• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.4
• /
• pp.57-65
• /
• 1997

The purpose of this paper is to propose the HMM (hidden markov model) based on multi-observation sequence for the isolated word recognition. The proosed model generates the codebook of MSVQ by dividing each word into several sections followed by dividing training data into several sections. Then, we are to obtain the sequential value of multi-observation per each section by weighting the vectors of distance form lower values to higher ones. Thereafter, this the sequential with high probability value while in recognition. 146 DDD area names are selected as the vocabularies for the target recognition, and 10LPC cepstrum coefficients are used as the feature parameters. Besides the speech recognition experiments by way of the proposed model, for the comparison with it, the experiments by DP, MSVQ, and genral HMM are made with the same data under the same condition. The experiment results have shown that HMM based on multi-observation sequence proposed in this paper is proved superior to any other methods such as the ones using DP, MSVQ and general HMM models in recognition rate and time.

• Progress in Medical Physics
• /
• v.18 no.2
• /
• pp.65-72
• /
• 2007

An echo planar imaging (EPI)-based spin-echo sequence Is often used to obtain diffusion tensor imaging (DTI) data on most of the clinical MRI systems, However, this sequence is confounded with the susceptibility artifacts, especially on the temporal lobe in the human brain. Therefore, the objective of this study was to design a pulse sequence that relatively immunizes the susceptibility artifacts, but can map diffusion tensor components in a single-shot mode. A multi-slice multi-echo pulsed-gradient spin-echo (MePGSE) sequence with eight echoes wasdeveloped with selective refocusing pulses for all slices to map the full tensor. The first seven echoes in the train were diffusion-weighted allowing for the observation of diffusion in several different directions in a single experiment and the last echo was for crusher of the residual magnetization. All components of diffusion tensor were measured by a single shot experiment. The sequence was applied in diffusive phantoms. The preliminary experimental verification of the sequence was illustrated by measuring the apparent diffusion coefficient (ADC) for tap water and by measuring diffusion tensor components for watermelon. The ADC values in the series of the water phantom were reliable. The MePGSE sequence, therefore, may be useful in human brain studies.

• The Journal of the Acoustical Society of Korea
• /
• v.12 no.1
• /
• pp.55-60
• /
• 1993

본 논문은 불특정 화자의 단모음 인식에 관한 연구로써, VQ(Vectro Quantization)와 MLP(multi-layer perceptron)에 의한 음성 인식 방법을 제안한다. 이 방법은 VQ codebook을 구하고 이를 이용해서 관측열(observation sequence)을 구해각 codeword가 데이터로부터 가질 수 있는 확률값을 계산하여 이 값을 신경 회로망의 입력으로 사용하는 방법이다. 인식 대상으로는 한국어 단모음을 선정하였으며 10명의 남성 화자가 8개의 단모음을 10번씩 발음한 것으로 시스템의 효율성을 알아보기 위해 VQ/HMM(hidden markov model)에 의한 인식과 비교 실험한다. 실험 결과에 의하면, 시스템의 단순성에도 불구하고 학습능력애 뛰어난 관계로 VQ/HMM보다 VQ와 MLP에 의한 음성 인식률이 향상됨을 보여준다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.1
• /
• pp.30-39
• /
• 1997

In this paper, a new practical coherent detection scheme for biorthogonal signals, which uses multi-symbol observation interval, is proposed and its performances are analyzed and simulated. The technique jointly estimates both the demondulated data and the channel from received signal only while reducing computation complexity by an approximate maximum-likelihood sequence estimation rather than symbol-by-symbol detection as in previous noncoherent detection. The scheme achieves performance close to that of ideal coherent detection with perfect channel estimates when select the appropriate observation symbol interval N in the given symbol alphabet wize M. What is particularly interesting is that the requeired average signal-to-noise ratio per bit ${\gamma}_{b}$ can be reducedd by as much as 1.4dB and the capacity can be increase by as much as 38% when we use this system in the CDMA cellular reverse link.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.6
• /
• pp.90-97
• /
• 2003

This paper proposes a speech recognition based on HMM(Hidden Markov Model) using FSVQ(First Section Vector Quantization) and fuzzy concept. In the proposed paper, we generate codebook of First Section, and then obtain multi-observation sequences by order of large propabilistic values based on fuzzy rule from the codebook of the first section. Thereafter, this observation sequences of first section from codebooks is trained and in case of recognition, a word that has the most highest probability of first section is selected as a recognized word by same concept. Train station names are selected as the target recognition vocabulary and LPC cepstrum coefficients are used as the feature parameters. Besides the speech recognition experiments of proposed method, we experiment the other methods under same conditions and data. Through the experiment results, it is proved that the proposed method based on HMM using FSVQ and fuzzy concept is superior to tile others in recognition rate.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.6
• /
• pp.674-677
• /
• 2004

