2022

• Record 97 of
  
  Title:Semisupervised Consistent Projection Metric Learning for Person Reidentification
  
  Author(s):Sun, Bangyong(1); Ren, Yutao(2); Lu, Xiaoqiang(1)
  Source: IEEE Transactions on Cybernetics  Volume: 52  Issue: 2  DOI: 10.1109/TCYB.2020.2979262  Published: February 1, 2022  
  Abstract:Person reidentification is a hot topic in the computer vision field. Many efforts have been paid on modeling a discriminative distance metric. However, existing metric-learning-based methods are a lack of generalization. In this article, the poor generalization of the metric model is argued as the biased estimation problem that the independent identical distribution hypothesis is not valid. The verification experimental result shows that there is a sharp difference between the training and test samples in the metric subspace. A semisupervised consistent projection metric-learning method is proposed to ease the biased estimation problem by learning a consistent constrained metric subspace in which the identified pairs are forced to follow the distribution of the positive training pairs. First, a semisupervised method is proposed to generate potential matching pairs from the {k} -nearest neighbors of test samples. The potential matching pairs are used to estimate the distances' distribution center of the positive test pairs. Second, the metric subspace is improved by forcing this estimation to be close to the center of the positive training pairs. Finally, extensive experiments are conducted on five datasets and the results demonstrate that the proposed method reaches the best performance, especially on the rank-1 identification rate. © 2013 IEEE.
  Accession Number: 20220911728758
• Record 98 of
  
  Title:Semisupervised Spectral Degradation Constrained Network for Spectral Super-Resolution
  
  Author(s):Chen, Wenjing(1); Zheng, Xiangtao(1); Lu, Xiaoqiang(1)
  Source: IEEE Geoscience and Remote Sensing Letters  Volume: 19  Issue:   DOI: 10.1109/LGRS.2021.3079961  Published: 2022  
  Abstract:Recently, various deep learning-based methods have been designed to improve the spectral resolution of the multispectral image (MSI) to obtain the hyperspectral image (HSI). These methods usually rely on sufficient MSI/HSI pairs for supervised training. However, collecting plentiful HSIs is time-consuming. In this letter, a semisupervised spectral degradation constrained network (SSDCN) is proposed to improve the spectral resolution of MSI. SSDCN is an autoencoder-like network that is composed of an encoder subnetwork for estimating HSI from input MSI and a decoder subnetwork for reconstructing MSI from the estimated HSI. A semisupervised training method is proposed to explore both MSI/HSI pairs and MSIs without ground-truth HSIs to optimize SSDCN. Simulated and two real databases are employed to demonstrate the effectiveness of SSDCN. © 2004-2012 IEEE.
  Accession Number: 20212310464071
• Record 99 of
  
  Title:A wideband self-decoupled multi-input multi-output antenna with a high isolation
  
  Author(s):Lu, Hao(1); Zhang, Li(1); Cao, Zhi-Xun(1); Sun, Ji-Qiu(1); Weng, Zibin(1); Wang, Hao(2)
  Source: International Journal of RF and Microwave Computer-Aided Engineering  Volume: 32  Issue: 7  DOI: 10.1002/mmce.23190  Published: July 2022  
  Abstract:A wideband MIMO antenna, which consists of two adjacent antenna elements with a very close distance (2 mm or 0.026λ), is suggested to achieve a good isolation by the principle of common mode (CM, in-phase signals) and differential mode (DM, out-of-phase signals) cancelation without using any additional decoupling structure in this article. Compared with its reference design, the measured results demonstrate that the isolation of proposed design can be improved to better than 20 dB in a wider bandwidth of 3.2–5.9 GHz (59%), with S11 and S22 © 2022 Wiley Periodicals LLC.
  Accession Number: 20221611982458
• Record 100 of
  
