The analysis and segmentation of an electrocardiogram (ECG) signal is a hard and difficult task due to its artifacts, noise and form. In this paper;we analyze the ECG signal in Frequency, applying Fourier transform, a...
详细信息
ISBN:
(纸本)9781424406272
The analysis and segmentation of an electrocardiogram (ECG) signal is a hard and difficult task due to its artifacts, noise and form. In this paper;we analyze the ECG signal in Frequency, applying Fourier transform, autoregressive moving average (ARMA), Multiple signal Classifications (MUSIC), as well as the short-term Fourier transform STFT, Choi-Williams and Wigner-Ville for Time frequency analysis and wavelet analysis. The analysis has been done in modified lead II (MLII) of ECGs data files of the MIT-BIH database, obtaining better results of segmentation of QRS complex by wavelet analysis.
Multiresolution estimates of classification complexity estimate the relative ease with which multivariate data belonging to multiple classes can be separated by non-linear boundaries in high dimensional spaces. In thi...
详细信息
ISBN:
(纸本)0889865477
Multiresolution estimates of classification complexity estimate the relative ease with which multivariate data belonging to multiple classes can be separated by non-linear boundaries in high dimensional spaces. In this paper we propose the concept of using multiple classifiers in feature subspaces that are generated by feature space partitioning. We find that the advantage gained by training multiple classifiers for a given data set is far greater than the disadvantage of having less number of samples in each feature subspace to train them. In this paper we take a number of data sets from the UCI repository and show the classification advantage gained by using multiple subspace classifiers in parallel. We also demonstrate that the multi-resolution estimates of classification complexity correlate well with this classification performance averaged across all subspaces.
Content based image and video retrieval research focuses on the development of novel features and similarity metrics for improving the retrieval performance. There are only a few well-established data benchmarks on wh...
详细信息
ISBN:
(纸本)0889865477
Content based image and video retrieval research focuses on the development of novel features and similarity metrics for improving the retrieval performance. There are only a few well-established data benchmarks on which video retrieval tools can be tested. In this paper we propose a novel benchmark for video retrieval that researchers can use in their studies for comparing features and algorithms. The benchmark comes with frame indexing for objects to assist the process of algorithm development, i.e. research can focus on higher level analysis of matching videos as opposed to spending long periods of research on low level image processing operations such as image segmentation and object definitions.
A method for presenting the result of ground penetrating radar (GPR) Bata processing in the form of structural model of the medium being probed that is obtained on the basis of the quantitative analysis of radar trace...
详细信息
ISBN:
(纸本)1424405130
A method for presenting the result of ground penetrating radar (GPR) Bata processing in the form of structural model of the medium being probed that is obtained on the basis of the quantitative analysis of radar traces showing the reflecting boundaries and electrophysical parameters is considered in this paper. The basis for the method developed are results of studying the features of the reflected signal formations in inhomogeneous media. The method enables to algorithmize the construction process for the structural model of the model of the medium being probed, to decrease labour consumption for radar trace processing and to reduce human factor effect.
The proceedings contain 127 papers. The topics discussed include: encryption models for patient recorddatabases;on the feasibility of a statewide patient recorddatabase;an algorithm for massive loading of raster geo...
ISBN:
(纸本)9806560841
The proceedings contain 127 papers. The topics discussed include: encryption models for patient recorddatabases;on the feasibility of a statewide patient recorddatabase;an algorithm for massive loading of raster geo-spatial data;patient flow optimization model;properties investigation of thin films photovoltaic hetero-structures;ultra-fast optical signalprocessing based on nonlinear optical effects in quantum dot semiconductor optical amplifiers (QD SOAs);visualization of nanoscale processes by image processing methods;nano indentation inspection of the mechanical properties of gold nitride thin films;computerized analysis of flowing conditions for use of chemical sticks in natural gas wells;implementation of a neural network controller on a real time system;type-2 fuzzy set based neuro-fuzzy model for identification and control of nonlinear systems;and intelligent control of blood glucose for diabetes mellitus patient.
The proceedings contain 127 papers. The topics discussed include: encryption models for patient recorddatabases;on the feasibility of a statewide patient recorddatabase;an algorithm for massive loading of raster geo...
