• Phonetics and Speech Sciences
• /
• v.6 no.4
• /
• pp.127-132
• /
• 2014

This paper describes a robust speech recognition technique by reconstructing spectral components mismatched with a training environment. Although the cluster-based reconstruction method can compensate the unreliable components from reliable components in the same spectral vector by assuming an independent, identically distributed Gaussian-mixture process of training spectral vectors, the presented method exploits the temporal dependency of speech to reconstruct the components by introducing a hidden-Markov-model prior which incorporates an internal state transition plausible for an observed spectral vector sequence. The experimental results indicate that the described method can provide temporally consistent reconstruction and further improve recognition performance on average compared to the conventional method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.5095-5111
• /
• 2016

The Probability Hypothesis Density (PHD) filter is a suboptimal approximation and tractable alternative to the multi-target Bayesian filter based on random finite sets. However, the PHD filter fails to track newborn targets when the target birth intensity is unknown prior to tracking. In this paper, a dual detection-guided newborn target intensity PHD algorithm is developed to solve the problem, where two schemes, namely, a newborn target intensity estimation scheme and improved measurement-driven scheme, are proposed. First, the newborn target intensity estimation scheme, consisting of the Dirichlet distribution with the negative exponent parameter and target velocity feature, is used to recursively estimate the target birth intensity. Then, an improved measurement-driven scheme is introduced to reduce the errors of the estimated number of targets and computational load. Simulation results demonstrate that the proposed algorithm can achieve good performance in terms of target states, target number and computational load when the newborn target intensity is not predefined in multi-target tracking systems.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.5 s.311
• /
• pp.52-61
• /
• 2006

Orthogonal wavelet tansform which is generally used in image and signal processing applications has limited performance because of lack of shift invariance and low directional selectivity. To overcome these demerits complex wavelet transform has been proposed. In this paper, we present an efficient image denoising method using dual-tree complex wavelet transform and Bernoulli-Gauss prior model. In estimating hyper-parameters for Bernoulli-Gaussian model, we present two simple and non-iterative methods. We use hypothesis-testing technique in order to estimate the mixing parameter, Bernoulli random variable. Based on the estimated mixing parameter, variance for clean signal is obtained by using maximum generalized marginal likelihood (MGML) estimator. We simulate our denoising method using dual-tree complex wavelet and compare our algorithm to well blown denoising schemes. Experimental results show that the proposed method can generate good denoising results for high frequency image with low computational cost.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.162-170
• /
• 2013

In recent years, the interests and needs of the Pedestrian Protection System (PPS), which is mounted on the vehicle for the purpose of traffic safety improvement is increasing. In this paper, we propose a pedestrian candidate window extraction and unit cell histogram based HOG descriptor calculation methods. At pedestrian detection candidate windows extraction stage, the bright ratio of pedestrian and its circumference region, vertical edge projection, edge factor, and PCA reconstruction image are used. Dalal's HOG requires pixel based histogram calculation by Gaussian weights and trilinear interpolation on overlapping blocks, But our method performs Gaussian down-weight and computes histogram on a per-cell basis, and then the histogram is combined with the adjacent cell, so our method can be calculated faster than Dalal's method. Our PCA reconstruction error based pedestrian detection candidate window extraction method efficiently classifies background based on the difference between pedestrian's head and shoulder area. The proposed method improves detection speed compared to the conventional HOG just using image without any prior information from camera calibration or depth map obtained from stereo cameras.

• Journal of Internet Computing and Services
• /
• v.8 no.4
• /
• pp.1-9
• /
• 2007

Decision fusion in wireless sensor networks under non-Gaussian noise channels is studied. To consider the tail behavior noise distributions, we use a exponentially-tailed distribution as a wide class of noise distributions. Based on a canonical parallel fusion model with fading and noise channels, the likelihood ratio(LR) based fusion rule is considered as an optimal fusion rule under Neyman-Pearson criterion. With both high and low signal-to-noise ratio (SNR) approximation to the optimal rule, we obtain several suboptimal fusion rules. and we propose a simple fusion rule that provides robust detection performance with a minimum prior information, Performance evaluation for several fusion rules is peformed through simulation. Simulation results show the robustness of the Proposed simple fusion rule.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2
• /
• pp.10-15
• /
• 2001

In this paper, we describe the comparison between the combination of features using a speech and music discrimination, which is classifying between speech and music on audio signals. Audio signals are classified into 3classes (speech, music, speech and music) and 2classes (speech, music). Experiments carried out on three types of feature, Mel-cepstrum, energy, zero-crossings, and try to find a best combination between features to speech and music discrimination. We using a Gaussian Mixture Model (GMM) for discrimination algorithm and combine different features into a single vector prior to modeling the data with a GMM. In 3classes, the best result is achieved using Mel-cepstrum, energy and zero-crossings in a single feature vector (speech: 95.1％, music: 61.9％, speech & music: 55.5％). In 2classes, the best result is achieved using Mel-cepstrum, energy and Mel-cepstrum, energy, zero-crossings in a single feature vector (speech: 98.9％, music: 100％).

