• The Journal of the Korea Contents Association
• /
• v.16 no.5
• /
• pp.520-532
• /
• 2016

In this paper, we propose a novel method to analyze and quantitatively evaluate the impression of landscapes and soundscapes about traditional area of Chungnam Asan city in Korea. For this, we collated data about impression evaluation of Korean and Japanese students and, based on these result, we proved the validity and usefulness of the evaluation system that we proposed through this study. The research process conducted in this study as follows: first of all, we had selected five traditional places of interest in Asan City, and investigated various conditional elements of the area including soundscapes which is forming the landscape using video camera. And we made both Korean and Japanese students evaluate the impression for these videotaped landscape data through using the SD(semantic differential) method. And then, we quantified these qualitative data through applying the quantification theory and cluster analysis method to them. Though this process, we could obtain both data as a result derived by classifying each sample and categorizing levels of those impression evaluation. Because of totally difference between those two analyzing processes, which one is for sample classification and the other is for determining impression level, we could validate the usefulness of our evaluation system through conducting comparative analysis of results from both methods. Analysis showed that our novel evaluation system for landscape is effective and, in most part of the traditional landscape, Japanese students' responses are different far from the Korean students' responses.

• Journal of Korea Multimedia Society
• /
• v.4 no.5
• /
• pp.414-422
• /
• 2001

This paper proposes a classifier of music genre using time delay neural network(TDNN) fur an audio data retrieval systems. The classifier considers eight kinds of genres such as Blues, Country, Hard Core, Hard Rock, Jazz, R&B(Soul), Techno and Trash Metal. The comparative unit to classify the genres is a melody between bars. The melody pattern is extracted based un snare drum sound which represents the periodicity of rhythm effectively. The classifier is constructed with the TDNN and uses fourier transformed feature vector of the melody as input pattern. We experimented the classifier on eighty training data from ten musics for each genres and forty test data from five musics for each genres, and obtained correct classification rates of 92.5% and 60%, respectively.

• Electrical & Electronic Materials
• /
• v.17 no.8
• /
• pp.25-32
• /
• 2004

센서는 인간의 오감에 해당하는 감지기로서 온도, 소리, 빛 등의 물리, 화학적 신호를 전기적 신호로 변환시켜주는 일련의 장치를 말한다. 물론 센서에 대한 정의 및 분류 방법은 다양하게 표현될 수 있으나 본 논문에서는 MEMS(Micro Electro-Mechanical System) 기술을 적용하여 Smart화 혹은 Wireless화에 필요한 Micro 센서로 그 범위를 한정하고자 한다. 만일 센서를 아주 작게(Micro Sensor) 만들 수 있고 그것들끼리 무선으로 연결될 수(Wireless Network) 있다면 우리의 삶의 형태는 지금과는 매우 다른 모습으로 전개 될 것이다. 그림 1은 2004년에 정보통신부가 발표한 Ubiquitous Sensor Network의 개념도이다.(중략)

• Proceedings of the KSLP Conference
• /
• 1994.06a
• /
• pp.93-100
• /
• 1994

애성은 후두질환 특히 성대의 질환으로 인하여 소리의 구성요소인 음질(음색), 높이 세기(크기) 및 지속시간등에 이상이 있는 것을 말하며 그 정도의 성질은 여러 가지가 있으나 이것을 여러 방법으로 분류하는 것은 애성의 "메카니즘"을 알기 위해서도, 또 애성의 치료 효과를 알기 위해서도 유익하며 필요하다. 이제까지는 애성은 단지 주관적으로, 애성 (+), (++), (+)등으로 기록하여 왔으나 애성을 잘 들으면 보다 많은 진단에 도움이 되는 정보가 포함되어 있는 것을 알 수 있다. 이런 것을 상세하게 판정하여 기재하는 것은 장애의 진행, 개선, 수술후의 경과 등을 알기 위해서 임상상 크게 참고가 된다. (중략)

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.05a
• /
• pp.616-620
• /
• 2002

가청범위의 주파수보다 낮은 소리, 즉 초저주파음과, 가청음 중 주파수가 낮은 음을 합하여 저주파음으로 분류하기도 하는데, 대략 100Hz 이하의 주파수를 갖는 음을 가리킨다. 인간이 느끼는 저주파음에 대한 영향은 여러 범위로 조사되고 있으며 주로 청력손실과 불쾌함에 미치는 영향에 초점을 맞추고 있다.(중략)