ISBN:
(纸本)9806560833
The proceedings contain 127 papers. The topics discussed include: encryption models for patient recorddatabases;on the feasibility of a statewide patient recorddatabase;an algorithm for massive loading of raster geo-spatial data;patient flow optimization model;properties investigation of thin films photovoltaic hetero-structures;ultra-fast optical signalprocessing based on nonlinear optical effects in quantum dot semiconductor optical amplifiers (QD SOAs);visualization of nanoscale processes by image processing methods;nano indentation inspection of the mechanical properties of gold nitride thin films;computerized analysis of flowing conditions for use of chemical sticks in natural gas wells;implementation of a neural network controller on a real time system;type-2 fuzzy set based neuro-fuzzy model for identification and control of nonlinear systems;and intelligent control of blood glucose for diabetes mellitus patient.
All of wavelet functions are designed for a particular form of dynamics, therefore, a choice of one particular function may not be appropriate to capture heart rate variability (HRV) dynamics. The aim of this paper is...
详细信息
ISBN:
(纸本)0863416586
All of wavelet functions are designed for a particular form of dynamics, therefore, a choice of one particular function may not be appropriate to capture heart rate variability (HRV) dynamics. The aim of this paper is to examine a set of wavelet functions (wavelets) for implementation in HRV analysis and to highlight the benefit of this transform relating to today's methods. The basis functions of the wavelet transforms should be able to represent HRV signal feature locally and adapt to slow and fast variations of the signal. This paper discusses the important features of wavelet transform in heart rate variability analysis, including the extent to which the limitations of nonparametric methods like data stationarity and detection of transient episodes can be do away with. The effects of different wavelet functions and their order are assessed and the Daubechies (DW-3) has been proposed as most suitable basis on the basis of performance of various basis and their orders under supine resting and deep-breathing test.
Studies show that patients who have in-hospital cardiac arrests or unexpectedly require admission to an Intensive Care Unit frequently show physiological signs of deterioration prior to the event. This deterioration f...
详细信息
ISBN:
(纸本)0863416586
Studies show that patients who have in-hospital cardiac arrests or unexpectedly require admission to an Intensive Care Unit frequently show physiological signs of deterioration prior to the event. This deterioration frequently goes unnoticed and hence is not acted on. To combat this there has been increased use of mandated vital sign measurement and medical emergency teams (MET) - groups of clinical experts who are called according to criteria relating to changes in physiological parameters. An automated system for detecting patient deterioration through data fusion of heart rate, breathing rate, oxygen saturation, temperature, and blood pressure has been developed. Its performance is tested against current techniques for generating MET calls and early warning of such events is demonstrated.
The watermarking technology provides an ideal tool of copy and copyright protection. It relies on embedding a secret imperceptible signal, a watermark, into the host data in a way that it always remains presence. In t...
详细信息
ISBN:
(纸本)9780889866362
The watermarking technology provides an ideal tool of copy and copyright protection. It relies on embedding a secret imperceptible signal, a watermark, into the host data in a way that it always remains presence. In this paper, we presented a new method of watermarking, which embeds a binary sequence into the DCT domain of a grey-level host image. The embedding procedure is carried out by slightly modifying the correlation coefficients of those middle frequency bands in the DCT blocks. Experiment results show that our algorithm is robust to certain types of attacks, such as the JPEG compression.
Today, the performance of even the best state-of-the-art Automatic Speech Recognition (ASR) tends to deteriorate obviously when speech is transmitted over telephone lines. How to improve ASR robustness in noisy channe...
详细信息
ISBN:
(纸本)0889865477
Today, the performance of even the best state-of-the-art Automatic Speech Recognition (ASR) tends to deteriorate obviously when speech is transmitted over telephone lines. How to improve ASR robustness in noisy channel environments becomes a life and death problem for many real applications. The challenge in addressing such network environments is that they change every moment and show quite different characteristics in terms of signal-to-noise ratio (SNR), stationarity and spectral structure. Previous adaptation methods with complex parameterization could not follow these channel-related variations reliably during the process of a single utterance. So an online adaptation especially designed for noisy channel environments is necessary. In this paper, a prototype library is established to describe acoustic similarities by exploring large amount of channel-contaminated data. The pre-calculated statistics of this library makes it possible to implement a fast channel selection reliably. Furthermore, a Bayesian learning scheme is developed to compensate channel distortion dynamically through a linear interpolation across the library. In our experiments, the new method leads to 10% relative reduction in Word Error Rate (WER) with respect to conventional Maximum Likelihood Linear Regression (MLLR).
暂无评论