• 한국지능시스템학회논문지
• /
• 제18권5호
• /
• pp.586-591
• /
• 2008

본 논문은 음성을 사용한 인간의 감정 인식 시스템의 성능을 향상시키기 위하여 감정 변화에 강인한 음성 인식 시스템과 결합된 감정 인식 시스템에 관하여 연구하였다. 이를 위하여 우선 다양한 감정이 포함된 음성 데이터베이스를 사용하여 감정 변화가 음성 인식 시스템의 성능에 미치는 영향에 관한 연구와 감정 변화의 영향을 적게 받는 음성 인식 시스템을 구현하였다. 감정 인식은 음성 인식의 결과에 따라 입력 문장에 대한 각각의 감정 모델을 비교하여 입력 음성에 대한 최종감정 인식을 수행한다. 실험 결과에서 강인한 음성 인식 시스템은 음성 파라메터로 RASTA 멜 켑스트럼과 델타 켑스트럼을 사용하고 신호편의 제거 방법으로 CMS를 사용한 HMM 기반의 화자독립 단어 인식기를 사용하였다. 이러한 음성 인식기와 결합된 감정 인식을 수행한 결과 감정 인식기만을 사용한 경우보다 좋은 성능을 나타내었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.178.5-178
• /
• 2001

We, Human beings, use both powers of reason and emotion simultaneously, which surely help us to obtain flexible adaptability against the dynamic environment. We assert that this principle can be applied into the general system. That is, it would be possible to improve the adaptability by covering a digital oriented information processing system with an analog oriented emotion layer. In this paper, we proposed a vertical slicing model with an emotion layer in It. And we showed that the emotion-based control allows us to improve the adaptability of a system at least under some conditions.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2008년도 학술대회 1부
• /
• pp.7-12
• /
• 2008

사람과 로봇의 인터랙션에 있어 정서의 역할은 중요하다. 그러므로 사람과 유사한 정서 메커니즘이 로봇에게 필요하다. 본 논문에서는 사람의 정서에 대한 심리적 연구를 바탕으로 사람의 내적 요소에 기반한 새로운 인공 정서 표현 시스템을 제안한다. 제안된 시스템은 external stimulus, emotion, mood, personality, tendency, machine rhythm 의 6가지 정보 요소를 활용하며, 각 요소들은 그들의 특성에 따라 결과적으로 emotion 표현 패턴의 변화에 영향을 준다. 그 결과 통일한 외부 자극들은 내적 상태에 따라 emotion의 변화를 만들어 낸다. 제안된 시스템은 사람과 로봇의 자연스러운 인터랙션을 유지하고, 친밀한 관계를 형성할 수 있도록 도움을 줄 것이다.

• 한국콘텐츠학회논문지
• /
• 제15권3호
• /
• pp.21-31
• /
• 2015

본 논문에서는 기존에 연구되었던 Thayer의 인간의 감성 모델을 바탕으로 음악을 듣고 느끼는 감성을 8가지 감성으로 정의한 음악 감성모델로 제시하였다. 감성에 영향을 주는 음악의 요소는 음악의 템포, 역동성, 진폭변화, 밝기, 잡음 등 5가지로 선택하였다. 8가지 감성으로 이루어진 감성 모델에 따라서 8가지 감성으로 분류된 160곡의 노래를 선곡하여 실제 데이터를 추출하여 분석하였다. 실제 데이터의 분석을 통해 5가지 요소의 가중치로 이루어진 감성 수식을 도출하였고 임의의 음악에 대하여 감성 수식을 통해서 2차원 감성 좌표계에 매핑 하여 감성을 예측할 수 있도록 알고리즘을 설계하였다. 또한 2차원 감성 좌표계에서의 좌표 값을 이동시켜 감성을 제어할 수 있는 방법을 제시하였다.

• 한국의류산업학회지
• /
• 제7권2호
• /
• pp.187-195
• /
• 2005

The purposes of this study were to investigate the effects of the elements of textile design on consumer emotion and to develop the emotion model which is suitable for the textile design. The descriptive system of textile design was developed based on the previous studies. Emotion measurement scale was developed to analyze the consumer emotion for the textile design. 20 representative types of textile design were collected as stimuli set for this study, consumer emotion on each design type was examined and was analyzed through the survey. For the data analysis, principal component analysis was employed. As a result, 8 emotional factors such as 'Modern', 'Fun', 'Natural', 'Elegance', 'Classic', 'Ethnic', 'Wild' and 'Sporty' were derived from the results of the survey. Emotion measurement scale which consisted of 8 factors was developed to analyze the effects of the elements of textile design on consumer emotion and 80 representative types of textile design were collected. In addition, the emotion which consumers feel for the textile design types was investigated and each textile design was described according to the descriptive system of textile design. Statistical methods of pearson correlation and multiple regression were employed to analyze the relationship between the elements of textile design and consumer emotion. The results of this study revealed that 15 design elements which affected consumer emotion were the size of motives, the shape of motives, the degree of tone contrast among motives etc. This study findings can provide specific design methods for the effectiveness of consumer emotion.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
• /
• pp.1243-1246
• /
• 2003