• The Korean Journal of Zoology
• /
• v.39 no.2
• /
• pp.207-214
• /
• 1996

Five call types of the genus Hyla in habitats around South Korea were classified according to some attributes of their advertisement calls (note duration, note intenral, dominant frequency, sonagram patterns. Among the call types, the E-type was more distinctive than the other call types in that it had a metal sound and much longer note duration and note intenral. This result indicated that some divergence had occurred in the advertisement call of the genus Hylo, though this was found in alimited number of regions and its occurrence was small.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.293-299
• /
• 1995

본 논문은 컴퓨터 시각을 이용하여 동적 제스쳐를 인식하기 위한 효율적인 지식 표 현 기법의 개발을 목표로 한다. 제스쳐란 시각적인 언어로서 소리를 대신하여 몸짓이나 손 짓을 통하여 자신의 생각이나 의도를 전달하는 보조적인 의사 전달 수단이다. 제안된 기법 은 여러 다양한 지식을 통합하여 총체적으로 표현하기에 적합한 프레임 구조를 기반으로 한 다. 프레임 지식을 물체의 특성을 표현하는 객체 지식, 물체의 움직임을 표현하는 행동 지 식, 그리고 객체 지식과 행동 지식의 순서화 된 집함으로써 동적인 제스쳐를 표현하는 스키 마로 분류한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.4-6
• /
• 2003

A holistic representation, such as sparse ceding or independent component analysis (ICA), was successfully applied to explain early auditory processing and sound classification. In contrast, Part-based representation is an alternative way of understanding object recognition in brain. In this paper. we employ the non-negative matrix factorization (NMF)［1］which learns parts-based representation for sound classification. Feature extraction methods from spectrogram using NMF are explained. Experimental results show that NMF-based features improve the performance of sound classification over ICA-based features.

• Cartoon and Animation Studies
• /
• s.35
• /
• pp.329-346
• /
• 2014

Art is the product from the combination of politics, economy, and social and cultural aspects. Recent development of digital media has affected on the expansion of visual expression in art. Digital media allow artists to use sound and physical interaction as well as image as an plastic element for making a work of art. Also, digital media help artists create an interactive, synaesthetic and visual perceptive environment by combining viewers' physical interaction with the reconstruction of image, sound, light, and among other plastic elements. This research was focused on the analysis of the relationship between images in art work and the viewer and data visualization using sound from the perspective of visual perception. This research also aimed to develop an interactive art by visualizing physical data with sound generating from outer stimulus or the viewer. Physical data generating from outer sound can be analyzed in various aspects. For example, Sound data can be analyzed and sampled within pitch, volume, frequency, and etc. This researcher implemented a new form of media art through the visual experiment of LED light triggered by sound frequency generating from viewers' voice or outer physical stimulus. Also, this researcher explored the possibility of various visual image expression generating from the viewer's reaction to illusionary characteristics of light(LED), which can be transformed within external physical data in real time. As the result, this researcher used a motif from Piet Mondrian's Broadway Boogie Woogie in order to implement a visual perceptive interactive work reacting with sound. Mondrian tried to approach at the essence of visual object by eliminating unnecessary representation elements and simplifying them in painting and making them into abstraction consisting of color, vertical and horizontal lines. This researcher utilized Modrian's simplified visual composition as a representation metaphor in oder to transform external sound stimulus into the element of light(LED), and implemented an environment inducing viewers' participation, which is a dynamic composition maximizing a synaesthetic expression, differing from Modrian's static composition.

• Journal of Korea Multimedia Society
• /
• v.11 no.9
• /
• pp.1195-1203
• /
• 2008

Onomatopoeia of korean documents is to represent from natural or artificial sound to human language and it can express onomatopoeia language which is the nearest an object and also able to utilize as standard for clustering of Multimedia data. In this study, We get frequency of onomatopoeia in the experiment subject and select 100 onomatopoeia of use to our study In order to cluster onomatopoeia's relation, we extract feature of similarity and distance metric and then represent onomatopoeia's relation on vector space by using PCA. At the end, we can clustering onomatopoeia by using k-means algorithm.