• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.15 no.3
• /
• pp.145-152
• /
• 2015

We consider an online selective-sample learning problem for sequence classification, where the goal is to learn a predictive model using a stream of data samples whose class labels can be selectively queried by the algorithm. Given that there is a limit to the total number of queries permitted, the key issue is choosing the most informative and salient samples for their class labels to be queried. Recently, several aggressive selective-sample algorithms have been proposed under a linear model for static (non-sequential) binary classification. We extend the idea to hidden Markov models for multi-class sequence classification by introducing reasonable measures for the novelty and prediction confidence of the incoming sample with respect to the current model, on which the query decision is based. For several sequence classification datasets/tasks in online learning setups, we demonstrate the effectiveness of the proposed approach.

• ETRI Journal
• /
• v.46 no.2
• /
• pp.250-262
• /
• 2024

Air-writing recognition is relevant in areas such as natural human-computer interaction, augmented reality, and virtual reality. A trajectory is the most natural way to represent air writing. We analyze the recognition accuracy of words written in air considering five features, namely, writing direction, curvature, trajectory, orthocenter, and ellipsoid, as well as different parameters of a hidden Markov model classifier. Experiments were performed on two representative datasets, whose sample trajectories were collected using a Leap Motion Controller from a fingertip performing air writing. Dataset D₁ contains 840 English words from 21 classes, and dataset D₂ contains 1600 English words from 40 classes. A genetic algorithm was combined with a hidden Markov model classifier to obtain the best subset of features. Combination ftrajectory, orthocenter, writing direction, curvatureg provided the best feature set, achieving recognition accuracies on datasets D₁ and D₂ of 98.81% and 83.58%, respectively.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.1
• /
• pp.145-151
• /
• 2014

This paper proposes an image haze removal algorithm for a single image. The conventional Dark Channel Prior(DCP) algorithm estimates a transmission map using the dark information in an image, and the haze regions are then detected using a matting algorithm. However, since the DCP algorithm uses block-based processing, block artifacts are invariably formed in the transmission map. To solve this problem, the proposed algorithm generates a modified transmission map using a Hidden Markov Random Field(HMRF) and Expectation-Maximization(EM) algorithm. Experimental results confirm that the proposed algorithm is superior to conventional algorithms in image haze removal.

• Journal of Korea Multimedia Society
• /
• v.7 no.11
• /
• pp.1540-1549
• /
• 2004

In this paper, we propose an algorithm to detect semantic events in golf video using Hidden Markov Model. The purpose of this paper is to identify and classify the golf events to facilitate highlight-based video indexing and summarization. In this paper we first define 4 semantic events, and then design HMM model with states made up of each event. We also use 10 multiple visual features based on MPEG-7 visual descriptors to acquire parameters of HMM for each event. Experimental results showed that the proposed algorithm provided reasonable detection performance for identifying a variety of golf events.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.375-378
• /
• 2000

In this paper, we propose a novo] content-based image retrieval system using both Hidden Markov Model(HMM) and an improved chain code. The Gaussian Mixture Model(GMM) is applied to statistically model a color information of the image, and Deterministic Annealing EM(DAEM) algorithm is employed to estimate the parameters of GMM. This result is used to segment the given image. We use an improved chain code, which is invariant to rotation, translation and scale, to extract the feature vectors of the shape for each image in the database. These are stored together in the database with each HMM whose parameters (A, B, $\pi$) are estimated by Baum-Welch algorithm. With respect to feature vector obtained in the same way from the query image, a occurring probability of each image is computed by using the forward algorithm of HMM. We use these probabilities for the image retrieval and present the highest similarity images based on these probabilities.