  Title:Computer-aided alignment method for AIMS solar telescope
  
  Author(s):E., Kewei(1); Fu, Xin(1); Shen, Yuliang(2); Zhao, Jianke(3); Wang, Tao(1); Chang, Ming(1); Liu, Shangkuo(1); Xue, Xun(1); Zhou, Yan(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 12166  Issue:   DOI: 10.1117/12.2617389  Published: 2022  
  Abstract:AIMS is an infrareds optical system for the accurate measurement of solar magnetic field, which is a national major scientific research project currently developed. The guiding optical system of AIMS is an off-axis Gregorian system, due to the designed minimum angle between the optical axis of the optical system and horizontal plane is 14.036°, a sub-aperture stitching test approach is developed to test the wavefront of the system. That makes the process of precision alignment of the system very difficult and laborious. Therefore, we developed a two-step alignment approach that based on merit function regression method, the developed method can make alignment of AIMS guiding optical system efficiency and accuracy. In this paper, we explain the detailed two-step alignment method and apply it to the real alignment of AIMS guiding system. Aided with sub-aperture stitching measurements, the AIMS guiding system is aligned and the results show that in 0.076λrms wavefront error in effective aperture was achieved. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.
  Accession Number: 20220911734915
• Record 101 of
  
  Title:Remote Sensing Scene Classification by Local-Global Mutual Learning
  
  Author(s):Chen, Xiumei(1); Zheng, Xiangtao(1); Zhang, Yue(1); Lu, Xiaoqiang(1)
  Source: IEEE Geoscience and Remote Sensing Letters  Volume: 19  Issue:   DOI: 10.1109/LGRS.2022.3150801  Published: 2022  
  Abstract:Remote sensing scene classification (RSSC) attempts to label an image with a specific scene category. Recently, convolutional neural networks (CNNs) have shown the powerful feature extraction capability to combine local and global features. However, both the local and global features are extracted independently, which ignore the complementary representation. In this letter, a local-global mutual learning (LML) method is proposed to capture both the global and local features. Specifically, local regions are first generated by highlighting the semantic areas in the corresponding original image. Then, a two-branch architecture is used to extract features for the local regions and global image, respectively. Both the classification loss and mutual learning loss are exploited to train the local-global branches simultaneously, which constrain the two branches to promote each other. Experiments on two popular datasets demonstrate the effectiveness of the proposed method. © 2004-2012 IEEE.
  Accession Number: 20220811679038
• Record 102 of
  
  Title:Mixed noise removal based on Stokes residual noise removal for division of focal plane polarimetric images
  
  Author(s):Jiang, Tuochi(1,2); Wen, Desheng(1); Song, Zongxi(1); Gao, Wei(1); Liu, Gang(1)
  Source: Optics and Lasers in Engineering  Volume: 159  Issue:   DOI: 10.1016/j.optlaseng.2022.107220  Published: December 2022  
  Abstract:The division of focal plane (DoFP) polarimeter has the advantages of simultaneous imaging and compact optical structure. It is inevitable to introducing the noise in the process of micro-polarizer array integration, polarization image acquisition and transmission. In this paper, we propose a novel mixed noise suppression method based on Stokes residual noise removal (SRNR) for mixed additive white Gaussian noise (AWGN) and impulse noise (IN) removal in DoFP polarimetric images. The Laplacian scale mixture (LSM) model is introduced to estimate the IN and the nonlocal low-rank regularization (NLR) is adopted to enhance the denoising performance. The residual noise of the Stokes parameters is taken into consideration. This crucial processing achieves the acquisition of high-quality polarization data. The experiments demonstrate the superior denoising performance for polarimetric imagery in terms of objective assessment and visual evaluation. © 2022
  Accession Number: 20223312578305
• Record 103 of
  
  Title:Infrared dim target detecting algorithm based on multi-feature and spatio-temporal fusion
  
