• 전자공학회논문지
• /
• 제54권6호
• /
• pp.100-105
• /
• 2017

본 논문은 개인 인증 알고리즘에 관한 것으로 심전도를 이용한 생체 인증 방식은 특정 보정기준점을 추출하는 방법과 그렇지 않은 방법으로 분류할 수 있으며 본 논문에서 제안하는 방법은 특정 보정기준점을 추출하지 않는 방법으로 이산 코사인 변환과 랜덤 포레스트 분류기를 사용하였다. 심전도 신호는 R-Peak 점을 기준으로 단일 심박으로 나누었으며 각 심박의 특징 추출을 위해 이산 코사인 변환을 적용하였다. 이산 코사인 변환 계수는 정보가 저주파에 집중되는 특성이 있으므로 초기 저주파에 해당하는 40까지 값을 특징으로 랜덤 포레스트 분류기를 구성하였다. 랜덤 포레스트는 의사결정 트리의 앙상블 분류기로 결정 트리를 기본으로 하고 있으므로 빠른 학습 속도와 많은 양의 데이터 처리 능력, 다양한 클래스를 분류할 수 있어 실생활에 적용 가능하며 무엇보다 ID의 승인과 거절을 위한 임계값을 분류기 내부에서 조절할 수 있어 오 분류에 강건한 알고리즘을 구성할 수 있다. 18개의 심전도 파일로 구성된 MIT-BIT Normal Sinus Rhythm 데이터베이스를 선정하여 성능을 평가하였으며 99.99%의 심전도 인식률을 보였다.

• 한국음향학회지
• /
• 제27권3호
• /
• pp.111-118
• /
• 2008

본 논문에서는 음악 장르 분류를 위한 새로운 자동 Taxonomy 구축 알고리즘을 제안한다. 제안된 알고리즘은 모든 가능한 노드들의 분류 확률을 예측하여 예측된 분류 성능값이 가장 좋은 조합을 Taxonomy로 구축하는 것이다. 제안된 알고리즘에서의 분류 확률 예측은 훈련 데이터를 k-fold cross validation을 이용하여 분류기에 적용함으로써 이루어진다. 제안된 알고리즘을 기반으로 한 분류 성능 측정은 2 클래스로 이루어진 각각의 노드에 2개 범주 분류에 효과적인 support vector machine을 적용함으로써 이루어진다. 제안된 알고리즘의 성능 검증을 위해 음색, 리듬, 피치 등 오디오 신호의 특징을 나타내는 다양한 파라미터를 오디오 신호로부터 추출하여 제안된 알고리즘과 기존의 다중 범주 분류기들을 이용하여 분류성능을 평가하였다. 다양한 실험결과 제안된 알고리즘은 기존의 알고리즘에 비하여 5%에서 25%정도의 분류 성능이 향상된 것을 확인할 수 있었고 특히 낮은 차원의 특징벡터를 이용한 분류 실험에서는 10% 에서 25% 향상된 좋은 성능을 보였다.

• BMB Reports
• /
• 제49권11호
• /
• pp.587-589
• /
• 2016

The circadian clock system enables organisms to anticipate the rhythmic environmental changes and to manifest behavior and physiology at advantageous times of the day. Transcriptional/translational feedback loop (TTFL) is the basic feature of the eukaryotic circadian clock and is based on the rhythmic association of circadian transcriptional activator and repressor. In Drosophila, repression of dCLOCK/CYCLE (dCLK/CYC) mediated transcription by PERIOD (PER) is critical for inducing circadian rhythms of gene expression. Pacemaker neurons in the brain control specific circadian behaviors upon environmental timing cues such as light and temperature cycle. We show that amino acids 657-707 of dCLK are important for the transcriptional activation and the association with PER both in vitro and in vivo. Flies expressing dCLK lacking AA657-707 in $Clk^{out}$ genetic background, homologous to the mouse Clock allele where exon 19 region is deleted, display pacemaker-neuron-dependent perturbation of the molecular clockwork. The molecular rhythms in light-cycle-sensitive pacemaker neurons such as ventral lateral neurons ($LN_vs$) were significantly disrupted, but those in temperature-cycle-sensitive pacemaker neurons such as dorsal neurons (DNs) were robust. Our results suggest that the dCLK-controlled TTFL diversify in a pacemaker-neuron-dependent manner which may contribute to specific functions such as different sensitivities to entraining cues.

