2020

• Record 157 of
  
  Title:An improved method for 3D reconstruction based on uniform point drift registration estimation
  
  Author(s):Zhang, Fan(1); Wang, Xin(1); Hu, Chao(1); Qu, YouShan(2)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 11437  Issue:   DOI: 10.1117/12.2543198  Published: 2020  
  Abstract:It is indispensable to obtain more information such as the 3D structure of the space target by detecting and identifying the target, when complete the on-orbit servicing and on-orbit control tasks. Both lidar and binocular stereo vision can provide three dimensional information of the environment. But it is very sensitive to the illuminance of environment and difficult to image registration at weak texture region, when we are using the binocular stereo vision in space. And lidar also has some defects such as the lidar data is sparse and the scanning frequency is low. So lidar and binocular stereo vision should be used together. The data of the lidar and binocular stereo vision are fused to make up for each others flaws. In this paper, uniform point drift registration method is used in the fusion of point cloud which is sampled by lidar and binocular stereo vision. In this method, the two groups of point cloud are considered as one which submit to mixed probability distribution and the other one which is sampled from the points submit to mixed probability distribution. The transformation estimation between the two groups of the point cloud is maximum likelihood estimation. The transformation is required to take overall smoothness. In other words, the point clouds should be uniformed. The uniform point drift method can solve the registration problem efficiently for 3D reconstruction. Usually the time can be compressed by 10%. © 2020 SPIE.
  Accession Number: 20201508383657
• Record 158 of
  
  Title:Beyond Rectangle Boundingbox: Visual Tracking Using Characteristic Points
  
  Author(s):Zhang, Ximing(1); Fan, Xuewu(1); Luo, Shujuan(2)
  Source: 2020 IEEE 6th International Conference on Computer and Communications, ICCC 2020  Volume:   Issue:   DOI: 10.1109/ICCC51575.2020.9345078  Published: December 11, 2020  
  Abstract:Visual Tracking plays a key role in computer vision application and artificial intelligent research. The main representation of tracking results come to rectangle boundingbox leading to inaccurate performance, which may not meet the requirements of computer vision application nowadays. We are more likely to obtain the pose estimation of tracking objects in order to achieve more complex computer vision mission including behavior detection and video analysis. While, rectangle boundingbox representation mostly dominates the output model when describing the target appearance in existing DNN-based trackers, leading to precision. We introduce the characteristic points(ChaPoints) to represent the both feature extraction and output model in order to accomplish pose estimation during training and tracking procedure. To this end, we build the mapping method between our proposed characteristic points and rectangle boundingbox. The multi-branches Siamese networks can further matching the candidates and the templates for final tracking results. We extensively prove the effectiveness of the proposed method through the ablation studies of the tracking benchmark, including OTB-2015 and UAV123. © 2020 IEEE.
  Accession Number: 20210910006569
• Record 159 of
  
  Title:Real Time Detection and Identification of UAV Abnormal Trajectory
  
  Author(s):Wang, Ziyuan(1,2); Zhang, Geng(1); Hu, Bingliang(1); Feng, Xiangpeng(1)
  Source: ACM International Conference Proceeding Series  Volume:   Issue:   DOI: 10.1145/3430199.3430212  Published: June 26, 2020  
  Abstract:Abnormal behavior detection based on video sequence is a hot field. At the same time, monitoring and tracking the UAV (Unmanned Aerial Vehicle) and identifying its abnormal behavior are great significance for the UAV defense. This paper focuses on the detection and recognition of the UAV abnormal trajectory based on real-time video sequence. By tracking and analyzing the characteristics of the UAV, the detection and recognition of abnormal trajectory are divided into two stages. First, by analyzing the UAV's abnormal trajectory satisfying the change conditions is extracted by the quantitative analysis of the UAV's directional angle change features. Second, the normalized polar path fourier spectrum feature of abnormal trajectory is established, and the feature is combined with window search length to accelerate the classification and identification of the UAV trajectory types. Through the contrast experiment, it shows that the method in this paper has good real-time performance and accuracy for trajectory recognition with scale and translation changes. © 2020 ACM.
  Accession Number: 20210309786047
• Record 160 of
  
