检索结果-内蒙古大学图书馆

Design of temperature compensation for silicon-sapphire pressure sensor

学校读者我要写书评

暂无评论

Design of temperature compensation for silicon-sapphire pres...

2017 IEEE International Conference on Imaging systems and Techniques, IST 2017

作者： Manguo, Huang Defeng, Liu Chao, Lu Zhanshe, Guo Xin, Li Advanced Sensor Technology Center AVIC Beijing Changcheng Aeronautic Measurement Control Technology Research Institute Aviation Key Laboratory of Science and Technology Special Condition Monitoring Sensor Technology Beijing China School of Instrument Science and Opto-electronics Engineering State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China

ISBN: (纸本)9781538616208

In order to solve the problem that the temperature drift of the silicon-on-sapphire pressure sensor is too large to reduce the accuracy of pressure measurement at high temperature, a temperature compensation scheme based on a single chip is proposed. The temperature of the working environment is obtained by a PT100, and integrated with the output of pressure sensor to calculate the compensated pressure sensor. Problems of power supply and signal acquisition are solved by the high precision, low drift voltage and current references, programmable amplifiers and high precision ADCs on chip. According to the working characteristics of the silicon-on-sapphire pressure sensor, an efficient temperature compensation algorithm is designed, and written to the on-chip ARM processor. The experimental results show that the maximum measurement error of pressure in the temperature range of -20∼250°C and measurement range of 0∼28 MPa is 8.37%FS before compensation and reduced to 0.3%FS after compensation. The accuracy of the sensor is greatly improved by the compensation system. © 2017 IEEE.

关键词： Pressure sensors

Learning a saliency evaluation metric using crowdsourced perceptual judgments

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Xia, Changqun Li, Jia Su, Jinming Borji, Ali State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China International Research Institute for Multidisciplinary Science Beihang University Beijing China Center for Research in Computer Vision Computer Science Department University of Central Florida OrlandoFL United States

In the area of human fixation prediction, dozens of computational saliency models are proposed to reveal certain saliency characteristics under different assumptions and definitions. As a result, saliency model benchmarking often requires several evaluation metrics to simultaneously assess saliency models from multiple perspectives. However, most computational metrics are not designed to directly measure the perceptual similarity of saliency maps so that the evaluation results may be sometimes inconsistent with the subjective impression. To address this problem, this paper first conducts extensive subjective tests to find out how the visual similarities between saliency maps are perceived by humans. Based on the crowdsourced data collected in these tests, we conclude several key factors in assessing saliency maps and quantize the performance of existing metrics. Inspired by these factors, we propose to learn a saliency evaluation metric based on a two-stream convolutional neural network using crowdsourced perceptual judgements. Specifically, the relative saliency score of each pair from the crowdsourced data is utilized to regularize the network during the training process. By capturing the key factors shared by various subjects in comparing saliency maps, the learned metric better aligns with human perception of saliency maps, making it a good complement to the existing metrics. Experimental results validate that the learned metric can be generalized to the comparisons of saliency maps from new images, new datasets, new models and synthetic data. Due to the effectiveness of the learned metric, it also can be used to facilitate the development of new models for fixation prediction. Copyright © 2018, The Authors. All rights reserved.

关键词： Crowdsourcing

Autonomous Aerial Manipulation Using a Hexacopter Equipped with a Robotic Arm

学校读者我要写书评

暂无评论

Autonomous Aerial Manipulation Using a Hexacopter Equipped w...

IEEE International Conference on robotics and Biomimetics

作者： Ran Jiao Mingjie Dong Wusheng Chou Hailong Yu Hao Yu College of Mechanical Engineering and Automation Beihang University Beijing P.R. China College of Mechanical Engineering and Applied Electronics Technology Beijing University of Technology Beijing P.R. China The State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing P.R. China

ISBN: (纸本)9781728103785;9781728103778

This paper presents the implementation of an autonomous aerial manipulation using a hexacopter equipped with a two DOF robotic arm. The kinematic and dynamic models are developed by considering the dynamic characteristics of the combined manipulation platform. A novel adaptive sliding mode controller is proposed for both position and velocity control. By building SSD (Single Shot Detection) deep neural network based on deep learning, an object detection solution is developed. The three dimensional coordinates of the target object relative to the multirotor are obtained via combining the vertical plane position obtained from object detection law with the depth value from stereo camera. Finally, with the proposed controller and object detection law, an autonomous flight experiment is accomplished including approaching, grabbing and delivering the target object. The proposed approaches are demonstrated with effectiveness and could be utilized in various manipulation applications.