This paper deals with a systematic approach for the construction of the fixed-delay tree search (FDTS) detector in the intersymbol interference channel. The approach is based on the efficient multi-dimensional space partitioning. The Voronoi diagram (VoD) and the Delaunay tessellation (DT) of the multi-dimensional space are applied to implement the algorithm. In the proposed approach, utilizing the geometric information contained in the VOD/DT, the relative location of the observation sequence is determined which has been shown to reduce the implementation complexity. Detailed construction procedures are discussed followed by an example from the intersymbol interference communication channel.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.6
• /
• pp.5-11
• /
• 1996

This paper is the study for recognizing single vowels of speaker-independent, and we suppose a method of speech recognition using VQ(Vector Quantization)/NN(Neural Network). This method makes a VQ codebook, which is used for obtaining the observation sequence, and then claculates the probability value by comparing each codeword with the data, finally uses these probability values for the input value of the neural network. Korean signle vowels are selected for our recognition experiment, and ten male speakers pronounced eight single vowels ten times. We compare the performance of our method with those of fuzzy VQ/HMM and conventional VQ/NN According to the experiment result, the recognition rate by VQ/NN is 92.3%, by VQ/HMM using fuzzy is 93.8% and by VQ/NN using fuzzy is 95.7%. Therefore, it is shown that recognition rate of speech recognition by fuzzy VQ/NN is better than those of fuzzy VQ/HMM and conventional VQ/HMM because of its excellent learning ability.

• Mycobiology
• /
• v.47 no.3
• /
• pp.319-328
• /
• 2019

Cytospora is a genus including important phytopathogens causing severe dieback and canker diseases distributed worldwide with a wide host range. However, identification of Cytospora species is difficult since the currently available DNA sequence data are insufficient. Aside the limited availability of ex-type sequence data, most of the genetic work is only based on the ITS region DNA marker which lacks the resolution to delineate to the species level in Cytospora. In this study, three fresh strains were isolated from the symptomatic branches of Elaeagnus angustifolia in Xinjiang Uygur Autonomous Region, China. Morphological observation and multi-locus phylogenetic analyses (ITS, LSU, ACT and RPB2) support these specimens are best accommodated as a distinct novel species of Cytospora. Cytospora elaeagnicola sp. nov. is introduced, having discoid, nearly flat, pycnidial conidiomata with hyaline, allantoid conidia, and differs from its relatives genetically and by host association.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.35.2-35.2
• /
• 2021

Blue straggler stars (BSs) are "rejuvenated" main sequence stars first recognized by Allan Sandage from his observation of the prominent northern globular cluster M3 in the year of 1953. BSs are now known to be present in diverse stellar environments including open clusters, globular clusters, dwarf galaxies, and even the field populations of the Milky Way. This makes them a very useful tool in a wide range of astrophysical applications: Particularly BSs are considered to have a crucial role in the evolution of stellar clusters because they affect on the dynamics, the binary population, and the history of the stellar evolution of the cluster they belong to. Here we report a part of the preliminary results from our ongoing research on the BSs in the two metal-poor globular clusters (GCs) in the Large Magellanic Cloud (LMC), Hodge 11 and NGC1466. Using the high precision multi-band images obtained with the Advanced Camera for Survey (ACS) onboard the Hubble Space Telescope (HST), we extract time-series photometry to search for the signal of periodic variations in the luminosity of the BSs. Our preliminary results confirm that several BSs are intrinsic "short period (0.05 < P < 0.25 days)" variable stars with either pulsating or eclipsing types. We will discuss our investigation on the properties of those variable BS candidates in the context of the formation channels of these exotic main sequence stars, and their roles in the dynamical evolution of the host star clusters.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.181-185
• /
• 1998

Recently, research and developmental direction of communication system is concurrent adopting voice data and face image in speaking to provide more higher recognition rate then in the case of only voice data. Therefore, we present a method of lipreading in speech image sequence by using the 3-D facial shape model. The method use a feature information of the face image such as the opening-level of lip, the movement of jaw, and the projection height of lip. At first, we adjust the 3-D face model to speeching face image sequence. Then, to get a feature information we compute variance quantity from adjusted 3-D shape model of image sequence and use the variance quality of the adjusted 3-D model as recognition parameters. We use the intensity inclination values which obtaining from the variance in 3-D feature points as the separation of recognition units from the sequential image. After then, we use discrete HMM algorithm at recognition process, depending on multiple observation sequence which considers the variance of 3-D feature point fully. As a result of recognition experiment with the 8 Korean vowels and 2 Korean consonants, we have about 80% of recognition rate for the plosives and vowels.