• Convergence Security Journal
• /
• v.12 no.4
• /
• pp.41-47
• /
• 2012

For competitiveness of national B2C (Business to Customer) service industry, improvement of process and analysis focused on customer and change of service system are needed. In other words, a business and an organization should deduce and provide what kind of services customers want. Then, evaluate customers' satisfaction and improve the service quality. To achieve this goal, accurate feedbacks from customers play an important role; however, there are not quantitative and standard systems a lot in nation. Recently, the researches about ICT (Information and Communication Technology) that can recognize emotion of human being are on the increase. The facial expression recognition among them is known as most efficient and natural human interface. This research analyzes about more efficient facial expression recognition and suggests a customer satisfaction feedback system using that.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1789-1790
• /
• 2007

이 논문에서 우리는 생체모방 시스템을 구현하기 위해 일반적인 기계 시스템과 인간의 신경 진동자 모델을 결합하였다. 이러한 시스템은 외부환경의 변화에 따른 효과적인 자율 적응 운동 형태를 생성할 수 있다. 인간 및 동물의 주기적 자율 운동을 관장하는 Central Pattern Generator (CPG)는 신경 진동자 네트웍에 의해서 표현가능하고 이는 신경 진동자 모델 내부의 sensory 피드백 신호를 통해, 주기성을 같은 외란에 상호 작용하여 적절한 운동을 생성해 낸다. 따라서 이를 기계 시스템에 결합하면 이러한 시스템은 변화되는 환경이나 잘 알지 못하는 외란에 대하여 자율적으로 적응된 운동을 보일 수 있다. 이를 위해 본 논문은 이러한 신경 진동자 모델과 결합된 realtime 시스템을 구현하고 그 자율 적응 운동의 생성 가능성을 살펴본다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.679-682
• /
• 2010

피부 영역 검출 기술은 생체 인식 기술의 하나로서 얼굴 자동 인식 혹은 손 모양 자동 인식 등을 위해 사용되고 있다. 일반적으로 색상을 이용하여 피부 영역을 검출하기 위해서는 다양한 피부색 샘플을 이용해 구해진 피부색 모델을 이용한다. 하지만 피부색은 사람마다 다르고, 조명과 같은 주변 환경의 영향도 받기 때문에 다양한 영상에 하나의 고정된 피부색 모델을 적용하여 피부 영역을 검출하기에는 한계가 있다. 이러한 문제점을 해결하기 위해 본 논문에서는 영상 피드백 방법을 이용하여 영상에 적응적인 피부색 모델을 구한 뒤 이를 적용하여 피부 영역을 추출하는 방법을 제안한다.

• The Journal of the Korea Contents Association
• /
• v.5 no.1
• /
• pp.19-26
• /
• 2005

This paper introduces a new design method of realizing the machine control interface by using bio-signals(EEG/EOG). This method can be further expanded to be applied to the computer system responding to EEG or EOG signals and the general bio-feedback system. For this reason, we made the remotely controlled toy system controlled by the EEG spectrums, their combination indexes, and EOG parameters. And the headset that has bio-signal processing modules built-in offers convenience for users, and this make much more advanced system than any other existing BCI and BMI system.

• Journal of Biomedical Engineering Research
• /
• v.20 no.2
• /
• pp.183-190
• /
• 1999

The purpose of this study was to develop a portable and convenient closed-loop contrel type electrical stimulator for patients with foot drop. This system restores walking movement as well as prevents from atrophy or necrosis of lower limb muscles and increases blood circulation in hemiplegic patients caused by traffic accident, industrial disaster or stoke. This system detects the changes of the ankle joint angle during walking, and then controls the stimulus intensity automatically to maintain the programmed level of the ankle joint angle. Also, this automatic system controls the stimulus intensity which is affected by increased electrode impedance resulting from long time use. The system detects the joint angle by an optical sensor and includes modified PID control which adjusts the stimulus intensity if the joint angle deviates from the preset value. Stimulus parameters are 30~80 volt, 40 Hz, and 0.2 ms. The system was applied to five hemiplegic patients for 42 days. Duration of stimulation was 15 min/day for the first week and then the duration was gradually increased to 30, 60, 90 and 120 min/day. The muscle force was increased up to 29.7%, muscle fatigue was decreased compared with the level before stimulation and the pattern of locomotion was improved. These results suggest that the electrical stimulator with closed-loop control type is more convenient and effective in restoration of locomotion of patients with foot drop than open-loop system.

