• The Transactions of the Korea Information Processing Society
• /
• v.7 no.7
• /
• pp.2246-2253
• /
• 2000

In this paper, we show that one can enhance the performance of blind segmentation of phoneme boundaries by adopting the knowledge of Korean syllabic structure and the regions of voiced/unvoiced sounds. eh proposed method consists of three processes : the process to extract candidate phoneme boundaries, the process to detect boundaries of voiced/unvoiced sounds, and the process to select final phoneme boundaries. The candidate phoneme boudaries are extracted by clustering method based on similarity between two adjacent clusters. The employed similarity measure in this a process is the ratio of the probability density of adjacent clusters. To detect he boundaries of voiced/unvoiced sounds, we first compute the power density spectrum of speech signal in 0∼400 Hz frequency band. Then the points where this paper density spectrum variation is greater than the threshold are chosen as the boundaries of voiced/unvoiced sounds. The final phoneme boundaries consist of all the candidate phoneme boundaries in voiced region and limited number of candidate phoneme boundaries in unvoiced region. The experimental result showed about 40% decrease of insertion rate compared to the blind segmentation method we adopted.

• Journal of the Institute of Convergence Signal Processing
• /
• v.9 no.4
• /
• pp.329-336
• /
• 2008

The facilities with new technologies in the recent power plants become larger and need a lot of high cost for maintenance due to stop operations and accidents from the operating machines. Therefore, it claims new systems to monitor the operating status and predict the faults of the machines. This research classifies the normal/abnormal status of the machines into 5 levels which are normal-level/abnormal-level/care-level/dangerous-level/fault and develops the new system that predicts faults without stop operation in power plants. We propose the regional segmentation technique in the frequency domain from the data of the operating machines and generate the Peak-codes similar to the Bar-codes, The high efficient and economic operations of the power plants will be achieved by carrying out the pre-maintenance at the care level of 5 levels in the plants. In order to be utilized easily at power plants, we developed the algorithm appling to a notebook computer from signal acquisition to the discrimination.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.774-783
• /
• 2020

For the issue of subsurface target localization by reverse projection, a new approach of target localization with different distances based on symmetric sub-array multiple-input multiple-output (MIMO) radar is proposed in this paper. By utilizing the particularity of structure of the two symmetric sub-arrays, the received signals are jointly reconstructed to eliminate the distance information from the steering vectors. The distance-independent direction of arrival (DOA) estimates are acquired, and the localizations of subsurface targets with different distances are realized by reverse projection. According to the localization mechanism and application characteristics of the proposed algorithm, the grid zooming method based on spatial segmentation is used to optimize the locaiton efficiency. Simulation results demonstrate the effectiveness of the proposed localization method and optimization scheme.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1918-1919
• /
• 2007

In this study, dynamic time warping(DTW) is utilized especially for automatically segmenting ECG(Electrocardiogram) signal to extract a periodic time information. For the possible medical application for diagnosing the abnormalities of ECG, the relative metric distance of the warped ECG signals are computed to decide whether the abrupt variations of ECG signal occur or not.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.801-804
• /
• 2001

This article is concerned with automatic segmentation of two adjacent vowels for speech signals. All kinds of transition case of adjacent vowels can be characterized by spectrogram. Firstly the voiced-speech is extracted by the histogram analysis of vowel indicator which consists of wavelet low pass components. Secondly given phonetic transcription and transition pattern spectrogram, the voiced-speech portion which has consecutive vowels automatically segmented by the template matching. The cross-correlation function is adapted as a template matching method and the modified correlation coefficient is calculated for all frames. The largest value on the modified correlation coefficient series indicates the boundary of two consecutive vowel sounds. The experiment is performed for 154 vowel transition sets. The 154 spectrogram templates are gathered from 154 words(PRW Speech DB) and the 161 test words(PBW Speech DB) which are uttered by 5 speakers were tested. The experimental result shows the validity of the method.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.110-112
• /
• 2002