  Author(s):Bai, Mei(1,2); Zhang, Jian(1); Zhao, Hui(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 12166  Issue:   DOI: 10.1117/12.2617906  Published: 2022  
  Abstract:Focusing on the detection of infrared dim targets in space-based imaging systems, a multi-feature and spatio-temporal fusion algorithm is proposed. By analyzing characteristics of target and background in the image, firstly, an algorithm combining TOP-HAT and improved partial differentiation method is proposed for image preprocessing; Secondly, the local entropy feature and local gradient feature of the small target image are extracted from images to fuse, an improved interframe method is used for spatio-temporal fusion to enhance the target signal, then threshold segmentation is used to obtain the detection result. Theoretical analysis and experiments show that the algorithm proposed in this paper can not only suppress the background and enhance the target well; in addition, it is possible to realize the on-chip transplantation of hardware. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.
  Accession Number: 20220911734986
• Record 104 of
  
  Title:Adaptive feedback connection with a single-level feature for object detection
  
  Author(s):Ruan, Zhongling(1); Cao, Jianzhong(1); Wang, Hao(1); Guo, Huinan(1); Yang, Xin(2)
  Source: IET Computer Vision  Volume: 16  Issue: 8  DOI: 10.1049/cvi2.12121  Published: December 2022  
  Abstract:From the perspective of detector optimisation, detecting objects using only a one-level feature cannot provide good performance for a wide range of scales. Various complex feature pyramidal structures address this problem using the divide-and-conquer strategy and multi-scale feature fusion. However, this requires adding too many additional convolutional layers and fusion operations. To address the issue, a simple detection part is proposed, which includes three components, namely a one-level feature map for detection, the encoder structure with feedback connection, and a decoupled head. The redesigned encoder and decoupled head can successfully address the performance decline caused by the one-level feature-based detection. Moreover, the proposed method can accelerate the convergence of the detector and achieve a faster inference time. Based on the optimised detection part, an adaptive feedback connection with a single-level feature (AFS) is proposed for object detection. The experiments conducted on the MS COCO 2017 benchmark show that the proposed method can achieve comparable results with its multi-scale pyramid counterpart, You Only Look Once v4 (YOLOv4). In addition, AFS can help the YOLOv4 achieve 44.9 mAP at 27 frame per second and converging 82 epochs earlier under the image size of 608×608, which represents a 42.1% improvements in the convergence speed. © 2022 The Authors. IET Computer Vision published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.
  Accession Number: 20222712326239
• Record 105 of
  
  Title:Optimized Design of Slab Width for High Performance Silicon Modulator
  
  Author(s):Chang, Chang(4,5); Cui, Jishi(1); Chen, Hongmin(1); Cui, Wenjing(1); Yang, Fenghe(1); Xu, Xiaofu(2,3)
  Source: SSRN  Volume:   Issue:   DOI: 10.2139/ssrn.4251033  Published: October 18, 2022  
  Abstract:In this paper, we extended the n-type slab width of the silicon modulator to 2.8 times compared with the p-type. The unbalanced slab will make a larger electro-optical bandwidth. Benefitted from the mobility difference between electrons and holes, the charging and discharging duration in the p-type region would be longer than the holes. Therefore, a rational extension of the n-type slab will not affect the charge-discharge duration of the modulator. Meanwhile, extending the n-type slab will reduce the parasitic capacitance of the transmission line electrodes and then the bandwidth of the modulator would be increased. The extending of the n-type slab could also reduce the insertion loss thanks to the reduction of the light absorption introduced by the electrodes at the same time. This design principle provides a reference for the design of the n- and p-type regions for other active devices. © 2022, The Authors. All rights reserved.
  Accession Number: 20220409517
• Record 106 of
  
  Title:Flat-Bottom Dark Gap Modes as a New Localized State Supported by Periodic Nanostructures
  
