检索结果-内蒙古大学图书馆

International Joint Conference on Neural Networks (IJCNN)

作者： Gregory Ditzler Robi Polikar Signal Processing and Pattern Recognition Laboratory in the Electrical & Computer Engineering Department Rowan University Glassboro NJ USA

Learning in nonstationary environments, also called learning concept drift, has been receiving increasing attention due to increasingly large number of applications that generate data with drifting distributions. These applications are usually associated with streaming data, either online or in batches, and concept drift algorithms are trained to detect and track the drifting concepts. While concept drift itself is a significantly more complex problem than the traditional machine learning paradigm of data coming from a fixed distribution, the problem is further complicated when obtaining labeled data is expensive, and training must rely, in part, on unlabelled data. Independently from concept drift research, semi-supervised approaches have been developed for learning from (limited) labeled and (abundant) unlabeled data; however, such approaches have been largely absent in concept drift literature. In this contribution, we describe an ensemble of classifiers based approach that takes advantage of both labeled and unlabeled data in addressing concept drift: available labeled data are used to generate classifiers, whose voting weights are determined based on the distances between Gaussian mixture model components trained on both labeled and unlabeled data in a drifting environment.

关键词： Classification algorithms Data models Clustering algorithms Training data Training Testing Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

A block-based 2D Discrete Wavelet Transform structure with new scan method for overlapped sections

A block-based 2D Discrete Wavelet Transform structure with n...

引用

Middle East Conference on Biomedical engineering (MECBME)

作者： Sayed Ahmad Salehi Rasoul Amirfattahi Department of Electrical and Computer Engineering Digital Signal Processing Research Laboratory Isfahan University of Technology Isfahan Iran

Regarding internal memory size, we proposed a new scan method for block-based structure to provide an efficient parametric structure for two Dimensional Discrete Wavelet Transform (2-D DWT). The main challenge in 2-D DWT structure is the amount of internal memory required to produce wavelet coefficients. The previous structures need fixed memory size depend on input frame size, but in proposed architecture, memory size is parametric and in trade off with the number of output memory access. Lifting scheme is incorporated to implement DWT.

关键词： Discrete wavelet transforms Memory management Hardware Pixel

来源：评论

学校读者我要写书评

暂无评论

Evaluation of some physiological statements about seizure, using processing of epileptic EEG signals

Evaluation of some physiological statements about seizure, u...

引用

Iranian Conference on electrical engineering (ICEE)

作者： F. Shayegh R. Amirfattahi S. Sadri K. Ansari-Asl Digital signal Processing Laboratory Department of Electrical and Computer Engineering Isfahan University of Technology Iran Electrical Department Engineering Faculty Shahid Chamran University of Ahvaz Iran

Seizure genesis is usually attributed to imbalance of the excitation and inhibition activities of neurons. It is claimed that slow inhibition mechanism is responsible for controlling the variations of the excitation mechanism, and any deficiency in this function will lead to disturbance of their natural balance. Fast inhibition mechanism is also known to collaborate with slow inhibitory processes. Any proof about the excitation and inhibition mechanisms refers to the ex-vivo or in-vitro experiments, i.e. from the microscopic point of view. Up to now, there is not a straight road into evaluating these hypotheses from the macroscopic view. In this paper we try to interpret the excitation-inhibition sequences extracted from some depth-EEG signals through an identification-based procedure. The aim is to check out whether the statements gained from microscopic observations are re-established in population of neurons in the case of MTLE or not. Using Neyman-Pearson hypothesis test technique we try to answer these questions. We observe that even in population of neurons slow inhibition is responsible for compensating excitatory processes to balance inhibitory /excitatory ratio, such that when it is impaired causes seizures to arise; also the average fast inhibitory process of population of neurons follows its slow inhibition dynamics. Also, we check the differentiation of the excitation-inhibition ratio between normal and ictal states, as well as between normal and pre-ictal states. Ictal activity is properly a state of abnormal high excitation-inhibition ratio but before beginning of seizures there are not unique patterns.

关键词： Brain modeling Electroencephalography Noise Mathematical model Neurons Cost function Equations

来源：评论

学校读者我要写书评

暂无评论

New image compression algorithm using proposed quantization approach

New image compression algorithm using proposed quantization ...

引用

International Conference on Pattern Analysis and Intelligent Robotics (ICPAIR)

作者： Seyed Mehdi Moghadas Hossein Pourghasem Rasoul Amirfattahi Department of Electrical Engineering Islamic Azad University Najafabad Iran Digital Signal Processing Research Laboratory Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan Iran

A new quantization method is proposed in this paper. This method is useful for enhancement of compression quality when each kind of neural network is used to compress the image. By quantizing the image with the proposed method, the numbers of samples which must be reconstructed by neural network is reduced. This causes a remarkable increase in quality of the reconstructed image. For testing the proposed method we use autoassociative transform coding and by merging it with the proposed quantization method a new compression algorithm is obtained. Then results of compression by the merged method are compared with some previous works. Obtained results show that the proposed compression algorithm increases the compression quality of the images remarkably. Compression time and complexity in the merged method is also better than JPEG and make it suitable for the systems with low processor and hardware implementation.

