging system and to obtain reconstructed high-resolution image after degradation. However, when detecting weak or low light object in space, Poisson noise, as the main source of noise, has a serious impact on the accuracy of the PD's two main function. Hence, we firstly propose a modified PD combined with Non-local Means (NLM) algorithm to reduce the sensitivity of PD towards the Poisson noise. The numerical simulations demonstrate that our approach compared with the traditional PD has a significant improvement in terms of both the wave-front residual root-mean-square error (RMS) and the structural similarity index metrics (SSIM). The wave-front residual RMS decreases by approximately 51% across the Poisson noise levels ranging from 24.48 dB to 61.02 dB. The overall decline range of SSIM significantly decreases from 47% to 17%, and the average of SSIM increases from 84% to 91%. The modified PD would be useful in the AO system with active Poisson noise. © 2019 COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.
    Accession Number: 20190506430680
  • Record 87 of

    Title:Research on temperature compensation technology of micro-electro-mechanical systems gyroscope in strap-down inertial measurement unit
    Author(s):Liu, Ying(1); Liu, Cong(1); Xu, Jintao(2); Zhao, Xiaodong(2)
    Source: Advances in Intelligent Systems and Computing  Volume: 891  Issue:   DOI: 10.1007/978-3-030-03766-6_2  Published: 2019  
    Abstract:Due to the characteristics of MEMS gyroscope and the influence of the peripheral driving circuit, the MEMS gyroscope is easily affected by temperature and the accuracy is deteriorated. The compensation delay is caused by the complexity of the model in practical engineering applications. A second-order polynomial compensation model for temperature-divided regions is proposed by analyzing the mechanism of gyroscope zero-bias temperature drift. The Model first divides the temperature region of the gyroscope work, and then uses the least squares method to solve the parameters according to multiple linear regression analysis. Finally, the model was verified by experiments. The results show that the model can effectively reduce the drift temperature drift caused by temperature changes, which can reduce the temperature drift after compensation by 73.3%. © Springer Nature Switzerland AG 2019.
    Accession Number: 20190106335318
  • Record 88 of

    Title:Tripartite Einstein-Podolsky-Rosen steering with linear and nonlinear beamsplitters in four-wave mixing of Rubidium atoms
    Author(s):Liu, Yang(1,3,4); Cai, Yin(1); Xiang, Yu(2,5,6); Li, Feng(1); Zhang, Yanpeng(1); He, Qiongyi(2,5,6)
    Source: Optics Express  Volume: 27  Issue: 23  DOI: 10.1364/OE.27.033070  Published: November 11, 2019  
    Abstract:Multipartite Einstein-Podolsky-Rosen (EPR) steering is an essential resource for secure one-sided device-independent quantum secret sharing. Here, we analyze the EPR steering properties exhibited in three-mode Gaussian states created by four-wave mixing (FWM) in Rubidium atoms combined with a linear beamsplitter and a nonlinear beamsplitter (second FWM), respectively. By quantifying Gaussian steerability based on a measure determined by the covariance matrix of the produced states, we compare the performance of two schemes to achieve one-way, collective, and genuine tripartite steering, as well as the monogamy constraints for distributing steering among three parties. We show that the scheme with nonlinear beamsplitter is feasible to create stronger bipartite steering and genuine tripartite steering and has more flexibility to manipulate the monogamy relation through the cooperation of the two cascaded FWM processes. © 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
    Accession Number: 20194707726966
  • Record 89 of