  Title:Spectral-Spatial Attention Network for Hyperspectral Image Classification
  
  Author(s):Sun, Hao(1); Zheng, Xiangtao(1); Lu, Xiaoqiang(1); Wu, Siyuan(1)
  Source: IEEE Transactions on Geoscience and Remote Sensing  Volume: 58  Issue: 5  DOI: 10.1109/TGRS.2019.2951160  Published: May 2020  
  Abstract:Hyperspectral image (HSI) classification aims to assign each hyperspectral pixel with a proper land-cover label. Recently, convolutional neural networks (CNNs) have shown superior performance. To identify the land-cover label, CNN-based methods exploit the adjacent pixels as an input HSI cube, which simultaneously contains spectral signatures and spatial information. However, at the edge of each land-cover area, an HSI cube often contains several pixels whose land-cover labels are different from that of the center pixel. These pixels, named interfering pixels, will weaken the discrimination of spectral-spatial features and reduce classification accuracy. In this article, a spectral-spatial attention network (SSAN) is proposed to capture discriminative spectral-spatial features from attention areas of HSI cubes. First, a simple spectral-spatial network (SSN) is built to extract spectral-spatial features from HSI cubes. The SSN is composed of a spectral module and a spatial module. Each module consists of only a few 3-D convolution and activation operations, which make the proposed method easy to converge with a small number of training samples. Second, an attention module is introduced to suppress the effects of interfering pixels. The attention module is embedded into the SSN to obtain the SSAN. The experiments on several public HSI databases demonstrate that the proposed SSAN outperforms several state-of-The-Art methods. © 1980-2012 IEEE.
  Accession Number: 20201908611065
• Record 161 of
  
  Title:A Joint Relationship Aware Neural Network for Single-Image 3D Human Pose Estimation
  
  Author(s):Zheng, Xiangtao(1); Chen, Xiumei(1); Lu, Xiaoqiang(1)
  Source: IEEE Transactions on Image Processing  Volume: 29  Issue:   DOI: 10.1109/TIP.2020.2972104  Published: 2020  
  Abstract:This paper studies the task of 3D human pose estimation from a single RGB image, which is challenging without depth information. Recently many deep learning methods are proposed and achieve great improvements due to their strong representation learning. However, most existing methods ignore the relationship between joint features. In this paper, a joint relationship aware neural network is proposed to take both global and local joint relationship into consideration. First, a whole feature block representing all human body joints is extracted by a convolutional neural network. A Dual Attention Module (DAM) is applied on the whole feature block to generate attention weights. By exploiting the attention module, the global relationship between the whole joints is encoded. Second, the weighted whole feature block is divided into some individual joint features. To capture salient joint feature, the individual joint features are refined by individual DAMs. Finally, a joint angle prediction constraint is proposed to consider local joint relationship. Quantitative and qualitative experiments on 3D human pose estimation benchmarks demonstrate the effectiveness of the proposed method. © 1992-2012 IEEE.
  Accession Number: 20201208319148
• Record 162 of
  
  Title:Single Space Object Image Super Resolution Reconstructing Using Convolutional Networks in Wavelet Transform Domain
  
  Author(s):Feng, Xubin(1); Su, Xiuqin(1); Xu, Zhengpu(2); Xie, Meilin(1); Liu, Peng(1); Lian, Xuezheng(1); Jing, Feng(1); Cao, Yu(1)
  Source: 2020 IEEE 3rd International Conference on Electronics Technology, ICET 2020  Volume:   Issue:   DOI: 10.1109/ICET49382.2020.9119660  Published: May 2020  
  Abstract:With the increasing importance of space exploration, the research of space object is becoming more and more important because high-quality space object images are meaning for space attack and defense confrontation. However, high-quality space object images are very difficult to obtain because of the large number of various rays in the space environment and the inadequacy of optical lenses and detectors on satellites to support high-resolution imaging. Image super resolution reconstruction methods are the most cost-effective way to solve the problem. In this paper, we propose a deep convolutional neural network based method to improve the resolution of space object image. The implementation of our method is in wavelet transform domain rather than spatial domain because wavelet transformation could decompose different frequencies of the image very effectively and this could further more enhance the performance. The experiment result shows that our method could achieve a very good performance. © 2020 IEEE.
  Accession Number: 20202808913838
• Record 163 of
  