• The Korean Journal of Applied Statistics
• /
• v.28 no.2
• /
• pp.251-267
• /
• 2015

The Indian Buffet Process is a stochastic process on equivalence classes of binary matrices having finite rows and infinite columns. The Indian Buffet Process can be imposed as the prior distribution on the binary matrix in an infinite feature model. We describe the derivation of the Indian buffet process from a finite feature model, and briefly explain the relation between the Indian buffet process and the beta process. Using a Gaussian linear model, we describe three algorithms: Gibbs sampling algorithm, Stick-breaking algorithm and variational method, with application for finding features in image data. We also illustrate the use of the Indian Buffet Process in various type of analysis such as dyadic data analysis, network data analysis and independent component analysis.

• Journal of Korea Multimedia Society
• /
• v.12 no.8
• /
• pp.1155-1163
• /
• 2009

This paper proposes an automatic facial expression generation system of vector graphic character using gaussian process model. Proposed method extracts the main feature vectors from twenty-six facial data of character redefined based on Russell's internal emotion state. Also by using new gaussian process model, SGPLVM, we find low-dimensional feature data from extracted high-dimensional feature vectors, and learn probability distribution function (PDF). All parameters of PDF are estimated by maximization the likelihood of learned expression data, and these are used to select wanted facial expressions on two-dimensional space in real time. As a result of simulation, we confirm that proposed facial expression generation tool is working in the small facial expression datasets and can generate various facial expressions without prior knowledge about relation between facial expression and emotion.

• Journal of The Korean Astronomical Society
• /
• v.55 no.5
• /
• pp.149-172
• /
• 2022

We present a new method which constructs an Hɪ super-profile of a galaxy which is based on profile decomposition analysis. The decomposed velocity profiles of an Hɪ data cube with an optimal number of Gaussian components are co-added after being aligned in velocity with respect to their centroid velocities. This is compared to the previous approach where no prior profile decomposition is made for the velocity profiles being stacked. The S/N improved super-profile is useful for deriving the galaxy's global Hɪ properties like velocity dispersion and mass from observations which do not provide sufficient surface brightness sensitivity for the galaxy. As a practical test, we apply our new method to 64 high-resolution Hɪ data cubes of nearby galaxies in the local Universe which are taken from THINGS and LITTLE THINGS. In addition, we also construct two additional Hɪ super-profiles of the sample galaxies using symmetric and all velocity profiles of the cubes whose centroid velocities are determined from Hermite h3 polynomial fitting, respectively. We find that the Hɪ super-profiles constructed using the new method have narrower cores and broader wings in shape than the other two super-profiles. This is mainly due to the effect of either asymmetric velocity profiles' central velocity bias or the removal of asymmetric velocity profiles in the previous methods on the resulting Hɪ super-profiles. We discuss how the shapes (𝜎_n/𝜎_b, A_n/A_b, and A_n/A_tot) of the new Hɪ super-profiles which are measured from a double Gaussian fit are correlated with star formation rates of the sample galaxies and are compared with those of the other two super-profiles.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.655-668
• /
• 2023

In radiography, an antiscatter grid is a well-known device for eliminating unexpected x-ray scatter. We investigate a new stationary grid artifact suppression method based on a nonsubsampled contourlet transform (NSCT) incorporated with Gaussian band-pass filtering. The proposed method has an advantage that extracts the Moiré components while minimizing the loss of image information and apply the prior information of Moiré component positions in multi-decomposition sub-band images. We implemented the proposed algorithm and performed a simulation and an experiment to demonstrate its viability. We did this experiment using an x-ray tube (M-113T, Varian, focal spot size: 0.1 mm), a flat-panel detector (ROSE-M Sensor, Aspenstate, pixel dimension: 3032 × 3800 pixels, pixel size: 0.076 mm), and carbon graphite-interspaced grids (JPI Healthcare, 18 cm × 24 cm, line density: 103 LP/inch and 150 LP/inch, ratio: 5:1, focal distance: 65 cm). Our results indicate that the proposed method successfully suppressed grid artifacts by reducing them without either reducing the spatial resolution or causing negative side effects. Consequently, we anticipate that the proposed method can improve image acquisition in a stationary grid x-ray system as well as in extended x-ray imaging.