检索结果-内蒙古大学图书馆

28th International conference on Computer Graphics and vision, GraphiCon 2018

作者： Gordeev, D.S. Institute of Informatics Systems SB RAS Novosibirsk Russia

This paper describes the solution of problem of visualization of changes in graph model of internal representation of programs in order to reflect processes which happens during calculation of programs or processing with graph algorithms. We use model which interprets every change of graph model as corresponding change of visual graphical styles, which corresponds to changes of attributes of graph model elements or to structure of graph model. Each single changing of an attribute value, adding of node to graph model or removing of edge from graph model is reflected over visual representation as animation of graphical styles of pieces of image. The key point of described model is an idea of context of visual animation as surrounding visual information in relation to happening change in graph model. © GraphiCon 2018 - 28th International conference on Computer Graphics and vision. All rights reserved.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

The Tangential Velocity MTI algorithms in Space-borne systems for Remote Sensing of the Earth

引用

Journal of Physics: conference Series 2020年第1期1632卷

作者： V V Kostrov E F Tolstov K K Khramov Murom Institute of Vladimir State University Murom Russia Head of department AEROCON Company Zhukovsky Russia

In this paper, the problem of moving target indication (MTI) using synthetic aperture radar (SAR) is considered. The focus of the article is the tangential component of velocity. Two tangential velocity MTI algorithms are considered. The first algorithm uses two apertures with various synthetic time of the radar image (AVST algorithm), and the second uses two apertures displaced along trajectory (ADAT algorithm). The structure of the MTI system based on the analysis of phase and amplitude radar images is considered. For S band and X band SAR, the phase change in the trajectory signal of a moving target, the effects of shift and bifurcation of target responses on the radar image are analyzed in detail. It was found that the AVST algorithm has a small working range of unambiguous velocity estimate (up to ±10 m/s). It is shown that the ADAT algorithm has a higher quality of work in a wide velocity range and can effectively suppress the signals of stationary objects by 20...30 dB. The obtained characteristics allow us to make demands on the parameters of space-borne systems for remote sensing of the Earth and processing systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Smart Annotation Tool for Multi-sensor Gait-based Daily Activity Data

Smart Annotation Tool for Multi-sensor Gait-based Daily Acti...

引用

IEEE International conference on Pervasive Computing and Communications (PerCom)

作者： Martindale, Christine F. Roth, Nils Hannink, Julius Sprager, Sebastijan Eskofier, Bjoern M. Friedrich Alexander Univ Erlangen Nurnberg Machine Learning & Data Analyt Lab Comp Sci Dept Erlangen Germany Univ Ljubljana Fac Comp & Informat Sci Ljubljana Slovenia

ISBN: (纸本)9781538632277

The monitoring of patients within a natural, home environment is important in order to close knowledge gaps in the treatment and care of neurodegenerative diseases, such as quantifying the daily fluctuation of Parkinson's patients' symptoms. The combination of machine learning algorithms and wearable sensors for gait analysis is becoming capable of achieving this. However, these algorithms require large, labelled, realistic datasets for training. Most systems used as a ground truth for labelling are restricted to the laboratory environment, as well as being large and expensive. We propose a study design for a realistic activity monitoring dataset, collected with inertial measurement units, pressure insoles and cameras. It is not restricted by a fixed location or capture volume and still enables the labelling of gait phases or, where non-gait movement such as jumping occur: on-the-ground, off-the-ground phases. Additionally, this paper proposes a smart annotation tool which reduces annotation cost by more than 80%. This smart annotation is based on edge detection within the pressure sensor signal. The tool also enables annotators to perform assisted correction of these labels in a post-processing step. This system enables the collection and labelling of large, fairly realistic datasets where 93% of the automatically generated labels are correct and only an additional 10% need to be inserted manually. Our tool and protocol, as a whole, will be useful for efficiently collecting the large datasets needed for training and validation of algorithms capable of cyclic human motion analysis in natural environments.