  Author(s):Li, Jiawei(1,2,3); Zhang, Yanpeng(2); Zeng, Jianhua(1,3)
  Source: SSRN  Volume:   Issue:   DOI: 10.2139/ssrn.4160576  Published: July 12, 2022  
  Abstract:We introduce the conception of flat-bottom dark gap modes in one-dimensional periodic nanostructures configured as periodic alternations (nanoscale stripes) of the local linear and nonlinear properties, resembling the Kronig-Penney model with combined nanoscopic linear-nonlinear lattices. We argue that, the flat-bottom gap modes, representing a new type of localized dark state in forbidden bandgap, can be implemented in BoseEinstein condensates and optics with periodic potentials like optical lattices and photonic crystals at nanoscale, with the existing experimental technologies. In addition, the fundamental localized dark gap modes and their higher-order ones, dark gap solitons and soliton clusters, are also predicted. Dynamics and stability of all the localized gap modes are scrutinized in two ways, linear-stability analysis and direct perturbed simulations. Our results open a new route to studying soliton physics in the context of periodic nanostructures where new localized states like flat-bottom dark gap solitons may exist. © 2022, The Authors. All rights reserved.
  Accession Number: 20220257206
• Record 107 of
  
  Title:Simulations of the Radiative Lifetime in Surface Acoustic Wave driven Piezoelectric Semiconductor Devices
  
  Author(s):Pang, Ziliang(1,2); Cao, Weiwei(1); Bai, YongLin(1)
  Source: 2022 23rd International Conference on Electronic Packaging Technology, ICEPT 2022  Volume:   Issue:   DOI: 10.1109/ICEPT56209.2022.9873441  Published: 2022  
  Abstract:The traveling wave potential wells that are generated from a surface acoustic wave in the piezoelectric materials of semiconductor efficiently trap the created electron hole pairs, which is a potential method of the single-photon detectors. In this study, we report on a model that is used to study the correlation between radiative lifetime and periodical potential wells. We devised a Python program to numerically find the eigenstates in 1D periodical potential well system. The Schrödinger equation is solved to obtain electron and hole wave functions that are related to charge distribution. Then, we analyzed the impact of the potential well on the corresponding radiative lifetime of the system. © 2022 IEEE.
  Accession Number: 20224012840724
• Record 108 of
  
  Title:Double-Functionalization of Water Repellence and Anti-Reflectance by Multiple-Laser-Based Fabrication of Triple-Scale Hierarchical Surface Structures
  
  Author(s):He, Jianguo(1,2,3); Li, Ming(4); Dai, Shoujun(1,2); Huang, Min(1,3); Liu, Yang(1,3); Li, Yang(1,3); Fan, Lianwen(5); Yu, Jin(1,2)
  Source: SSRN  Volume:   Issue:   DOI:   Published: March 4, 2022  
  Abstract:A novel strategy of laser ablation followed by tungsten-based pulsed laser deposition was proposed and experimentally verified for available surface functionalization with superhydrophobicity and anti-reflectance on 316L stainless steel. Three surface patterns, crater, parallel lines (PL), and grid, were manufactured by nanosecond laser ablation, while the surface morphology was controlled by the transverse traverse index. After the pulsed laser deposition treatment, tungsten particles and clusters were densely coated on the laser-ablated surfaces. The fabricated substrates were characterized by triple-scaled hierarchical structures having different patterns, corrugated structures, and broccoli-like nano-protrusions. An aging treatment was performed to improve wettability. The results verified that the proposed multiple laser treatments in combination with the aging treatment allow the quick implementation of superhydrophobicity while maintaining good reflectance over the spectral range. Moreover, the topology was characterized using an optical profiler and through scanning electron microscopy, while a video-based optical contact angle measuring device and spectrophotometer were used to measure the contact angle and optical reflectance, respectively. X-ray photoelectron spectroscopy was performed to analyze the chemical components. This method can prepare water-repellent anti-reflectance hybrid substrates for industrial and academic applications. © 2022, The Authors. All rights reserved.
  Accession Number: 20220055968