关键词： Image coding Biological neural networks Image reconstruction Transform coding Neurons PSNR Training

来源：评论

学校读者我要写书评

暂无评论

Bayesian Spatiotemporal Multitask Learning for Radar HRRP Target Recognition

引用

IEEE Transactions on signal processing 2011年第7期59卷 3182-3196页

作者： Lan Du Penghui Wang Hongwei Liu Mian Pan Feng Chen Zheng Bao Department of Electrical and Computer Engineering Duke University Durham NC USA National Laboratory of Radar Signal Processing Xidian University Xi'an China School of Electronic Information and Electrical Engineering Shanghai Jiaotong University Shanghai China

A Bayesian dynamic model based on multitask learning (MTL) is developed for radar automatic target recognition (RATR) using high-resolution range profile (HRRP). The aspect-dependent HRRP sequence is modeled using a truncated stick-breaking hidden Markov model (TSB-HMM) with time-evolving transition probabilities, in which the spatial structure across range cells is described by the hidden Markov structure and the temporal dependence between HRRP samples is described by the time evolution of the transition probabilities. This framework imposes the belief that temporally proximate HRRPs are more likely to be drawn from similar HMMs, while also allowing for possible distant repetition or “innovation”. In addition, as formulated the stick-breaking prior and MTL mechanism are employed to infer the number of hidden states in an HMM and learn the target-dependent states collectively for all targets. The form of the proposed hierarchical model allows efficient variational Bayesian (VB) inference, of interest for large-scale problems. To validate the formulation, example results are presented for an illustrative synthesized dataset and our main application-RATR, for which we consider the measured HRRP data. For the latter, we also make comparisons to the model with the independent state-transition statistics and some other existing statistical models for radar HRRP data.

关键词： Hidden Markov models Data models Radar Bayesian methods Target recognition Analytical models

来源：评论

学校读者我要写书评

暂无评论

On gas detection and concentration estimation via mid-IR-based gas detection system analysis model

On gas detection and concentration estimation via mid-IR-bas...

引用

Annual Conference on Information Sciences and Systems (CISS)

作者： Yi Xin Joel M. Morris Communications and Signal Processing Laboratory (CSPL) Computer Science and Electrical Engineering Department University of Maryland Baltimore County Catonsville MD USA

In recent decades, Laser-based spectroscopy (LAS) has been used in a wide range of research and application fields due to developments in laser technology and infrared spectroscopy. A particular application of interest is mid-IR laser-based gas detection systems for health and environment assessment. In this paper, we use our statistical analysis model for a generic mid-IR pulsed-laser gas detection system to predict trace gas detection and concentration estimation performance, and their sensitivity to system parameters. Based on PNNL data and the Beer-Lambert law, we use the three main spectral peaks of a trace gas, as the basis for gas detection, and use the relationship between gas transmittance β, molar absorptivity ε, concentration, and the sample-mean measurement, x N , from the photodetector, as the basis for concentration estimation using a standard confidence interval method.

关键词： Gas lasers Estimation Optical pulses Photodetectors Laser modes Cavity resonators

来源：评论

学校读者我要写书评

暂无评论

Effect of Different Diffusion Maps on Registration Results

Effect of Different Diffusion Maps on Registration Results

引用

Iranian Machine Vision and Image processing (MVIP)

作者： Maryam Afzali Emad Fatemizadeh Hamid Soltanian-Zadeh Department of Electrical Engineering Biomedical Signal and Image Processing Laboratory (BiSIPL) Sharif University of Technology Tehran Iran Image Analysis Laboratory Radiology Department Henry Ford Health System Detroit MI USA Control and Intelligent Processing Center of Excellence (CIPCE) School of Electrical and Computer Engineering University of Tehran Tehran Iran

In this paper, we compare registration results obtained using different diffusion maps extracted from diffusion tensor imaging (DTI). Fractional Anisotropy (FA) and Ellipsoidal Area Ratio (EAR) are two diffusion maps (indices) that may be used for image registration. First, we use FA maps to find deformation matrix and register diffusion weighted images. Then, we use EAR maps and finally we use both of FA and EAR maps to register diffusion weighted images. The difference between FA values before deformation and after registration using the FA alone or EAR alone has a median of 0.57 and using both of them has a median of 0.29. Therefore, the results of registration using both of the FA and EAR indices are superior to those obtained using only one of them alone.