关键词： image edge detection Tools Cameras Labeling Pressure sensors Intelligent sensors

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Model for Error Modeling with Pseudoinverse Learning Algorithm

An Ensemble Model for Error Modeling with Pseudoinverse Lear...

引用

IEEE International conference on systems, Man and Cybernetics

作者： Sibo Feng Xiaodan Deng Ping Guo Bo Zhao Qian Yin Hongfeng Wang Image Processing and Pattern Recognition Laboratory Beijing Normal University Beijing China School of Information Management Dezhou University Shandong China

In Bayesian theory, the maximum posterior estimator uses prior information to estimate the noise in the machine learning model by adding the regularization term. The regularization terms L 1 and L 2 correspond to Laplacian prior and Guassian prior, respectively. In existing deep learning models, in order to use the gradient descent optimization algorithm and achieve good results, most models take L 2 regularization as the regularization term of the network model to fit the complex Guassian noise. However in practice, the Laplace noise and the Guassian noise are both considered as data noise. For multi-layer perceptrons, the difficulty caused by adding L 1 and L 2 into the optimization function of the network is solved by proposing an ensemble model for error modeling through adopting the divide and conquer strategy. First, several base learners are trained to fit different noise distributions of data, then the final results can be obtained by taking the results of each base leaner as new data to train a meta leaner, and get the final results. Among them, coordinate regression method is used to solve L 1 loss, while the pseudo-inverse learning algorithm is employed to solve L 2 loss. Both methods are nongradient optimization algorithms. The comparison results of the model on several data sets show that the proposed ensemble model achieves better performance.

关键词： Training Optimization Data models Neural networks Computational modeling Matrix decomposition Estimation

来源：评论

学校读者我要写书评

暂无评论

Sensitivity Estimation and image Reconstruction for Sparse PET with Deep Learning

Sensitivity Estimation and Image Reconstruction for Sparse P...

引用

IEEE Nuclear Science Symposium and Medical Imaging conference (NSS/MIC) / 25th International Symposium on Room-Temperature Semiconductor X-Ray and Gamma-Ray Detectors

作者： Feng, Tao Wang, Jizhe Li, Hongdi UIH Amer Inc Houston TX 77054 USA

ISBN: (纸本)9781538684948

The use of a sparse crystal setting would reduce the cost of the PET scanner and has advantages such as less RF shielding in PET/MR. It also allows a longer axial field of view (FOV) using the same crystal volume. In this paper, the sensitivities of the coincidence events of PET systems with the sparse crystal configuration, thin crystal setting, and the conventional design using a fixed total crystal volume were analytically estimated. The sinograms of a sparse system (with 50% crystal removed and fixed axial FOV) were simulated using patient data. Reconstruction algorithms were developed by modeling the effects of reduced crystals in the system matrix. A convolutional neural network (CNN) based noise reduction approach was used for post-processing. A total of 14 patient data were included and were truncated to 3 minutes scan for consistency. Leave-one-out cross- validation was used for evaluation purpose. A patch based data input/output was used for model training to increase the number of training samples. images reconstructed using OSEM followed by Gaussian denoising was also used as a comparison. The percentage summed square difference (SSD) between images of sparse crystal configuration and non-sparse systems were used for quantitative evaluation. When using the same total volume of crystals, the difference of sensitivity at the center of FOV was within 10% among three different settings, with the rank from highest to lowest being the thin detector, sparse detector, and conventional detector. When using the same axial FOV, reconstructed images of the sparse crystal configuration showed increased noise due to reduced sensitivity. The percentage SSD for image processed with the Gaussian filter was 30% on average and was reduced to 16% with CNN on average. The results show with the same amount of crystal, the use of sparse crystal configuration provides a slightly larger sensitivity and much larger axial FOV. CNN processed images was able to partially recover los

关键词： Crystals Sensitivity image reconstruction Detectors Noise reduction Training image quality

来源：评论

学校读者我要写书评

暂无评论

Driver safety approach using efficient image processing algorithms for driver distraction detection and alerting 6th

Driver safety approach using efficient image processing algo...