• Journal of Information Processing Systems
• /
• 제13권2호
• /
• pp.400-416
• /
• 2017

Music emotion is an important component in the field of music information retrieval and computational musicology. This paper proposes an approach for automatic emotion classification, based on rough set (RS) theory. In the proposed approach, four different sets of music features are extracted, representing dynamics, rhythm, spectral, and harmony. From the features, five different statistical parameters are considered as attributes, including up to the $4^{th}$ order central moments of each feature, and covariance components of mutual ones. The large number of attributes is controlled by RS-based approach, in which superfluous features are removed, to obtain indispensable ones. In addition, RS-based approach makes it possible to visualize which attributes play a significant role in the generated rules, and also determine the strength of each rule for classification. The experiments have been performed to find out which audio features and which of the different statistical parameters derived from them are important for emotion classification. Also, the resulting indispensable attributes and the usefulness of covariance components have been discussed. The overall classification accuracy with all statistical parameters has recorded comparatively better than currently existing methods on a pair of datasets.

• 한국멀티미디어학회논문지
• /
• 제4권5호
• /
• pp.414-422
• /
• 2001

본 논문에서는 오디오 데이터의 효과적인 검색을 위하여, 시간지연신경망을 이용한 음악 장르 분류 시스템을 제안한다. 분류 대상 장르는 Blues, Country, Hard Core, Hard Rock, Jazz, R&B(Soul), Techno, Trash Metal의 8종류이다. 장르를 분류하기 위한 비교단위는 곡 중에서의 한 마디이다. 이러한 마디는 리듬의 특성을 효과적으로 반영하는 스네어 드럼 소리를 기준으로 추출한다. 분류기는 시간 지연 신경망을 이용하여 구성하며 입력은 추출된 마디에 대한 주파수 특징벡터이다. 제안한 시스템의 유효성을 검증하기 위한 실험에서, 장르별 10곡씩 총 80곡의 학습 데이터와 장르별 5곡씩 총 40곡의 테스트 데이터에 대하여 각각 92.5%와 60%의 정인식율을 보였다

• 복식
• /
• 제53권5호
• /
• pp.13-22
• /
• 2003

The purposes of this study are to develop a Korean image clothing design for cultural products by using the literary artist´s works and to satisfy the demand of the local self-government and of Gwangju ㆍChonnam area´s small and medium-sized companies. The image of traditional Korean clothing can be expressed with adjectives like Pure, plain, rhythmical, neatness, plentiful, and calm etc. These attributes are similar to the characteristics of the literary artist´s style, which is expressed as suggestive, animated, spatial, possibility of stationary, serene and still. The features of the traditional Korean clothing design can be described as ´The Beauty of Elegance´, ´The Beauty of Plainness´, ´The Beauty of Asymmetric Balance´ and ´The Beauty of rhythm´. It can be expressed in the modern fashion design, which is developed through the combination of the most important feature like ´The Beauty of Space´ in the literary artist´s style. At the time of production we tried to express spatial beauty and plentiful taste of the literary artist's style in the clothing design of Korean image for cultural products by considering the balance between them. Six works of the literary artist were selected which satisfied the purpose of the developed clothing design item. The results were nine clothing designs of Korean image, like a skirt suit made up of a blouse and a skirt, a pants suit made up of pants and a jacket, two evening dresses with a shawl, and also T-shirts and one-pieces etc.