    Title:Kerr combs and telecommunications components for the generation and high-dimensional quantum processing of d-level cluster states
    Author(s):Roztocki, Piotr(1); Reimer, Christian(1,2); Sciara, Stefania(1,3); Islam, Mehedi(1); Cortés, Luis Romero(1); Zhang, Yanbing(1); Fischer, Bennet(1); Loranger, Sébastien(4,5); Kashyap, Raman(4,6); Cino, Alfonso(3); Chu, Sai T.(7); Little, Brent E.(8); Moss, David J.(9); Caspani, Lucia(10); Munro, William J.(11,12); Azaña, José(1); Kues, Michael(1,13); Morandotti, Roberto(1,14,15)
    Source: International Conference on Transparent Optical Networks  Volume: 2019-July  Issue:   DOI: 10.1109/ICTON.2019.8840346  Published: July 2019  
    Abstract:Large and complex optical quantum states are a key resource for fundamental science and applications such as quantum communications, information processing, and metrology. In this context, cluster states are a particularly important class because they enable the realization of universal quantum computers by means of the so-called 'one-way' scheme, where processing operations are performed through measurements on the state. While two-level (i.e. qubit) cluster states have been realized thus far, further boosting this computational resource by increasing the number of particles comes at the price of significantly reduced coherence time and detection rates, as well as increased sensitivity to noise. In contrast, the realization of d-level (with d > 2) cluster states offers the possibility to increase quantum resources without changing the number of particles, enables the implementation of efficient computational protocols, as well as coincides with a reduction in the noise sensitivity of the states. Here, we experimentally realize, characterize, and perform one-way processing operations on three-level, four-partite cluster states formed by two photons in the time and frequency domain. We make use of a unique approach based on integrated photonic chips and optical fiber communications components, enabling scalable, deterministic new functionalities. © 2019 IEEE.
    Accession Number: 20194107527391
  • Record 90 of

    Title:Joint optimization of apodizer and lyot stop for coronagraph with four-quadrant phase mask
    Author(s):Ge, Rui(1); Zhao, Hui(1); Li, Chuang(1); Fan, Xue-Wu(1); Yu, Cong-Cong(1,2)
    Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 11341  Issue:   DOI: 10.1117/12.2538039  Published: 2019  
    Abstract:Exploring exoplanets using stellar coronagraph requires coronagraph with a contrast of 10-10 or even lower, because the difference in light intensity between the planet and its parent star is very large. To this end, we optimized the coronagraph imaging system with Four-Quadrant Phase-Mask (FQPM) proposed by D. ROUAN et al in 2000. The FQPM has the advantages of high extinction efficiency for the coherent light from the main source, low sensitivity to atmospheric turbulence and large dynamic range. This paper proposed an apodizer with continuous transmission and a Lyot stop optimized in conjunction with the apodizer for FQPM coronagraph, which enhance the nulling ability of FQPM to achieve a high contrast ratio of 10-12 at 1.25λ/D and a contrast of 10-13 at larger distance. Moreover, this optimization method can optimize the non-circular symmetric mask, that is, the optimization method is two-dimensional, rather than one-dimensional optimization in the case of circular symmetry. Then we compare the joint optimization with the optimization of the apodizer only, the results show that the former has better diffraction suppression and concentrating ability than the latter, which makes the energy more concentrated and the peak signal obtained on the detector is stronger. In the follow-up work, we will continue to complete the FQPM coronagraph system, such as adding adaptive optical to correct wavefront distortion, adding considerations for manufacturing precision of optical components and so on. © 2019 SPIE.
    Accession Number: 20200408063316
  • Record 91 of