引用

6th International conference on Frontiers of Intelligent Computing: Theory and Applications, FICTA-2017

作者： Wathiq, Omar Ambudkar, Bhavna D. E&TC Dr. D.Y. Patil Institute of Engineering & Technology Pimpri Pune411018 India

ISBN: (纸本)9789811075650

Currently, due to different reasons, the road accidents are increasing. Road accidents are prone to number human deaths. There are different reasons which lead to road accidents, but drivers fatigue or distraction is main threat in major accidental cases. Therefore, recently various methods are explained by many authors for untimely identification of driver sleepiness in the manner of prohibiting mischance on road. In this paper, we are presenting the novel approach called hybrid method in which automatic care of driver safety and hospitality management services. Our approach aims at determining first if a driver is distracted or not based yawing, eye position, head position, mouth position etc., second if driver is detected as distracted instance alarming will perform on both driver side and near hospital services in order to be available in case of accident happen. Based on computer vision techniques, we propose four different modules for features extraction, focusing on arm position, face orientation, facial expression and eye behaviour, and then, the outputs of all these phases combined together and feed to the classifier feed-forward neural network (FFNN) for alarming the distraction detection and type of distraction. The outcome of this paper is efficient driver safety approach by considering the RGB-D sensor and image processing algorithms. © Springer Nature Singapore Pte Ltd. 2018.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Fully automated CADx for early breast cancer detection using image processing and machine learning 30

Fully automated CADx for early breast cancer detection using...

引用

30th International conference on Microelectronics, ICM 2018

作者： Gamil, Monica Ezzat Mohamed Fouad, Mariam Abd El Ghany, Mohamed A. Hoffinan, Klaus Electronics Engineering Dept German University in Cairo Cairo Egypt Electronics Department German University Cairo Egypt Integrated Electronic Systems Lab TU Darmstadt Germany

ISBN: (纸本)9781538681671

Breast cancer accounts for 16% of all cancers among females. Current early detection methods are expensive or computationally complex and thus unsuitable for developing countries. For this reason, a real-time fully automated Computer Aided Diagnosis system for Breast Cancer early detection from Ultrasound images is built in this paper. The proposed and implemented design comprises into its modules state of the art techniques and methods. The implemented design includes preprocessing/filtering of the input ultrasound image, segmentation of the region of interest from the background image and feature set calculation/extraction. Machine learning algorithms were implemented for classification of the tumour. Successful implementation with satisfactory run time is achieved with a final accuracy improved by 10% from previous work using the same set of features. Additional evaluation metrics like precision-recall plots and confusion matrices were also used to test and evaluate the system overall balanced performance. © 2018 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Laboratory Approbation of the Algorithm for Isolating People's Faces on a Thermal Infrared image in the Case of Their Quasi-Stationary Arrangement in a Room 8

Laboratory Approbation of the Algorithm for Isolating People...

引用

8th Annual International conference of the Biologically-Inspired-Cognitive Architectures-Society on Biologically Inspired Cognitive Architectures (BICA)

作者： Alyushin, M., V Alyushin, A. M. Kolobashkina, L., V Natl Res Nucl Univ MEPhI Moscow Engn Phys Inst Moscow Russia