  Title:Design and analysis of a moving mirror supporting mechanism for fourier transform spectroscopy
  
  Author(s):Tian, Feifei(1,2); Li, Siyuan(1)
  Source: Proceedings - 2020 3rd International Conference on Electron Device and Mechanical Engineering, ICEDME 2020  Volume:   Issue:   DOI: 10.1109/ICEDME50972.2020.00129  Published: May 2020  
  Abstract:Moving mirror supporting mechanism is a key component of Fourier transform spectrometer(FTS), its motion precision and the maximum range of travel affect the performance of the instrument. Based on the double parallelogram structure, a symmetrical flexible moving mirror supporting mechanism is designed. Simulation results show the system permits ± 5 mm of mirror travel with the tilt value of less than ±4.8urad and the shear value of less than 3um. Excellent shear performance and tilt performance make it well suited for most FTS instruments. © 2020 IEEE.
  Accession Number: 20202908946747
• Record 164 of
  
  Title:Research on real-time distance measurement of mobile eye tracking system based on neural network
  
  Author(s):Hu, Ling(1,2); Gao, Jiarui(1)
  Source: Proceedings of 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference, ITOEC 2020  Volume:   Issue:   DOI: 10.1109/ITOEC49072.2020.9141800  Published: June 2020  
  Abstract:With the development and application of eye-tracking technology, mobile eye-tracking systems have become more widely used due to their safety and portability. We combine eye-tracking systems with real-time object detection using machine learning. We propose a method of wearing an eye tracker in daily life to obtain the distance between the eye tracking system and the gaze target in real time. During the visual interaction of the eye tracking system, in order to obtain the distance from the eyeball fixation target to the eyeball in real time, the world camera of the mobile eye tracking system pupil labs first collects the position and scale information of the detected target image in real time, and uses camera calibration principle, pinhole camera model and camera distortion model to establish a ranging equation, and then the feasibility of the real-time ranging equation is verified through a specified distance experiment. The total average relative error after de-distortion at the position of 50cm-75cm is reduced to 1.25%, and the highest accuracy-0.9182cm distance measurement can be achieved within the effective distance. © 2020 IEEE.
  Accession Number: 20203809211462
• Record 165 of
  
  Title:A wide-band interference spectrometer based on bandpass sampling technology
  
  Author(s):Tian, Feifei(1,2); Li, Siyuan(1)
  Source: Proceedings of SPIE - The International Society for Optical Engineering  Volume: 11606  Issue:   DOI: 10.1117/12.2585504  Published: 2020  
  Abstract:In the process of wide-band spectrum detection, interferogram acquisition of the traditional Michelson interferometer needs to follow Nyquist sampling theorem, the static performance such as high resolution of moving mirror scanning and the dynamic performance such as transient response need to meet strict requirements, which usually make the spectrometer system structure complex. Meanwhile, the interference modulation efficiency of traditional Michelson interferometer will drop sharply with the increase of optical path difference(OPD). In this way, the interference data value at the long optical path difference will be submerged by noise, which will reduce the signal-to-noise ratio of reconstructed spectrum. In order to simultaneously achieve spectrum detection with wide-band spectrum, high resolution and high signal-to-noise ratio, this paper introduces a configuration of wide-band interference spectrometer based on band-pass sampling technology. The wide-band interference spectrometer includes dispersion unit and interference modulation unit. Firstly, the dispersion unit pre-disperses the wide spectrum into continuous spectrum distributed along wavelength and divides the interference modulation signal of continuous spectrum into several interference signals of narrow-band spectrum. Secondly, the interference modulation unit carries out interference modulation on the dispersed continuous spectrum and the interferograms of every narrow-band spectrum are sampled and obtain the interferogram sequence of every narrow-band spectrum according to the band-pass sampling theorem. Finally, the spectral distribution of the detection target can be obtained by data processing and spectral superposition. The interference spectrometer provides a new idea for the development of spectral detection with wide spectral range, high resolution and high signalto- noise ratio. © 2020 SPIE.
  Accession Number: 20210109716240
• Record 166 of
  
  Title:Orthogonal optimum design of parameters of flux used for low carbon bainitic steel
  