• Proceedings of the KOSOMBE Conference
• /
• v.1995 no.11
• /
• pp.282-285
• /
• 1995

In this paper, we present a feedback structured adaptive RF filter based on the recursive modified Gram-Schmidt algorithm for short-term prediction of EEG signal. And the performance of this proposed filter is compared with those of linear AR model, RF filter, Volterra filter and RBF neural network as single-step prediction and multi-step prediction. The results show the superiority of this proposed filter in prediction of EEG signals.

• Electronics and Telecommunications Trends
• /
• v.24 no.5
• /
• pp.38-51
• /
• 2009

정보통신망을 이용하여 언제 어디서나 모든 사람이 의료서비스를 제공받을 수 있고, 나아가서는 생활관리서비스를 제공하는 유비쿼터스 건강관리 사회를 구현하기 위하여 생체정보 센서, 생체정보 모니터링, 데이터 분석, 피드백 등의 4가지 핵심기술에 대한 개발이 진행되고 있다. 본 고에서는 현장진단이 가능한 바이오센서 기술에 대한 개요와 국내외 기술개발 동향 및 특허 동향을 분석하고 향후 현장진단용 바이오센서 시장에 대하여 전망하고자 한다.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2009.05a
• /
• pp.195-198
• /
• 2009

본 연구는 사용자가 경험하는 감성 이미지를 추출하여 생리신호를 통해 자동으로 분류하는 시스템을 구현하고자 한다. 본 연구에서 개발한 시스템은 이미지의 감성평가를 위하여 착용과 휴대가 간편한 PPG생체신호를 활용하였고, 생활 속의 장면들에서 감성 별 이미지를 무선으로 PC에 저장 및 분류하는 시스템을 개발하였다. 또한 이미지의 라이프 로그 정보를 제공하는 기술을 개발하였다. 본 시스템은 사용자의 경험과 이미지를 연관시켜 생활 속 정보의 경험 피드백을 하는 데에 의의가 있다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.185-190
• /
• 2020

The rapid aging is increasing as the shortage of medical facilities and the resulting of decline in the quality of public health. In order to ease the burden of rising medical expenses, advanced medical institutions are expanding their remote medical care to lower the cost of services. U-healthcare detects the changes in physical and chemical phenomena occurring in the human body and converts them into electrical signals that can be processed and feeds back to the results through analytical and visualization processes to select only the desired information from the measured signals. The service is provided through a process of providing an alarm to a user. However, traditional biometric methods of attaching sensors directly to the body can be annoying and rejected in daily life. Therefore, there is a need for a method of continuously measuring biometric information without causing inconvenience to daily life. In this paper, we propose an IR-UWB-based non-contact and non-responsive respiratory measurement system that can continuously monitor biological information without any inconveniences to daily life.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.312-315
• /
• 2008

Recently, owing to the development of ubiquitous, RFID and local area wireless communication technology, many studies on the system which can measure biomedical signals are being carried out. In this paper, we have designed and implemented an u-Healthcare system based on sensor network using biomedical signal measurement sensors such as ECG, blood pressure, and heartbeats. The biomedical signals from sensor nodes pass through the gateway and are finally transmitted to a healthcare renter. The acquired biomedical signals are processed in the healthcare center and the analyzed results are transmitted to the patients to improve patients' health using either kinesitherapy or dietary treatment.