Increasing the effectiveness of training and training sessions is possible through the implementation of so-called biological feedback. Such feedback allows the teacher, or the instructor, to continuously monitor the current psycho-emotional and functional state of the students. As a result, it becomes possible to adapt the style, pace, training mode and the volume of the material outlined, depending on the current receptivity and fatigue level of the listeners. The main element of systems that implement biological feedback in practice are remote non-contact technologies. Such technologies allow in a fully automatic mode to register the main most informative human bio-parameters. Among them, in the first place are the parameters characterizing the current state of the cardiovascular system of man, his breathing system, as well as his peripheral nervous system. The bulk of information is obtained by processing in real time the thermal infrared image of a person's face. Unfortunately, existing algorithms for distinguishing a person's face have a sufficiently high computational complexity and insufficient reliability. A typical example in this regard can be a family of algorithms based on the viola-Jones approach. The approach proposed in the work is based on taking into account additional information about the most likely location of a person's face on a thermal image. This approach is most appropriate to use in cases of quasi-stationary location of people in the room. A typical example is the location of students at the tables in the classroom. For such cases it is possible to determine the areas of the most probable location of the trainees' faces, as well as the possible boundaries of their movement. Laboratory tests of the developed program on the basis of the proposed algorithm have confirmed its high productivity, as well as efficiency in identifying students faces in the classroom. (C) 2018 The Authors. Published by Elsevier Ltd. This is an open access article

关键词： Thermal image of the face Face Selection Algorithm Quasi-Stationary Arrangement of People

来源：评论

学校读者我要写书评

暂无评论

Monomodal image Registration by Tensor Analysis: Going beyond the Brightness Constancy Assumption 15

Monomodal Image Registration by Tensor Analysis: Going beyon...

引用

15th IEEE/ACS International conference on Computer systems and Applications, AICCSA 2018

作者： Moran, M.B.H. Conci, A. Gonzalez, J.R. Saade, D.C.M. Fiirst, W. Araujo, A.S. Damiao, Charbel Lima, Giovanna Filho, Rubens Cruz Resmini, Roger Institute of Computing Universidade Federal Fluminense Niteroi Brazil Faculty of Medicine Hospital Universitário Antônio Pedro Universidade Federal Fluminense Niteroi Brazil Exact and Natural Sciences Institute Universidade Federal de Mato Grosso Rondonopolis Brazil

ISBN: (纸本)9781538691205

There is a great number of image applications, where the first step in the analysis is to associate the same point in two or more images that were taken from different viewpoints or at different times. This procedure is called image registration (IR). This work proposes a new efficient method for IR based on tensor calculus that overtakes the usual assumption of brightness consistency among the involved images. The main idea is to define transformations in serial images by using tensor behavior on coordinate transformations. It combines the simplicity and efficiency of the computer implementation with the correctness and elegance of the physic foundations. Moreover, it introduces a complete different way for management of this very important application of image processing: the use of the principal axis idea. The significant advantage of the proposed method is its promptitude and robustness for use in medical images in a way the as far as we known have never considered before. Experimental results in real examinations demonstrate that the proposed algorithm is highly accurate and outperforms a combination of the most updated IR approaches. © 2018 IEEE.

关键词： Luminance

来源：评论

学校读者我要写书评

暂无评论

Synchronized 2D SLAM and 3D Mapping Based on Three Wheels Omni-directional Mobile Robot

Synchronized 2D SLAM and 3D Mapping Based on Three Wheels Om...

引用

IEEE International conference on Cyber Technology in Automation, Control, and Intelligent systems

作者： Sibo Huang Chong Li Zhaoquan Cai Guijie Zhu Li Yao Zhun Fan Guangdong Provincial Key Laboratory of Digital Signal and Image Processing College of Engineering Shantou University Shantou China Huizhou University Huizhou China

ISBN: (数字)9781728107707

ISBN: (纸本)9781728107714

In this paper, we design a three wheels Omni-directional mobile robot (TOMR) and propose a method of simultaneous construction of 2D and 3D maps based on the mobile robot. To be more specific, we use information from a laser and a kinect to build 2D grid maps and 3D environment, respectively. The particle filter algorithm is used to achieve the pose of the robot, together with the OctoMap which is generated from a 3D point cloud map, to construct the 2D and 3D maps. An asymmetric environment is employed to test our proposed method and some state-of-the-art methods like RGB-D SLAM and ORB-SLAM. The experimental results show that the proposed method is efficient for synchronized 2D and 3D mapping and has better performance than other compared algorithms.

关键词： Three-dimensional displays Mobile robots Two dimensional displays Robot kinematics Wheels Simultaneous localization and mapping

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：