关键词： Manipulator dynamics Vehicle dynamics Object detection Robot kinematics Kinematics

How drones look: Crowdsourced knowledge transfer for aerial video saliency prediction

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Fu, Kui Li, Jia Shen, Hongze Tian, Yonghong State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China International Research Institute for Multidisciplinary Science Beihang University Beijing100191 China School of Electronics Engineering and Computer Science Peking University Beijing100871 China Cooperative Medianet Innovation Center China

In ground-level platforms, many saliency models have been developed to perceive the visual world as the human does. However, they may not fit a drone that can look from many abnormal viewpoints. To address this problem, this paper proposes a Crowdsourced Multi-path Network (CMNet) that transfer the ground-level knowledge for spatiotemporal saliency prediction in aerial videos. To train CMNet, we first collect and fuse the eye-tracking data of 24 subjects on 1,000 aerial videos to annotate the ground-truth salient regions. Inspired by the crowdsourced annotation in eye-tracking experiments, we design a multi-path architecture for CMNet, in which each path is initialized under the supervision of a classic ground-level saliency model. After that, the most representative paths are selected in a data-driven manner, which are then fused and simultaneously fine-tuned on aerial videos. In this manner, the prior knowledge in various classic ground-level saliency models can be transferred into CMNet so as to improve its capability in processing aerial videos. Finally, the spatial predictions given by CMNet are adaptively refined by incorporating the temporal saliency predictions via a spatiotemporal saliency optimization algorithm. Experimental results show that the proposed approach outperforms ten state-of-the-art models in predicting visual saliency in aerial videos. Copyright © 2018, The Authors. All rights reserved.

关键词： Crowdsourcing

The Data-Reusing MCC-Based Algorithm and Its Performance Analysis

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2016年第4期25卷 719-725页

作者： LIU Chunhui QI Yue DING Wenrui Research Institute of Unmanned Aerial Vehicle Beihang University State Key Laboratory of Virtual Reality Technology and System Beihang University

Maximum correntropy criterion(MCC)provides a robust optimality criterion for non-Gaussian signal *** this paper,the weight update equation of the conventional MCC-based adaptive filtering algorithm is modified by reusing the past K input vectors,forming a class of data-reusing MCC-based algorithm,called DRMCC *** with the conventional MCCbased algorithm,the DR-MCC algorithm provides a much better convergence performance when the input data is *** mean-square stability bound of the DRMCC algorithm has been studied *** both Gaussian noise case and non-Gaussian noise case,the expressions for the steady-state Excess mean square error(EMSE) of DR-MCC algorithm have been *** relationship between the data-reusing order and the steadystate EMSEs is also *** results are in agreement with the theoretical analysis.

关键词： Adaptive filtering Maximum correntropy criterion(MCC) Data-reusing Mean-square stability Steady-state excess mean square error

Spatial-Temporal Editing for Dynamic Hair Data

学校读者我要写书评

暂无评论

Spatial-Temporal Editing for Dynamic Hair Data

International Conference on virtual reality and Visualization (ICVRV)

作者： Yijie Wu Yongtang Bao Yue Qi State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Beihang University Qingdao Research Institute Qingdao China

ISBN: (纸本)9781538626375;9781538626368

Hair plays a unique role in depicting a person's character. Currently, most hair simulation techniques require a lot of computation time, or rely on complex capture settings. Editing and reusing of existing hair model data are very important topics in computer graphics. In this paper, we present a spatialtemporal editing technique for dynamic hair data. This method can generate a longer or even infinite length sequence of hair motion according to its motion trend from a short input. Firstly, we build spatial-temporal neighborhood information about input hair data. We then initialize the output according to the input exemplar and output constraints, and optimize the output through iterative search and assignment steps. To make the method be more efficient, we select a sparse part of the hair as the guide hair to simplify the model, and interpolate a full set of hair after the synthesis. Results show that our method can deal with a variety of hairstyles and different way of motions.

关键词： Hair Mathematical model Solid modeling Dynamics Computational modeling Market research Interpolation

Low-resolution face recognition in the wild via selective knowledge distillation

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Ge, Shiming Zhao, Shengwei Li, Chenyu Li, Jia Institute of Information Engineering Chinese Academy of Sciences Beijing100095 China Institute of Information Engineering Chinese Academy of Sciences School of Cyber Security at University of Chinese Academy of Sciences China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China