    Title:Deterministic generation and switching of dissipative Kerr soliton in a thermally controlled micro-resonator
    Author(s):Lu, Zhizhou(1,2); Wang, Weiqiang(1,2); Zhang, Wenfu(1,2); Chu, Sai T.(3); Little, Brent E.(1); Liu, Mulong(1,2); Wang, Leiran(1,2); Zou, Chang-Ling(4,5); Dong, Chun-Hua(4,5); Zhao, Bailing(1,2); Zhao, Wei(1,2)
    Source: AIP Advances  Volume: 9  Issue: 2  DOI: 10.1063/1.5080128  Published: February 1, 2019  
    Abstract:In this paper, we first experimentally demonstrate deterministic generation and switching of dissipative Kerr solitons (DKSs) in a thermally controlled micro-ring resonator based on high-index doped silica glass platform. In our scheme, an auxiliary laser is introduced to timely balance the intra-cavity heat fluctuation. By decreasing the operation temperature through a thermo-electric cooler, primary-, chaotic-comb and soliton crystal are firstly generated, then increasing the temperature, DKSs switching and single soliton are robustly accessed, which is independent of the tuning speed. During the switching process, varieties of DKSs are identified by tens of the characteristic "soliton-steps". Besides, by simply changing the operation temperature under which the DKSs are formed, the center wavelength of dispersive waves could be tuned in a broadband range. When the micro-ring resonator operating at temperature larger than 63.5 °C, avoided mode-crossing free soliton can be obtained. Our results are favorable for study of on-chip soliton dynamics and practical nonlinear applications. © 2019 Author(s).
    Accession Number: 20191006585432
  • Record 92 of

    Title:A Method for Detecting Small and Weak Defect Targets Based on Neighborhood Vector PCA Image Enhancement
    Author(s):Wang, Zheng-Zhou(1); Li, Gang(1); Wang, Wei(1); Xia, Yan-Wen(2); Wang, Li(1); Tan, Meng(1)
    Source: Guangzi Xuebao/Acta Photonica Sinica  Volume: 48  Issue: 7  DOI: 10.3788/gzxb20194807.0710001  Published: July 1, 2019  
    Abstract:A method for detecting small and weak damaged targets based on Neighborhood Vector PCA (NVPCA) image enhancement was proposed. The main idea is that each pixel and its 8 neighborhood pixels in the damaged image are treated as a column vector to participate in the operation. All column vectors generated by each pixel will construct a 9-dimensional data cube. After PCA transformation, the correlation between the middle and neighborhood pixels is eliminated, so that the main information of the 9-dimensional data cube will be set in the first dimension, and the transformed first dimension data is NVPCA image. When the NVPCA image is processed again by LCM method, a better image enhancement effect is obtained. In addition, the region growth method is used to separate the damage target from background. The experimental results show that the method can detect the damage target with the size of 1 pixel and located in local bright area, and meet the requirement of on-line optical damage detection system for the accuracy of the damage target. © 2019, Science Press. All right reserved.
    Accession Number: 20193407330507
  • Record 93 of

    Title:Cylindrically Focused Nonablative Femtosecond Laser Processing of Long-Range Uniform Periodic Surface Structures with Tunable Diffraction Efficiency
    Author(s):Huang, Ji(1); Jiang, Lan(1); Li, Xiaowei(1); Wei, Qunshuo(2); Wang, Zhipeng(1); Li, Bohong(1); Huang, Lingling(2); Wang, Andong(1); Wang, Zhi(1); Li, Ming(3); Qu, Liangti(4,5); Lu, Yongfeng(6)
    Source: Advanced Optical Materials  Volume: 7  Issue: 20  DOI: 10.1002/adom.201900706  Published: October 1, 2019  
    Abstract:Periodic surface structures are core components for controlling the dispersion and steering characteristics of light. Here, a mask-free approach using nonablative femtosecond laser processing is proposed and demonstrated to fabricate extremely long-range uniform periodic surface structures on silicon with tunable diffraction efficiency. First, a cylindrically focused femtosecond laser scans over silicon substrates to efficiently produce large-area periodic modified stripes in a nonablation regime. Second, the modified stripes act as fine etch stops to generate the desired structures on sample surfaces during the subsequent chemical etching process. The structures produced by the method achieve optimal long-range uniformity compared to the reported laser-induced periodic surface structures, which possess a minimum divergence of structure orientation angles of © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
    Accession Number: 20193207291608
  • Record 94 of

