• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.375-382
• /
• 2003

Lip-reading technique based on bimodal is to enhance speech recognition rate in noisy environment. It is most important to detect the correct lip-image. But it is hard to estimate stable performance in dynamic environment, because of many factors to deteriorate Lip-reading's performance. There are illumination change, speaker's pronunciation habit, versatility of lips shape and rotation or size change of lips etc. In this paper, we propose the IIR filtering in time-domain for the stable performance. It is very proper to remove the noise of speech, to enhance performance of recognition by digital filtering in time domain. While the lip-reading technique in whole lip image makes data massive, the Principal Component Analysis of pre-process allows to reduce the data quantify by detection of feature without loss of image information. For the observation performance of speech recognition using only image information, we made an experiment on recognition after choosing 22 words in available car service. We used Hidden Markov Model by speech recognition algorithm to compare this words' recognition performance. As a result, while the recognition rate of lip-reading using PCA is 64%, Time-domain filter applied to lip-reading enhances recognition rate of 72.4%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.1
• /
• pp.166-174
• /
• 2013

QRS detection of ECG is the most popular and easy way to detect cardiac-disease. But it is difficult to analyze the ECG signal because of various noise types. The important problem in recording ECG signal is a baseline wandering, which is occurred by rhythm of respiration and muscle contraction attaching to an electrode. Particularly, in the healthcare system that must continuously monitor people's situation, it is necessary to process ECG signal in realtime. In other words, the design of algorithm that exactly detects QRS region using minimal computation by analyzing the person's physical condition and/or environment is needed. Therefore, baseline wander removing method based on morphological filter for efficient QRS detection method is presented in this paper. For this purpose, we detected QRS through the preprocessing method using morphological filter, adaptive threshold, and window. The signal distortion ratio of the proposed method is compared with other filtering method. Also, R wave detection is evaluated by using MIT-BIH arrhythmia database. Experiment result show that proposed method removes baseline wanders effectively without significant morphological distortion.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.9
• /
• pp.2460-2470
• /
• 2009

In this study, the parameter extraction for evaluation of the anesthesia depth in each anesthesia stages was conducted. An object of the this experiment study has studied 5 adult patients (mean $\pm$ SD age:$42{\pm}9.13$), ASA classification I and II, undergoing surgery of obstetrics and gynecology. Anaesthesia was maintained with Enflurane. HRV signal was created by R-peak detection algorithm form ECG signal. The HRV data were preprocessing algorithm. It has tried find out the anesthesia parameter which responds the anesthesia events and shows objective anesthesia depth according to anesthesia stage including pre-anesthesia, induction, maintenance, awake and post-anesthesia. In this study, proposed algorithm to analysis the HRV(heart rate variability) signal using wavelet transform in anesthesia stage. Three sorts of wavelet functions applied to PSD. In the result, all of the results were showed similarly. But experiment results of Daubeches 10 is better. Therefore, this parameter is the best parameter in the evaluation of anesthesia stage.

• Journal of the Korean Child Neurology Society
• /
• v.26 no.4
• /
• pp.189-196
• /
• 2018

Purpose: Floppy infants or congenital hypotonia indicates decreased muscle tone in infants secondary to abnormalities of the central or the peripheral nervous system, or both. Previous literature classified its causes as those attributable to a central vs. peripheral origin; however, recent studies have introduced a newer classification describing a combined origin. We invenstigated floppy infants by applying the new etiological classification and reviewed the most common etiologies based on the age of presentation. We additionally reviewed the clinical characteristics, diagnoses, and the developmental outcomes in these infants. Methods: We retrospectively reviewed the electronic medical charts and recruited 116 infants diagnosed with floppy infant syndrome between January 2005 and December 2016 at Severance Children's Hospital. Among these infants, 66 with a confirmed diagnosis were reviewed for the etiological classification. Information regarding developmental outcomes was obtained via phone interviews with the infants' families. Results: Based on the new etiological classification, among 69 infants with a confirmed diagnosis, in 40 (34.5%) this syndrome was of central origin, in 19 (16.4%) of peripheral origin, and in 10 (8.6%) of combined origin. Prader-Willi syndrome, myotonic dystrophy, and spinal muscular atrophy were the most common disorders observed and combined hypotonia showed the poorest developmental outcome. Conclusion: The study states the importance of proper evaluation of etiological diagnosis and optimal intervention for developmental prognosis. The introduction of a new etiological group of combined hypotonia especially emphasizes regular monitoring and timely rehabilitative intervention in patients for the better quality of life in them as well as their caregivers.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.6
• /
• pp.33-43
• /
• 2003

In this paper, we proposed a real-time watermarking algorithm to be combined and to work with a DWT(Discrete Wavelet Transform)-based image compressor. To reduce the amount of computation in selecting the watermarking positions, the proposed algorithm uses a pre-established look-up table for critical values, which was established statistically by computing the correlation according to the energy values of the corresponding wavelet coefficients. That is, watermark is embedded into the coefficients whose values are greater than the critical value in the look-up table which is searched on the basis of the energy values of the corresponding level-1 subband coefficients. Therefore, the proposed algorithm can operate in a real-time because the watermarking process operates in parallel with the compression procession without affecting the operation of the image compression. Also it improved the property of losing the watermark and the efficiency of image compression by watermark inserting, which results from the quantization and Huffman-Coding during the image compression. Visual recognizable patterns such as binary image were used as a watermark The experimental results showed that the proposed algorithm satisfied the properties of robustness and imperceptibility that are the major conditions of watermarking.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.7
• /
• pp.343-362
• /
• 2020