• 한국실내디자인학회논문집
• /
• 제25권4호
• /
• pp.43-50
• /
• 2016

This study tries to categorize trends of "structure and skin integration" and understand the expressive characters of each architectural type. To do so, we listed up 8 architects who are quoted twice or more in related researches, then analyzed their 25 contemporary buildings which integrated structure and skin since 2000. As a result, this study defined four types based on the way of building tectonic system of structure and skin. Key feature of "linear structure-two dimensional skin" type is the communication with the surroundings as a result of .geometric architectural forms, patterned surface and reflection. Characters of "linear-three dimensional" type are organic architectural forms, sculptural skin, and the mysterious space. "Planar-two dimensional" type is a transformational geometry form to express the dramatic images through the skin, therefore gives a sense of rhythm and dynamics to space. "Planar structure-three dimensional" type highlights the texture, and exposes boundary of the inside and outside. In architectures we studied, the structure is the way to make a creative forms and space, and the skin to express various meanings. That said, the "structure and skin integration" is the means of aggressive design expression.

• 한국언어정보학회:학술대회논문집
• /
• 한국언어정보학회 2007년도 정기학술대회
• /
• pp.405-414
• /
• 2007

This paper describes a preliminary work on prosody modeling aspect of a text-to-speech system for Thai. Specifically, the model is designed to predict symbolic markers from text (i.e., prosodic phrase boundaries, accent, and intonation boundaries), and then using these markers to generate pitch, intensity, and durational patterns for the synthesis module of the system. In this paper, a novel method for annotating the prosodic structure of Thai sentences based on dependency representation of syntax is presented. The goal of the annotation process is to predict from text the rhythm of the input sentence when spoken according to its intended meaning. The encoding of the prosodic structure is established by minimizing speech disrhythmy while maintaining the congruency with syntax. That is, each word in the sentence is assigned a prosodic feature called strength dynamic which is based on the dependency representation of syntax. The strength dynamics assigned are then used to obtain rhythmic groupings in terms of a phonological unit called foot. Finally, the foot structure is used to predict the durational pattern of the input sentence. The aforementioned process has been tested on a set of ambiguous sentences, which represents various structural ambiguities involving five types of compounds in Thai.

• 패션비즈니스
• /
• 제5권4호
• /
• pp.101-117
• /
• 2001

Artist's creative works is accomplished in the very complicated network being in fluenced each other, getting celebrate from surround and center, and showing repetition and change. It took an important role of 1920's culture with new social and mental feature generated by revolution. It also has been influenced on an intellectual paradigm and artistic tendency. According to the feather the Art to Wear of Constructivism, at first, design as absolution of folk art in Russia prior to the concept is coming up to systematic design and useful article's performance. Courageous arrangement of color and rhythm is unique. The second is a tendency clothing design of productirism. It is connected with material culture and artists insisted creature of productive material that is a essential in life than Arts in museum. The third, tendency clothing design of minimalism emphasis functional aspect that is produced industrial aesthetics. The examples are simply designed and functional cloths for sorking uniform. This thesis is studied about clothing and textiles of major artists, formative property, and international reflection from October revolution in 1917 that is strengthen a system by introducing a new economic policy to a five year plan.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.90-98
• /
• 2018

심전도 신호는 기본적으로 심장의 전기적 활동에 포함되며 이를 통해 심박수 측정, 심장 박동의 리듬 검사, 심장 이상 진단, 정서 인식 및 생체 인식과 같은 다양한 목적으로 분석 및 활용된다. 본 논문의 목적은 다차원 데이터 배열인 텐서 특성을 가진 다선형 판별분석(MLDA: Multilinear Linear Discriminant Analysis) 기법을 이용하여 개인식별을 수행하고자 한다. MLDA는 상위 차원의 텐서를 포함하는 분류 문제에 대해서 차원 문제를 해결 할 수 있으며, 상호 연관된 부분 공간은 서로 다른 클래스를 구별하기 위해 사용될 수 있다. 제시된 방법의 성능을 검증하기 위해 Physionet의 MIT-BIH데이터베이스를 적용하였다. 이 데이터베이스에 대해 실험한 결과, MLDA는 기존 PCA와 LDA와 비교하여 개인식별 성능이 우수함을 확인하였다.