• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1330-1332
• /
• 2015

최근 지능형 감시 시스템에서 비정상 행동들을 자동으로 감지하는 연구가 활발히 진행되고 있다. 그러나 해결하기 힘든 몇 가지 이슈들이 있는데, 주어진 입력 영상에서 군중들이 중첩될 때 각각의 객체를 인식하는데 어려움이 있다는 점과 비정상 행동을 나타내는 훈련 데이터셋이 제한적이라는 점이다. 이러한 문제들을 해결하기 위해 우리는 군중 영상에서 비정상 행동들을 인식하는 새로운 프레임워크를 제안한다. 제안된 방법은 크게 특징추출모듈과 추출된 특징들을 이용한 행동인식모듈로 구성된다. 중첩문제를 해결하기 위해 움직임 에너지와 고정 에너지를 특성으로 정의하였고 위에 언급한 특징추출모듈에서 두 에너지 값을 계산한다. 그리고 정상/비정상 행동들은 HMM과 최적의 임계값을 도출하는 알고리즘을 사용하는 행동인식모듈에 의해 분류된다. 우리가 제안한 방법은 인공 데이터셋과 실제 비디오 영상 데이터셋을 이용한 실험에 의해 증명한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.10
• /
• pp.2137-2142
• /
• 2012

Many studies in detection and classification of the targets in the underwater environments have been conducted for military purposes, as well as for non-military purpose. Due to the complicated characteristics of underwater acoustic signal reflecting multipath environments and spatio-temporal varying characteristics, active sonar target classification technique has been considered as a difficult technique. And it has a difficult in collecting actual underwater data. In this paper, we implemented the simulator to synthesize the active target signal, to extract feature and to classify the target in the underwater environment. In target signal synthesis, highlight and three-dimensional model are used and multi-aspect based hidden markov model is used for target classification.

• Journal of Korea Water Resources Association
• /
• v.52 no.8
• /
• pp.545-554
• /
• 2019

The most drought assessments are based on a drought index, which depends on univariate variables such as precipitation and soil moisture. However, there is a limitation in representing the drought conditions with single variables due to their complexity. It has been acknowledged that a multivariate drought index can more effectively describe the complex drought state. In this context, this study propose a Copula-based drought index that can jointly consider precipitation and soil moisture. Unlike precipitation data, long-term soil moisture data is not readily available so that this study utilized a Gaussian Mixture Non-Homogeneous Hidden Markov chain Model (GM-NHMM) model to simulate the soil moisture using the observed precipitation and temperature ranging from 1973 to 2014. The GM-NHMM model showed a better performance in terms of reproducing key statistics of soil moisture, compared to a multiple regression model. Finally, a bivariate frequency analysis was performed for the drought duration and severity, and it was confirmed that the recent droughts over Jeollabuk-do in 2015 have a 20-year return period.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.64-68
• /
• 2020

In a conventional speech recognition system using Gaussian Mixture Model-Hidden Markov Model (GMM-HMM), the cepstral feature normalization method based on pole filtering was effective in improving the performance of recognition of short utterances in noisy environments. In this paper, the usefulness of this method for the state-of-the-art speech recognition system using Deep Neural Network (DNN) is examined. Experimental results on AURORA 2 DB show that the cepstral mean and variance normalization based on pole filtering improves the recognition performance of very short utterances compared to that without pole filtering, especially when there is a large mismatch between the training and test conditions.

• Journal of Internet Computing and Services
• /
• v.10 no.4
• /
• pp.55-70
• /
• 2009

Vision-based Human computer interaction is an emerging field of science and industry to provide natural way to communicate with human and computer. In that sense, inferring the emotional state of the person based on the facial expression recognition is an important issue. In this paper, we present a novel approach to recognize facial expression from a sequence of input images using emotional specific HMM (Hidden Markov Model) and facial motion tracking based on optical flow. Conventionally, in the HMM which consists of basic emotional states, it is considered natural that transitions between emotions are imposed to pass through neutral state. However, in this work we propose an enhanced transition framework model which consists of transitions between each emotional state without passing through neutral state in addition to a traditional transition model. For the localization of facial features from video sequence we exploit template matching and optical flow. The facial feature displacements traced by the optical flow are used for input parameters to HMM for facial expression recognition. From the experiment, we can prove that the proposed framework can effectively recognize the facial expression in real time.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.656-664
• /
• 2014

In the case of a haze image, transferring the information of the original image is difficult as the contrast leans toward bright regions. Thus, dehazing algorithms have become an important area of study. Normally, since it is hard to obtain a haze-free image, the output image is qualitatively analyzed to test the performance of an algorithm. However, this paper proposes a quantitative error comparison based on reproducing the haze image using a disparity map. In addition, a Hidden Random Markov Model and EM algorithm are used to remove any block artifacts. The performance of the proposed algorithm is confirmed using a variety of synthetic and natural images.

• 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 KIISE:Software and Applications
• /
• v.36 no.9
• /
• pp.683-690
• /
• 2009

One of the major problems that prevent us from building a good recognition system for large-scale on-line Chinese character recognition using HMMs is increasing recognition time. In this paper, we propose a clustering method to solve recognition speed problem and an efficient distance measure between HMMs. From the experiments, we got about twice the recognition speed and 95.37% 10-candidate recognition accuracy, which is only 0.9% decrease, for 20,902 Chinese characters defined in Unicode CJK unified ideographs.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.1
• /
• pp.46-51
• /
• 2007

This paper is to present gesture analysis for human-robot interaction. Understanding human emotions through gesture is one of the necessary skills fo the computers to interact intelligently with their human counterparts. Gesture analysis is consisted of several processes such as detecting of hand, extracting feature, and recognizing emotions. For efficient operation we used recognizing a gesture with HMM(Hidden Markov Model). We constructed a large gesture database, with which we verified our method. As a result, our method is successfully included and operated in a mobile system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.12
• /
• pp.2622-2627
• /
• 2010

The accuracy of face recognition used unmanned security system is very important and necessary. However, face recognition is a lot of restriction due to the change of distortion of face image, illumination, face size, face expression, round image. We propose a hybrid neural network for improve the performance of the face recognition. The proposed method is consisted of SOM and LVQ. In order to verify usefulness of the proposed method, we make a comparison between eigenface method, hidden Markov model method, multi-layer neural network.