关键词： Ear Diffusion tensor imaging Tensile stress Registers Eigenvalues and eigenfunctions Anisotropic magnetoresistance Image registration

来源：评论

学校读者我要写书评

暂无评论

Design, fabrication and Helium Ion Microscope patterning of suspended nanomechanical graphene structures for NEMS applications

Design, fabrication and Helium Ion Microscope patterning of ...

引用

International Conference on Solid State Sensors and Actuators (TRANSDUCERS)

作者： M. Annamalai S. Mathew V. Viswanathan C. Fang D. S. Pickard M. Palaniapan Centre of Integrated Circuit Failure Analysis & Reliability Singapore Plasmonics & Advanced Imaging Technology Laboratory Department of Electrical and Computer Engineering National University of Singapore Singapore Signal Processing & VLSI Lab Singapore

This paper investigates the potential uses of graphene nanomechanical devices for NEMS applications fabricated using Helium Ion Microscope (HIM). Suspended nanomechanical graphene drum structures of diameter (~2 - 3 μm) and thickness ranging from few atomic layers (~10 Å) to about 11 nm have been fabricated and characterized using Atomic Force Microscopy (AFM). The initial suspension profile and the thickness of the fabricated structures have been obtained using AFM. FEM results suggest that these structures can vibrate in the range of several MHz. Their small size and the higher operating frequencies make them well suitable for mass sensing applications with sensitivities greater than 10 -21 g Hz -1 which is more than three orders of magnitude improvement over silicon based structures. The initial suspended graphene membranes have been patterned using HIM to obtain multiple z-axis flexures with sub-10nm feature sizes.

关键词： Silicon Nanoelectromechanical systems Helium Microscopy Nanostructures Gold Fabrication

来源：评论

学校读者我要写书评

暂无评论

Upper Limb Joint Angle Tracking with Inertial Sensors

Upper Limb Joint Angle Tracking with Inertial Sensors

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society

作者： Mahmoud El-Gohary Lars Holmstrom Jessie Huisinga Edward King James McNames Fay Horak Biomedical Signal Processing Laboartory in the Department of Electrical and Computer Engineering at Portland State University (PSU) Chief Information Officer at APDM Inc. Balance Disorders Laboratory in the Department of Neurology at Oregon Health & Science University (OHSU)

ISBN: (纸本)9781424441211

Wearable inertial systems have recently been used to track human movement in and outside of the laboratory. Continuous monitoring of human movement can provide valuable information relevant to individual's level of physical activity and functional ability. Traditionally, orientation has been calculated by integrating the angular velocity from gyroscopes. However, a small drift in the measured velocity leads to large integration errors that grow with time. To compensate for that drift, complementary data from accelerometers are normally fused into the tracking systems using the Kalman or extended Kalman filter (EKF). In this study, we combine kinematic models designed for control of robotic arms with the unscented Kalman filter (UKF) to continuously estimate the angles of human shoulder and elbow using two wearable sensors. This methodology can easily be generalized to track other human joints. We validate the method with an optical motion tracking system and demonstrate correlation consistently greater than 0.9 between the two systems.

关键词： inertia sensor Human locomotion TRACKING SYSTEM robotic arm kinematics joint angle accelerometers Elbow Upper Extremity inertial systems Motor Activity

来源：评论

学校读者我要写书评

暂无评论

Detection of Collagenous Colitis Based on Histopathology Image Segmentation of Colon

Detection of Collagenous Colitis Based on Histopathology Ima...

引用

Iranian Machine Vision and Image processing (MVIP)

作者： Vahid Malekian Mojgan Mokhtari Saeid Sadri Rasoul Amirfattahi Alireza Aghaie Digital Signal Processing Research Laboratory Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan Iran School of Medicine Isfahan University of Medical Sciences Isfahan Iran Isfahan University of Technology Isfahan Isfahan IR

Collagenous colitis is prevalent among pathological diseases. Because of normal radiological and endoscopic appearances, the microscopic image of colon tissue is necessary and important material for investigation of this disease. In this paper, a computerized system is developed for collagenous colitis diagnosis. First, colon tissues are colored with Mason trichrome to make sub-epithelial collagen band to be appeared in blue. Second, preprocessing is done on the microscopic image of colon tissue, and then Color histogram features in HSV color space and Haar Wavelet energy in detail coefficients of sub-band images are implemented to characterize the histological structure representation for tissue classification. KNN Classifier is selected for this classification, which segments the sub-epithelial collagen band from the image, and finally, thickness of sub-epithelial collagen band is determined by morphological operations. Abnormal case will be recognized if the thickness of sub-epithelial collagen band be more than 10 μm. Fifty clinical image samples were employed in the training and testing procedures for the evaluation of system performance. Through the applied classification it was revealed that collagenous colitis can be diagnosed accurately.

关键词： Microscopy Image color analysis Colon Feature extraction Image segmentation Training Histograms

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：