It is known that structure of Human's brain information system is controlled by cerebral cortex mainly. Cerebral cortex is divided by sensory area, motor area and associated area largely. Sensory area takes part in information from environment and motor area is actuation by decision as associated area determined. It is possible to copy brain information system by input-output pattern. but there is difficulty in modeling of memorizing new information. Such action is performed by Limbic Lobe and Papez circuit which is controlled by intrinsic emotion. So we need of definition of emotion's role in decision. In this paper, we define roles of emotion in intrinsic decision using Dynamic Cognitive Maps(DCMs). The emotion is evaluated by outside information then intrinsic decision performed as how much emotion variated. The dynamic cognitive maps take part in emotion evaluating process.

• 산업경영시스템학회지
• /
• 제37권4호
• /
• pp.1-11
• /
• 2014

This paper addresses the emotion computing model for software affective agents. In this paper, emotion is represented in valence-arousal-dominance dimensions instead of discrete categorical representation approach. Firstly, a novel emotion model architecture for affective agents is proposed based on Scherer's componential theories of human emotion, which is one of the well-known emotion models in psychological area. Then a fuzzy logic is applied to determine emotional statuses in the emotion model architecture, i.e., the first valence and arousal, the second valence and arousal, and dominance. The proposed methods are implemented and tested by applying them in a virtual training system for children's neurobehavioral disorders.

• 전기학회논문지
• /
• 제66권8호
• /
• pp.1265-1271
• /
• 2017

In this paper, the speech parameter fusion method is studied to improve the performance of the conventional emotion recognition system. For this purpose, the combination of the parameters that show the best performance by combining the cepstrum parameters and the various pitch parameters used in the conventional emotion recognition system are selected. Various pitch parameters were generated using numerical and statistical methods using pitch of speech. Performance evaluation was performed on the emotion recognition system using Gaussian mixture model(GMM) to select the pitch parameters that showed the best performance in combination with cepstrum parameters. As a parameter selection method, sequential feature selection method was used. In the experiment to distinguish the four emotions of normal, joy, sadness and angry, fifteen of the total 56 pitch parameters were selected and showed the best recognition performance when fused with cepstrum and delta cepstrum coefficients. This is a 48.9% reduction in the error of emotion recognition system using only pitch parameters.

• 대한인간공학회지
• /
• 제31권2호
• /
• pp.271-279
• /
• 2012

Objective: This study classified three different emotional states(boredom, pain, and surprise) using physiological signals. Background: Emotion recognition studies have tried to recognize human emotion by using physiological signals. It is important for emotion recognition to apply on human-computer interaction system for emotion detection. Method: 122 college students participated in this experiment. Three different emotional stimuli were presented to participants and physiological signals, i.e., EDA(Electrodermal Activity), SKT(Skin Temperature), PPG(Photoplethysmogram), and ECG (Electrocardiogram) were measured for 1 minute as baseline and for 1~1.5 minutes during emotional state. The obtained signals were analyzed for 30 seconds from the baseline and the emotional state and 27 features were extracted from these signals. Statistical analysis for emotion classification were done by DFA(discriminant function analysis) (SPSS 15.0) by using the difference values subtracting baseline values from the emotional state. Results: The result showed that physiological responses during emotional states were significantly differed as compared to during baseline. Also, an accuracy rate of emotion classification was 84.7%. Conclusion: Our study have identified that emotions were classified by various physiological signals. However, future study is needed to obtain additional signals from other modalities such as facial expression, face temperature, or voice to improve classification rate and to examine the stability and reliability of this result compare with accuracy of emotion classification using other algorithms. Application: This could help emotion recognition studies lead to better chance to recognize various human emotions by using physiological signals as well as is able to be applied on human-computer interaction system for emotion recognition. Also, it can be useful in developing an emotion theory, or profiling emotion-specific physiological responses as well as establishing the basis for emotion recognition system in human-computer interaction.

• 한국멀티미디어학회논문지
• /
• 제18권9호
• /
• pp.1068-1082
• /
• 2015

Due to the rapid advancement of the mobile technology, smart phones have been widely used in the current society. This lead to an easier way to retrieve video contents using web and mobile services. However, it is not a trivial problem to retrieve particular video contents based on users' specific preferences. The current movie recommendation system is based on the users' preference information. However, this system does not consider any emotional means or perspectives in each movie, which results in the dissatisfaction of user's emotional requirements. In order to address users' preferences and emotional requirements, this research proposes a movie recommendation technology to represent a movie's emotion and its associations. The proposed approach contains the development of emotion ontology by representing the relationship between the emotion and the concepts which cause emotional effects. Based on the current movie metadata ontology, this research also developed movie-emotion ontology based on the representation of the metadata related to the emotion. The proposed movie recommendation method recommends the movie by using movie-emotion ontology based on the emotion knowledge. Using this proposed approach, the user will be able to get the list of movies based on their preferences and emotional requirements.