• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.593-595
• /
• 2013

유비쿼터스 컴퓨팅은 IT기술을 기반으로, 의료 도시환경 건축 의류 등 사회 전반적인 분야로 확대되고 있다. 특히, 의료 기술의 발전과 고령인구의 증가, 삶의 질 향상에 대한 사람들의 욕구는 점점 헬스케어 산업의 중요성을 부각시켜주고 있다. 본 논문에서는 사용자 생체정보 측정을 위해 통신이 가능한 디지털 의류를 제작하여 생체신호를 측정할 수 있는 시스템을 제안하였다. 디지털 의류는 심전도 센서, 디지털 실, 모듈을 통해 구성되어 있으며 이를 안드로이드 어플리케이션을 통하여 스마트폰과 통신하도록 개발 하였다.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.55 no.5
• /
• pp.638-644
• /
• 2022

The measurement of morphological features is essential in aquaculture, fish industry and the management of fishery resources. The measurement of fish requires a large investment of manpower and time. To save time and labor for fish measurement, automated and reliable measurement methods have been developed. Automation was achieved by applying computer vision and machine learning techniques. Recently, machine learning methods based on deep learning have been used for most automatic fish measurement studies. Here, we review the current status of automatic fish measurement with traditional computer vision methods and deep learning-based methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.550-552
• /
• 2010

최근 행위인식 기술은 u-헬스케어 분야를 통해서 실용화되는 단계에 이르고 있다. 실생활에 적용단계에 있는 행위인식 시스템은 기존의 서버 및 데스크톱 환경에서 벗어나 모바일 기기를 기반으로 수행되도록 변해가고 있다. 모바일 환경에 적합한 행위인식 시스템은 행위인식 센서 및 모바일 행위인식 기기의 저전력화와 모바일 기기에 적합한 행위인식 시스템이 요구된다. 본 논문은 모바일기기 기반의 스마트 환경에 알맞은 행위인식 시스템을 위해 행위인식 알고리즘 경량화 기법 연구를 수행한다. 행위별 최적의 낮은 주파수를 사용하여 센싱에 소요되는 자원을 줄인 경량화된 행위인식 방법을 제안하고 또한 주파수 변화에 따른 윈도우간 적절한 오버랩 구간 설정 방법과 윈도우에서의 특징 검출 시 오버랩 구간의 중복 연산을 제거한 경량화된 특징 검출 방법을 제안한다. 실험 결과는 행위 별 최적의 낮은 주파수를 사용하여 전력을 줄이면서도, 서로 다른 주파수의 데이터임에도 인식률은 그대로 유지됨을 보인다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.345-348
• /
• 2020

노년층의 건강관리를 위하여 몸상태를 확인할 수 있는 심박센서와 운동량을 확인할 수 있는 모션센서 및 간단한 센서를 활용하여 실시간으로 건강분석과 관리 및 처방을 할 수 있는 방안을 제시하고 플랫폼 개발에 도전함.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.315-316
• /
• 2022

본 논문은 스마트 인솔의 압력 데이터를 이용하는 컨볼루션 신경망 모델에 해석가능한 인공지능 방법인 Grad-CAM을 적용하는 방법을 제안한다. 학습된 각 모델에 Grad-CAM을 적용하여 모델에서 중요한 역할을 하는 압력센서와 중요하지 않은 압력센서를 알아내는 방법을 제안하고 데이터마다 학습을 진행하고 학습된 모델을 통해 실제로 중요한 압력센서와 그렇지 않은 압력센서에 대해서 알아본다.

• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.165-176
• /
• 2021

Recently, interest in home training and changes in lifestyle are expanding the market in the healthcare field, and high value-added fashion products reflecting the dumbbell economy phenomenon are being released. This study had the following objectives: to investigate the possibility of developing a fashion item that can be applied to the dumbbell economy phenomenon; to develop the UI/UX of a smartphone application for beginners who wish to work out their upper and lower bodies regardless of time and space; and to create a wearable customized smart athleisure fashion device. First, the study identified factors related to exercise methods, breathing techniques, and range of exercises for beginners by investigating the postures of workouts of the upper and lower bodies by angles. Based on the results, the study collected empirical data through a user needs analysis from muscle strengthening exercise experts to verify the significance of the study and use as fundamental data. Second, the study developed the UI/UX of a smartphone application with three different contents: counting, suggesting exercise postures, and providing exercise calendars. Further, the study analyzed necessary user-centered concepts and characteristics in terms of design and technology and developed a wearable customized smart athleisure fashion device based on the results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.11
• /
• pp.5506-5511
• /
• 2012

u-Wellness is widely applicable to individuals and medical service providers such as hospitals and it includes u-fitness and video health counselling services at the side of the provider and stress management, obesity management, and the amount of exercise at the side of the individual. In this paper, we design and implement a smart health care system which uses the authentication device to identify an individual and the user's smart phone. Our system records and manages the amount of exercise on the basis of the prescription of health care professionals through the exercise equipment and Wi-Fi communication. Therefore, our system helps user do optimized amount of exercise through the health care professional's prescription. And our system quantifies the results of the measurement of body fat measuring machines and experts to build the database and automatically schedule.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.6
• /
• pp.1191-1196
• /
• 2017

Lately, the three mobile telecom companies in Korea are competing for the launch of Internet of Things services for using home. Typical launched services are in the smart home related fields. However, Internet of Things as mobile telecom based are at an early stage, expected that various services will be started continuously. At this point, we have been planning to analyze the preference of Internet of Things for objects based on the services already launched. In order to apply the analytic hierarchy method, the first stage factors were designed as Safety, Security, Health care, Intelligence and Home appliances. In addition, the second stage factors were organized into 18 detailed services presented in the conceptual model. As a result, Health care (23.2%) was the most preferred priority. These results can be interpreted as the result of interest in health by improving income. We presented the theoretical and practical implications of these results.

• Journal of Advanced Navigation Technology
• /
• v.19 no.5
• /
• pp.446-451
• /
• 2015

Recently, researches on sensor-based U-health applications that provide personal health information such as blood pressure, body temperature, and glucose, have actively been studied. The health information obtained via sensors, however, may have accuracy problems so that they can not be used unprocessed. This paper proposes a sensor abstraction layer for enhancing the accuracy of sensor readings from biomedical sensors that interact with smart phones. This layer recognizes sensor types and converts sensor readings into a form as specified in ISO/IEEE 11073 Personal Health Standard. When necessary, not only a filtering method that eliminates outlier values from sensor readings but also a summarization method that transforms them into more suitable forms, can also be applied. An android-based development board is used for the evaluation of proposed sensor abstraction layer. The results obtained by applying filtering and summarization show improved accuracy over unprocessed sensor readings of the body temperature and heartbeat sensors.

• Journal of Digital Convergence
• /
• v.14 no.11
• /
• pp.249-255
• /
• 2016

Since the elderly voices include a lot of noise caused by physiological changes in respiration, phonation, and resonance, the performance of the convergence health-care equipments such as speech recognition, synthesis, analysis program done by elderly voice is deteriorated. Therefore it is necessary to develop researches to operate health-care instruments with elderly voices. In this study, a voice activity detection using a symmetric higher-order differential energy function (SHODEO) was developed and was compared with auto-correlation function(ACF) and the average magnitude difference function(AMDF). It was confirmed to have a better performance than other methods in the voice interval detection. The voice activity detection will be applied to a voice interface for the elderly to improve the accessibility of the smart devices.