In this paper, a new technique for Thai isolated speech phoneme segmentation is proposed. Based on Thai speech feature, the isolated speech is first divided into low and high frequency components by using the technique of wavelet decomposition. Then the energy contour of each decomposed signal is computed and employed to locate phoneme boundary. To verity the proposed scheme, some experiments have been performed using 1,000 syllables data recorded from 10 speakers. The accuracy rates are 96.0, 89.9, 92.7 and 98.9% for initial consonant, vowel, final consonant and silence, respectively.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.11
• /
• pp.874-882
• /
• 2003

The condition assessment of engineering systems has increased in importance because the manpower needed to operate and supervise various plants has been reduced. Especially, induction motors are at the core of most engineering processes, and there is an indispensable need to monitor their health and performance. So detection and diagnosis of motor faults is a base to improve efficiency of the industrial plant. In this paper, a model-based fault detection system is developed for induction motors, using steady state vibration signals. Early various fault detection systems using vibration signals are a trivial method and those methods are prone to have missed fault or false alarms. The suggested motor fault detection system was developed using a model-based reference value. The stationary signal had been extracted from the non-stationary signal using a data segmentation method. The signal processing method applied in this research is FFT. A reference model with spectra signal is developed and then the residuals of the vibration signal are generated. The ratio of RMS values of vibration residuals is proposed as a fault indicator for detecting faults. The developed fault detection system is tested on 800 hp motor and it is shown to be effective for detecting faults in the air-gap eccentricities and broken rotor bars. The suggested system is shown to be effective for reducing missed faults and false alarms. Moreover, the suggested system has advantages in the automation of fault detection algorithms in a random signal system, and the reference model is not complicated.

• Phonetics and Speech Sciences
• /
• v.10 no.3
• /
• pp.19-29
• /
• 2018

We propose to use independent component analysis (ICA) and deep neural network (DNN) to detect music sections in broadcast drama contents. Drama contents mainly comprise silence, noise, speech, music, and mixed (speech+music) sections. The silence section is detected by signal activity detection. To detect the music section, we train noise, speech, music, and mixed models with DNN. In computer experiments, we used the MUSAN corpus for training the acoustic model, and conducted an experiment using 3 hours' worth of Korean drama contents. As the mixed section includes music signals, it was regarded as a music section. The segmentation error rate (SER) of music section detection was observed to be 19.0%. In addition, when stereo mixed signals were separated into music signals using ICA, the SER was reduced to 11.8%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.474-476
• /
• 2021

Glioblastoma is the most common brain malignancies arising from glial cells. Early diagnosis and treatment plan establishment are important, and cancer is diagnosed mainly through T1CE imaging through injection of a contrast agent. However, the risk of injection of gadolinium-based contrast agents is increasing recently. Region segmentation that marks cancer regions in medical images plays a key role in CAD systems, and deep neural network models for synthesizing new images are also being studied. In this study, we propose a model that simultaneously learns the generation of T1CE images and segmentation of cancer regions. The performance of the proposed model is evaluated using similarity measurements including mean square error and peak signal-to-noise ratio, and shows average result values of 21 and 39 dB.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.414-423
• /
• 2020

In this paper, we propose a Dilated Convolution Gate Linear Unit (DCGLU) to mitigate the lack of sparsity and small receptive field problems caused by the segmentation map extraction process in sound event detection with weak labels. In the advent of deep learning framework, segmentation map extraction approaches have shown improved performance in noisy environments. However, these methods are forced to maintain the size of the feature map to extract the segmentation map as the model would be constructed without a pooling operation. As a result, the performance of these methods is deteriorated with a lack of sparsity and a small receptive field. To mitigate these problems, we utilize GLU to control the flow of information and Dilated Convolutional Neural Networks (DCNNs) to increase the receptive field without additional learning parameters. For the performance evaluation, we employ a URBAN-SED and self-organized bird sound dataset. The relevant experiments show that our proposed DCGLU model outperforms over other baselines. In particular, our method is shown to exhibit robustness against nature sound noises with three Signal to Noise Ratio (SNR) levels (20 dB, 10 dB and 0 dB).