English is a stress-timed language that has a phonetic system in which the speech is restructured by stress changes. On the other hand, Korean is a syllable-timed language in which each syllable is pronounced at almost the same length and intensity, and Korean and English have distinctly different metrical systems in general speech. However, as the language of the lyrics in K-Pop music is mixed in both languages, Korean and English, the Korean lyrics in K-Pop music have a metrical system by stress changes as in English. The writer's view is that the change in the metrical structure of Korean lyrics is inevitable in order to sustain the new Korean Wave. Therefore, in this study, dance music - a major genre of K-Pop music that focuses on rhythm expression - is classified into 1998, 2003, and 2009 according to the changes in the Korean Wave, and the metrical structure of each period is compared and analyzed. Based on this, the current K-Pop metrical structure features are derived and the K-Pop Korean writing method is proposed that deviates from the existing limited writing method which allocates one syllable per note. The author hopes this research will be used as a methodology for writing lyrics in Korean songs in K-Pop, as well as a way to encourage the use of Korean lyrics.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.7
• /
• pp.303-314
• /
• 2023

Voice conversion, a technology that allows an individual's speech data to be regenerated with the acoustic properties(tone, cadence, gender) of another, has countless applications in education, communication, and entertainment. This paper proposes an approach based on the StarGAN-VC model that generates realistic-sounding speech without requiring parallel utterances. To overcome the constraints of the existing StarGAN-VC model that utilizes one-hot vectors of original and target speaker information, this paper extracts feature vectors of target speakers using a pre-trained version of Rawnet3. This results in a latent space where voice conversion can be performed without direct speaker-to-speaker mappings, enabling an any-to-any structure. In addition to the loss terms used in the original StarGAN-VC model, Wasserstein distance is used as a loss term to ensure that generated voice segments match the acoustic properties of the target voice. Two Time-Scale Update Rule (TTUR) is also used to facilitate stable training. Experimental results show that the proposed method outperforms previous methods, including the StarGAN-VC network on which it was based.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.33-48
• /
• 2024

With the advent of 5G, characterized by Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), and Massive Machine Type Communications (mMTC), efficient network management and service provision are becoming increasingly critical. This paper proposes a novel approach to address key challenges of 5G networks, namely ultra-high speed, ultra-low latency, and ultra-reliability, while dynamically optimizing network slicing and resource allocation using machine learning (ML) and deep learning (DL) techniques. The proposed methodology utilizes prediction models for network traffic and resource allocation, and employs Federated Learning (FL) techniques to simultaneously optimize network bandwidth, latency, and enhance privacy and security. Specifically, this paper extensively covers the implementation methods of various algorithms and models such as Random Forest and LSTM, thereby presenting methodologies for the automation and intelligence of 5G network operations. Finally, the performance enhancement effects achievable by applying ML and DL to 5G networks are validated through performance evaluation and analysis, and solutions for network slicing and resource management optimization are proposed for various industrial applications.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.5
• /
• pp.55-68
• /
• 2021

The purpose of this study is to analyze the characters and narratives of series films and to extract corresponding elements of film music to compare and analyze how musical elements were utilized. The scope of the study was analyzing the story structure and characters of the films "Mary Poppins" and "Mary Poppins Returns" and the corresponding film music. After comparing the contents of the film "Mary Poppins" and "Mary Poppins Returns," the film matched the film music equivalent to the similar scenes of the two films. As a result, seven of the 11 songs of "Mary Poppins" overlap with those used in similar narratives of "Mary Poppins Returns", and eight songs overlap in "Mary Poppins Returns". Seven songs from "Mary Poppins" and eight songs from "Mary Poppins Returns" can be divided into nine scenes in total when connected to a common narrative. Among them, "A Spoonful of Sugar" from "Mary Poppins", "Jolly Holiday", "A cover is not the Book" from "Mary Poppins Returns" and "Triple light fantastic" were overlapping songs with narratives. Based on this, it analyzes leitmotiv film music, focusing on characters from the films "Mary Poppins" and "Mary Poppins Returns." The common leitmotivs in the two films are Mary Poppins leitmotiv, Lesson leitmotiv, Lullaby leitmotiv, World leitmotiv, Chimney Sweeper leitmotiv, Up & Down leitmotiv, Chimney Sweeper leitmotiv, and Sky leitmotiv. The characteristic rhythm and pitch used in Mary Poppins leitmotivs were used in the overall song featuring Mary Poppins. Through this, the elements of music symbolizing Mary Poppins, a key figure, were matched to the films "Mary Poppins" and "Mary Poppins Returns" and modified according to the narrative flow. The analysis results of this work have theoretical significance in that it is necessary to analyze the narratives and film music of series films to discover common features and consider how they are matched in theoretical terms.

• Korean Journal of Breeding Science
• /
• v.43 no.4
• /
• pp.233-240
• /
• 2011

Food security has been a main global issue due to climate changes and growing world population expected to 9 billion by 2050. While biodiversity is becoming more highlight, breeders are confronting shortage of various genetic materials needed for new variety to tackle food shortage challenge. Though biotechnology is still under debate on potential risk to human and environment, it is considered as one of alternative tools to address food supply issue for its potential to create a number of variations in genetic resource. The new technology, phenomics, is developing to improve efficiency of crop improvement. Phenomics is concerned with the measurement of phenomes which are the physical, morphological, physiological and/or biochemical traits of organisms as they change in response to genetic mutation and environmental influences. It can be served to provide better understanding of phenotypes at whole plant. For last decades, high-throughput screening (HTS) systems have been developed to measure phenomes, rapidly and quantitatively. Imaging technology such as thermal and chlorophyll fluorescence imaging systems is an area of HTS which has been used in agriculture. In this article, we review the current statues of high-throughput screening system in phenomics and its application for crop improvement.