  Author(s):Yang, Liang(1); Wang, Hong(2)
  Source: Applied Physics A: Materials Science and Processing  Volume: 126  Issue: 7  DOI: 10.1007/s00339-020-03730-z  Published: July 1, 2020  
  Abstract:L8 (27) orthogonal test table is applied to design eight kinds of flux, and arranged with SiO2, ZrO2 and TiO2 as three factors. The microstructure, morphology and mechanical properties of low carbon bainitic steels are investigated by means of optical microscope, scanning electron microscope, transmission electron microscope, tensile testing machine and instrumented drop weight impact tester with oscilloscope. The results indicate that the optimum composition rate is as follows: 20% SiO2, 10% ZrO2, and 6% TiO2, and the interaction between ZrO2 and TiO2 is obvious. The microstructure of deposited metal of low carbon bainitic steel is mainly formed by acicular ferrite and granular bainite. The acicular ferrite is paralleled distribution, and plenty of fine sheet Martensitic–Austenitic constituents are dispersing, which can hinder crack propagation and improve strength and toughness. The impact energy is up to 109.7 J at − 20 °C, and the ratio of brittle fracture termination load to maximum impact load is 0.378, ensuring excellent crack arrest toughness. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.
  Accession Number: 20202508856594
• Record 167 of
  
  Title:In-plane micro-displacement measurement based on secondary diffraction
  
  Author(s):Liu, Shengrun(1,2); Xue, Bin(1); Yu, Jirui(1,2); Xu, Guangzhou(1); Lv, Juan(1); Cheng, Ying(1); Yang, Jianfeng(1)
  Source: AIP Advances  Volume: 10  Issue: 4  DOI: 10.1063/1.5143339  Published: April 1, 2020  
  Abstract:For precision machinery, the measurement of the relative in-plane displacement of two parallel planes that are separated by several meters is important. In this paper, a theoretical model for measuring the relative in-plane microdisplacement between two parallel planes was developed on the basis of secondary diffraction. Based on this method, we employed a pinhole and a circular-ring as the diffraction screens. The influence of the structural parameters of diffraction screens on the secondary diffraction pattern was analyzed in detail, and the obtained parameters were then used in the experimental measurements. For experimental investigation, a laser beam at 532 nm was used to irradiate a pinhole; the diffracted light was then further diffracted using a circular-ring, and the final diffraction pattern was recorded using a CCD camera. The circular-ring was mounted on the plane to be measured, while the pinhole and the CCD camera remained stationary; the space between the pinhole and the circular-ring was set at 1200 mm. The displacement of the circular-ring can be calculated by comparing the central position of the two diffraction patterns before and after shifting the circular-ring. Over a measurement range of 0-90 μm, the absolute error in the displacement measurement was less than 1.97 μm. © 2020 Author(s).
  Accession Number: 20201608418587
• Record 168 of
  
  Title:Cascaded region proposal networks for proposal-based tracking
  
  Author(s):Zhang, Ximing(1); Fan, Xuewu(1); Luo, Shujuan(2)
  Source: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  Volume: 12015 LNCS  Issue:   DOI: 10.1007/978-3-030-54407-2_25  Published: 2020  
  Abstract:There still remains some problems which have not been solved in RPN-based trackers, including data imbalance, inappropriate proposals and poor robustness to spatial rotation even scale variation. We propose a cascaded region proposal network framework for visual tracking based on region proposal networks, spatial transformer networks and proposal selection strategy. We first to extract the features from deep and shallow layers via cascaded region proposal network to ensure the spatial information and semantic cue of the appearance model. Then, the feature extraction model based on spatial transformer networks is performed to calculate the parameters of spatial transformer and obtain the fused features. During the tracking and testing of proposed networks, the proposals are generated and re-ranked by formulating the proposals selection strategy to ensure the localization and scale of the estimated target. We extensively prove the effectiveness of the proposed method though the ablation studies of the tracking benchmark which include OTB2015, VOT2016 and UAV123. The experimental results perform that the accuracy and robustness of the proposed method as the real-time tracker and the long-term tracker as well. In the meantime, the test on the benchmark UAV123 shows that the tracker can be employed to some engineering area. © Springer Nature Switzerland AG 2020.
  Accession Number: 20203409067645