    Title:Femtosecond laser ablation and photo-induced effects of As40S60, Ga0.8As39.2S60 and Ga0.8As29.2Sb10S60 chalcogenide glasses
    Author(s):Liu, Lutao(1,2); Zheng, Xin(2,3); Xiao, Xusheng(1); Xu, Yantao(1,2); Cui, Xiaoxia(1,2); Cui, Jian(1,2); Guo, Chunlei(3,4); Yang, Jianjun(3); Guo, Haitao(1,2)
    Source: Optical Materials Express  Volume: 9  Issue: 9  DOI: 10.1364/OME.9.003582  Published: September 1, 2019  
    Abstract:Laser induced damage thresholds (LIDTs) and photo-induced changes of As40S60, Ga0.8As39.2S60 and Ga0.8As29.2Sb10S60 chalcogenide glasses are investigated by femtosecond laser of 800 nm. As40S60 glass has the highest LIDT as well as 1452.3 mJ/cm2, the introduction of small amount of Ga and Sb into glass decreases the LIDTs to 957.1 mJ/cm2 for Ga0.8As39.2S60 and 705.9 mJ/cm2 for Ga0.8As29.2Sb10S60, respectively. Microstructure analysis reveals that the decrease of LIDT is tightly related to the decrease of high strength chemical bonds and formation of lower ones in glass matrix. After multi pulses induced damage occurred, the structure of glass matrix became more random and a half of S was replaced by O approximately. The damage mechanism was proposed and it is helpful to develop high LIDT chalcogenide glasses and the photo-induced effects are the basis of waveguide writing in chalcogenide glasses by femtosecond laser. © 2019 Optical Society of America.
    Accession Number: 20194007489326
  • Record 95 of

    Title:Ultrafast time-response characteristics of AlGaAs materials
    Author(s):Yin, Fei(1,2); Yang, Qing(1); Wang, Tao(2); Gao, Guilong(2); He, Kai(2); Yan, Xin(2); Li, Shaohui(2)
    Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 10843  Issue:   DOI: 10.1117/12.2506353  Published: 2019  
    Abstract:The ultrafast all-optical solid-state framing camera(UASFC) based on semiconductor photorefractive effect is a new type of X-ray ultrafast imaging system. The temporal resolution of UASFC is determined by the response time of the semiconductor. We improve the pump-probe experiment to measure the time response of GaAs/AlGaAs. In our recent experiments, the full width of half maximum (FWHM) is about 2ps, and the dynamic test result of the UASFC system, which use these AlGaAs samples, is 2.5ps. The results verify feasibility of the measurement and provide necessary methods for the further construction of high performance UASFC system. © 2019 SPIE ·
    Accession Number: 20191006603240
  • Record 96 of

    Title:Robust Space-Frequency Joint Representation for Remote Sensing Image Scene Classification
    Author(s):Fang, Jie(1); Yuan, Yuan(2); Lu, Xiaoqiang(1); Feng, Yachuang(1)
    Source: IEEE Transactions on Geoscience and Remote Sensing  Volume: 57  Issue: 10  DOI: 10.1109/TGRS.2019.2913816  Published: October 2019  
    Abstract:Remote sensing image scene classification is a fundamental problem, which aims to label an image with a specific semantic category automatically. Recent progress on remote sensing image scene classification is substantial, benefitting mostly from the powerful feature extraction capability of convolutional neural networks (CNNs). Even though these CNN-based methods have achieved competitive performances, they only construct the representation of the image in location-sensitive space-domain. As a result, their representations are not robust to rotation-variant remote sensing images, which influence the classification accuracy. In this paper, we propose a novel feature representation method by introducing a frequency-domain branch to the traditional only-space-domain architecture. Our framework takes full advantages of discriminative features from space domain and location-robust features from the frequency domain, providing more advanced representations through an additional joint learning module, a property that is critically needed to perform remote sensing image scene classification. Additionally, our method produces satisfactory performances on four public and challenging remote sensing image scene data sets, Sydney, UC-Merced, WHU-RS19, and AID. © 1980-2012 IEEE.
    Accession Number: 20200408086928