• The Korean Journal of Applied Statistics
• /
• v.24 no.1
• /
• pp.177-183
• /
• 2011

Hidden Markov random eld(HMRF) is one of the most common model for image segmentation which is an important preprocessing in many imaging devices. The HMRF has unknown hyper-parameters on Markov random field to be estimated in segmenting testing images. However, in practice, due to computational complexity, it is often assumed to be a fixed constant. In this paper, we numerically show that the segmentation results very depending on the fixed hyper-parameter, and, if the parameter is misspecified, they further depend on the choice of the class-labelling algorithm. In contrast, the HMRF with estimated hyper-parameter provides consistent segmentation results regardless of the choice of class labelling and the estimation method. Thus, we recommend practitioners estimate the hyper-parameter even though it is computationally complex.

• Journal of Korea Multimedia Society
• /
• v.19 no.1
• /
• pp.22-29
• /
• 2016

Hidden Markov Model is one of the most successful and popular tools for modeling real world sequential data. Real world signals come in a variety of shapes and variabilities, among which temporal and spectral ones are the prime targets that the HMM aims at. A new problem that is gaining increasing attention is characterizing missing observations in incomplete data sequences. They are incomplete in that there are holes or omitted measurements. The standard HMM algorithms have been developed for complete data with a measurements at each regular point in time. This paper presents a modified algorithm for a discrete HMM that allows substantial amount of omissions in the input sequence. Basically it is a variant of Baum-Welch which explicitly considers the case of isolated or a number of omissions in succession. The algorithm has been tested on online handwriting samples expressed in direction codes. An extensive set of experiments show that the HMM so modeled are highly flexible showing a consistent and robust performance regardless of the amount of omissions.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.11-19
• /
• 1994

The matching probability P(ο/$\lambda$), of the signal sequence(ο) observed for a finite time interval with a HMM (Hidden Markov Model $\lambda$) indicates the probability that signal comes from the given model. By utilizing the fact that the probability represents matching score of the observed signal with the model we can recognize an unknown signal pattern by comparing the magnitudes of the matching probabilities with respect to the known models. Because the algorithm however uses floating point variables during the computing process hardware implementation of the algorithm requires floating point units. This paper proposes an integer algorithm which uses positive integer numbers rather than float point ones to compute the matching probability so that we can economically realize the algorithm into hardware. The algorithm makes the model parameters integer numbers by multiplying positive constants and prevents from divergence of data through the normalization of variables at each step. The final equation of matching probability is composed of constant terms and a variable term which contains logarithm operations. A scheme to make the log conversion table smaller is also presented. To analyze the qualitive characteristics of the proposed algorithm we attatch simulation result performed on two groups of 10 hypothetic models respectively and inspect the statistical properties with repect to the model order the magnitude of scaling constants and the effect of the observation length.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.6
• /
• pp.1852-1860
• /
• 2000

Distinct from the Markov random field or pseudo 2D HMM models for image analysis, this paper proposes a new model of 2D hidden Markov mesh chain(HMMM) model which subsumes the definitions of and the assumptions underlying the conventional HMM. The proposed model is a new theoretical realization of 2D HMM with the causality of top-down and left-right progression and the complete lattice constraint. These two conditions enable an efficient mesh decoding for model estimation and a recursive maximum likelihood estimation of model parameters. Those algorithms are developed in theoretical perspective and, in particular, the training algorithm, it is proved, attains the optimal set of parameters.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.24-31
• /
• 1994

In this paper, we report on the development of a speaker independent continuous speech recognition system using continuous hidden Markov models. The continuous hidden Markov model consists of mean and covariance matrices and directly models speech signal parameters, therefore does not have quantization error. Filter bank coefficients with their 1st and 2nd-order derivatives are used as feature vectors to represent the dynamic features of speech signal. We use the segmental K-means algorithm as a training algorithm and triphone as a recognition unit to alleviate performance degradation due to coarticulation problems critical in continuous speech recognition. Also, we use the one-pass search algorithm that Is advantageous in speeding-up the recognition time. Experimental results show that the system attains the recognition accuracy of $83\%$ without grammar and $94\%$ with finite state networks in speaker-indepdent speech recognition.