Typically, the deployment of face recognition models in the wild needs to identify low-resolution faces with extremely low computational cost. To address this problem, a feasible solution is compressing a complex face model to achieve higher speed and lower memory at the cost of minimal performance drop. Inspired by that, this paper proposes a learning approach to recognize low-resolution faces via selective knowledge distillation. In this approach, a two-stream convolutional neural network (CNN) is first initialized to recognize high-resolution faces and resolution-degraded faces with a teacher stream and a student stream, respectively. The teacher stream is represented by a complex CNN for high-accuracy recognition, and the student stream is represented by a much simpler CNN for low-complexity recognition. To avoid significant performance drop at the student stream, we then selectively distil the most informative facial features from the teacher stream by solving a sparse graph optimization problem, which are then used to regularize the fine-tuning process of the student stream. In this way, the student stream is actually trained by simultaneously handling two tasks with limited computational resources: approximating the most informative facial cues via feature regression, and recovering the missing facial cues via low-resolution face classification. Experimental results show that the student stream performs impressively in recognizing low-resolution faces and costs only 0.15MB memory and runs at 418 faces per second on CPU and 9, 433 faces per second on GPU. Copyright © 2018, The Authors. All rights reserved.

关键词： Students

Kinetic Simulation of Cardiac Motion with Patient-Specific Coronary Artery Vessels Attached for PCI Simulator

学校读者我要写书评

暂无评论

Kinetic Simulation of Cardiac Motion with Patient-Specific C...

International Conference on virtual reality and Visualization (ICVRV)

作者： Zhijun Xie Shuai Li Qing Xia Aimin Hao State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Beihang University Qingdao Research Institute Qingdao China

ISBN: (纸本)9781538626375;9781538626368

In this paper, we propose to simulate the dynamic motion of cardiovascular system attached with patient-specific vessel structure for personalized Percutaneous Coronary Intervention (PCI) simulation to train surgeons of skills and to help planning surgery. To obtain patient-specific vessel structure, a coarse segmentation with the centerlines extraction subsequently is applied to the computed tomography (CT) scans and the vessels along the centerlines is modeled using a lofted 2D segmentation method. The vessels are then combined with a template heart model to construct a cardiovascular system. For the cardiac motion, we estimate the ventricles motion from 4D Magnetic Resonance Imaging (MRI) sequences to drive the whole heart motion. And the position-based method coupling with a mass-spring model constructed with elastic spheres is used to simulate the cardiac motion cycle stably in the interactive PCI simulator. With our method, a personalized highly realistic beating motion of a whole heart is able to be created and applied to our patient-specific PCI surgery simulation system.

关键词： Solid modeling Heart Computational modeling Deformable models Surgery Strain Magnetic resonance imaging

R3-Net: A deep network for multi-oriented vehicle detection in aerial images and videos

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Li, Qingpeng Mou, Lichao Xu, Qizhi Zhang, Yun Zhu, Xiao Xiang State Key Laboratory of Virtual Reality Technology and Systems Beijing Key Laboratory of Digital Media School of Computer Science and Engineering Beihang University Beijing100191 China Remote Sensing Technology Institute German Aerospace Center Wessling82234 Germany Signal Processing in Earth Observation Technical University of Munich Munich80333 Germany Canada Research Chair Laboratory in Advanced Geomatics Image Processing Department of Geodesy and Geomatics Engineering University of New Brunswick FrederictonNBE3B 5A3 Canada

Vehicle detection is a significant and challenging task in aerial remote sensing applications. Most existing methods detect vehicles with regular rectangle boxes and fail to offer the orientation of vehicles. However, the orientation information is crucial for several practical applications, such as the trajectory and motion estimation of vehicles. In this paper, we propose a novel deep network, called rotatable region-based residual network (R3-Net), to detect multi-oriented vehicles in aerial images and videos. More specially, R3-Net is utilized to generate rotatable rectangular target boxes in a half coordinate system. First, we use a rotatable region proposal network (R-RPN) to generate rotatable region of interests (R-RoIs) from feature maps produced by a deep convolutional neural network. Here, a proposed batch averaging rotatable anchor (BAR anchor) strategy is applied to initialize the shape of vehicle candidates. Next, we propose a rotatable detection network (R-DN) for the final classification and regression of the R-RoIs. In RDN, a novel rotatable position sensitive pooling (R-PS pooling) is designed to keep the position and orientation information simultaneously while downsampling the feature maps of RRoIs. In our model, R-RPN and R-DN can be trained jointly. We test our network on two open vehicle detection image datasets, namely DLR 3K Munich Dataset and VEDAI Dataset, demonstrating the high precision and robustness of our method. In addition, further experiments on aerial videos show the good generalization capability of the proposed method and its potential for vehicle tracking in aerial videos. The demo video is available at https://***/xCYD-tYudN0. Copyright © 2018, The Authors. All rights reserved.

关键词： Remote sensing