Showing papers by "Liqing Zhang published in 2006"

ECG Arrhythmias Recognition System Based on Independent Component Analysis Feature Extraction

Abstract: Automatic recognition of cardiac arrhythmias is important for diagnosis of cardiac abnormalities. This paper presents a new approach to classification ECG signals based on feature extraction to diagnose heartbeat irregularities. We introduce the independent component analysis (ICA) feature extraction method and propose an over-complete feature extraction method combining ICA basis function's coefficients and wavelet transform coefficients. A set of relevant features are selected from the entire overcomplete features using a relevant feature selection method. The selected features are used to trained a support vector machine classifier to recognize different heartbeat arrhythmias. From computer simulations, the proposed method yields a more satisfactory classification results on the MIT-BIH Arrhythmia Database than the other existing methods, reaching an overall accuracy of 98.65%.

A Time-Dependent Model of Information Capacity of Visual Attention

Abstract: What a human's eye tells a human's brain? In this paper, we analyze the information capacity of visual attention. Our hypothesis is that the limit of perceptible spatial frequency is related to observing time. Given more time, one can obtain higher resolution - that is, higher spatial frequency information, of the presented visual stimuli. We designed an experiment to simulate natural viewing conditions, in which time dependent characteristics of the attention can be evoked; and we recorded the temporal responses of 6 subjects. Based on the experiment results, we propose a person-independent model that characterizes the behavior of eyes, relating visual spatial resolution with the duration of attentional concentration time. This model suggests that the information capacity of visual attention is time-dependent.

A two-stage based approach for extracting periodic signals

Abstract: In many applications, such as biomedical engineering, it is often required to obtain specific periodic source signals. In this paper, we propose a two-stage based approach for extracting periodic signals. At the first stage, the autocorrelation property of the desired source signal is exploited to roughly extract the desired source signal. At the second stage, the extracted signal is further processed as cleanly as possible, based on the higher-order statistics. Simulations on artificially generated data and real-world ECG data have showed its better performance, compared with many existing extraction algorithms.

Multichannel blind deconvolution using a novel filter decomposition method

Abstract: In our previous work [11], we introduced a filter decomposition method for blind deconvolution in non-minimum phase system To simplify the deconvolution procedure, we further study the demixing filter and modify the cascade structure of demixing filter In this paper, we introduce a novel two-stage algorithm for blind deconvolution In first stage, we present a permutable cascade structure which constructed by a causal filter and an anti-causal scalar filter Then, we develop SOS-based algorithm for causal filter and derive a natural gradient algorithm for anti-causal scalar filter At second stage, we apply an instantaneous ICA algorithm to eliminate the residual instantaneous mixtures Computer simulations show the validity and effectiveness of this approach

Two-Stage temporally correlated source extraction algorithm with its application in extraction of event-related potentials

Abstract: To extract source signals with certain temporal structures, such as periodicity, we propose a two-stage extraction algorithm. Its first stage uses the autocorrelation property of the desired source signal, and the second stage exploits the independence assumption. The algorithm is suitable to extract periodic or quasi-periodic source signals, without requiring that they have distinct periods. It outperforms many existing algorithms in many aspects, confirmed by simulations. Finally, we use the proposed algorithm to extract the components of visual event-related potentials evoked by three geometrical figure stimuli, and the classification accuracy based on the extracted components achieves 93.2%.

Multichannel Blind Deconvolution Using a Novel Filter Decomposition Method

Abstract: In our previous work [II], we introduced a filter decomposition method for blind deconvolution in non-minimum phase system. To simplify the deconvolution procedure, we further study the demixing filter and modify the cascade structure of demixing filter. In this paper, we introduce a novel two-stage algorithm for blind deconvolution. In first stage, we present a permutable cascade structure which constructed by a causal filter and an anti-causal scalar filter. Then, we develop SOS-based algorithm for causal filter and derive a natural gradient algorithm for anti-causal scalar filter. At second stage, we apply an instantaneous ICA algorithm to eliminate the residual instantaneous mixtures. Computer simulations show the validity and effectiveness of this approach.

A time-dependent model of information capacity of visual attention

Abstract: What a human's eye tells a human's brain? In this paper, we analyze the information capacity of visual attention. Our hypothesis is that the limit of perceptible spatial frequency is related to observing time. Given more time, one can obtain higher resolution – that is, higher spatial frequency information, of the presented visual stimuli. We designed an experiment to simulate natural viewing conditions, in which time dependent characteristics of the attention can be evoked; and we recorded the temporal responses of 6 subjects. Based on the experiment results, we propose a person-independent model that characterizes the behavior of eyes, relating visual spatial resolution with the duration of attentional concentration time. This model suggests that the information capacity of visual attention is time-dependent.

Extraction and Classification of Visual Evoked Potentials Based on a Two-Stage Source Extraction Algorithm

Abstract: In order to verify whether or not the EEG patterns can be classified when the subjects perceive different types of geometric figures, we perform some EEG experiments. In this paper, the evoked potentials by three types of geometric figures are extracted and classified using a series of approaches. First, a two-stage source extraction algorithm is proposed to extract the evoked potentials from the recorded EEG signals, and then a mutual information based feature selection method is presented to find effective features for classification. Finally, a multi-category support vector machine classifier is employed, which achieves the average classification performance of 93.2%.

Two-Stage Blind Deconvolution for V-BLAST OFDM System

Abstract: In this paper, we integrate orthogonal frequency-division multiplexing (OFDM) technique with vertical Bell Labs layered space-time (V-BLAST) architecture as a promising solution for enhancing the data rates of wireless communication systems, and propose a new blind deconvolution method. A two-stage algorithm is developed to estimate the channel parameters. At first stage, we propose an algorithm based on the second order statistics to decorrelate the sensor signals. After decorrelation, we apply instantaneous demixing algorithm to separate the signals at the second stage. Simulation results demonstrate the validity and the performance of the proposed algorithms.

Local independent factorization of natural scenes

Abstract: In this paper, we study sparse representation of large-size natural scenes via local spatial dependency decomposition. We propose a local independent factorization model of natural scenes and develop a learning algorithm for adaptation of the synaptic weights. We investigate the dependency of neighboring location of the natural scene patches and derive learning algorithm to train the visual neural network. Some numerical experiments on natural scenes are performed to show the sparse representation of the visual sensory information.

Multi-Modal Machine Attention: Sound Localization and Visual-Auditory Signal Synchronization

Abstract: This paper discusses machine attention based on multimodal information integration and synchronization. The aim is to introduce a computational model for robust multimodal attention mechanism in artificial systems. First, we introduce a new blind source separation algorithm based on temporal structures and high order statistics of the source signals. Combining the bandpass filter and blind source separation, we present a new sound localization approach to find the sound position in a complicated environment. Furthermore, we introduce a visual-auditory signal synchronization approach to focus visual attention in the artificial system to the interesting object that synchronizes the auditory and visual signals.

Two-Stage blind deconvolution for V-BLAST OFDM system

Abstract: In this paper, we integrate orthogonal frequency-division multiplexing (OFDM) technique with vertical Bell Labs layered space-time (V-BLAST) architecture as a promising solution for enhancing the data rates of wireless communication systems, and propose a new blind deconvolution method. A two-stage algorithm is developed to estimate the channel parameters. At first stage, we propose an algorithm based on the second order statistics to decorrelate the sensor signals. After decorrelation, we apply instantaneous demixing algorithm to separate the signals at the second stage. Simulation results demonstrate the validity and the performance of the proposed algorithms.