• Annual Conference of KIPS
• /
• 2017.11a
• /
• pp.795-798
• /
• 2017

근래에 들어, 웨어러블 기기의 발전으로 사람의 움직임에 대한 측정이 손쉬워 지면서, 워킹, 러닝, 사이클링 등의 인간의 신체 활동 상태를 감지하여 더 효율적인 운동을 할 수 있도록 정보를 획득, 제공하려는 연구가 계속되고 있다. 본 연구에서는 웨어러블 기기중 하나인 스마트 인솔을 통해서 수집되는 가속도 정보와 압력 정보를 사용하여 운동시에 사람의 운동 자세를 감지하고 측정하는 시스템을 구현하였다. 사람이 헬스센터에서 수행하는 각각의 자세는 운동의 특성에 따라 시계열 신호의 표현 패턴이 다르게 나타나며 이 패턴을 통한 정확한 자세의 감지를 위해서 본 연구에서는 다양한 신호처리 알고리즘을 사용하였으며 이 경우 더 정확한 자세를 측정할 수 있음을 알 수 있었다. 따라서 본 연구에서는 정확한 자세의 감지를 위해 운동의 특징에 따라 알고리즘을 선택하여 시계열 정보를 처리 분석 하는 시스템을 제안하였으며 이를 통해 보다 정확하게 사람의 신체활동을 분석할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.260-263
• /
• 2016

최근 스마트폰을 이용한 사용자의 운동량 계산 어플리케이션이 많이 출시되고 있다. 이러한 어플리케이션은 스마트폰의 GPS, 가속도 등의 내장 센서들을 이용하여 사용자의 행위를 분석하고, 분석된 내용을 토대로 운동량을 단순 계산해 주는 것들이 많다. 이러한 단순 계산은 사용자의 행동을 지정된 행동으로 인식하여 계산하기 때문에 복잡한 운동을 할 경우 어플리케이션에서 측정하는 칼로리 소모량과 실제 칼로리 소모량 계산과 비교해보면 오차가 많이 나는 경우가 많다. 이러한 오차를 줄이고자 신발에 압전센서와 블루투스 통신을 포함한 모듈을 삽입하여 사용자의 이동거리와 발걸음 수를 정확히 측정하여 사용자의 운동을 인식하고, 측정된 데이터를 이용하여 운동이 끝난 뒤 발걸음수와 이동거리를 이용하여 사용자의 소모된 칼로리를 정확하게 측정할 수 있는 방법을 제시하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.6
• /
• pp.1300-1305
• /
• 2012

In this research, we propose a momentum measurement system using the accelerometer sensor, MCU, and Bluetooth to measure the exact momentum. The proposed system can figure out information for the real time travel distance. We performed various experiments, and analyzed the results using the proposed momentum measurement system. In the simulation experiments, we compared the reliability and accuracy for the existing momentum measurement systems from the analyzed results. The proposed system showed travel distance error of less than 8% as compared to the existing system with the error of approximately 13%. We expect that the proposed system apply to the commercial products.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.40 no.3
• /
• pp.189-198
• /
• 2003

Videooculography (VOG) is one of the eye-movement measurement methods used for objective evaluation of vestibule -ocular reflex. A key feature of VOG is to estimate accurately the center of pupil and ocular torsion with being less influenced by the upper eyelid droop, eyelashes, corneal reflection, and eye blinks. Especially, it Is important to find the accurate center of the pupil in 3-D VOG because the inaccurate pupil center causes significant errors on measuring torsional eye movement. A new algorithm was proposed to find the center of pupil which is a little influenced by factors mentioned above. In this study, real time three-dimensional VOG which can measure horizontal, vortical, torsional eye movements, and the diameter of pupil was implemented using the proposed method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.3
• /
• pp.615-625
• /
• 2012

To measure the athletic information of exercisers, the applications of smartphone are programmed based on the sensing data from GPS device. These applications provide exercisers for running or walking distance, exercising time, calorie consumption, average speed, and so on. Among them, the exercising distance should measure accurately because it directly affects the other athletic information for exercisers. However, the existing methods for measuring the exercising distance makes errors because they are worked on the simple sphere or ellipse earth models. Actually, the surface of real earth is composed of inclined ground like hills and valleys. In this paper, a new exercising distance measuring algorithm is proposed to compensate the errors of existing method. It considers the altitude of slopes in exercising routes. To evaluate exercising distance measuring algorithms, we implement the athletic life-guide system based on the smartphone platform. In experiments, the proposed method shows that it provides more accurate distance measurement.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.8-16
• /
• 2017

There are various methods to count the number of repetitive motions such as jump rope. Most of the methods use features extracted from the time-varying waves of acceleration or angular velocity, which is the main feature in the count of rotations in jump rope. However, there exist several variables and it is not easy to find the count with a single sensor. For example, accelerometer is susceptible to noise and vibration, and the angular velocity may cause a drift phenomenon, which is the main cause of the inaccurate count of jump rope rotation. In this paper, complementary filter is used to consider two sensors simultaneously and complement each other, which results in more accurate count in jump rope rotation. The proposed method can count the exact number of jump rope rotation compared to other existing methods only using one sensor value, which is confirmed through experimental results.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.11
• /
• pp.2697-2702
• /
• 2015

In the existing studies providing a method which do exercise according to music speed, the method was used which a user changes playing musics. However, if the method which a user changes directly playing musics is used, flow of exercise can be discontinued during searching musics when anyone want to play a fast music. In this paper, we designed the module which measures the number of steps from a user of wearable smart shoes based on the Arduino, and implemented the module so that the number of steps can be measured correctly by sending data to smart phone based on Android. And it is possible to measure moving distance and pace speed by utilizing a GPS in order to get the more accurate momentum. Also, we can measure more accurate calorie consumption than existing products by measuring the mean value of the calorie consumption for moving distance and the calorie consumption for the number of steps, give motivation of exercise by applying an algorithm of changing music genre according to pace speed, and increase the exercise effect at the same time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1624-1630
• /
• 2019

With the increase in there are smart devices penetration around the world, services related to exercise checks are attracting attention. However, there is existing exercise amount measurement service does not use the altitude information, or because the use of an algorithm that does not corrected the GPS altitude error is not accurate movement distance provided have a problem. Therefore, in this paper, to improve the existing problems, Kalman-filter-based user movement distance measurement algorithm is designed and implementation of improved by using the Kalman-filter based GPS and barometric altimeter sensor fusion algorithm to improve the altitude value the accuracy and of calculate the coordinate plane distance. As a result of comparing the designed and implementation of algorithm with the existing algorithms, it is confirmed that the proposed algorithm improves the accuracy by about 2.17%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.283-285
• /
• 2012

In this research, we produced a simple momentum measurement system which figures out information for the real-time travel distance with modularizing a momentum measurement system by using the acceleration sensor, MCU and Bluetooth to measure the exact momentum. In the simulation experiments, we compared the reliability and accuracy of the new momentum measurement system with those of existing momentum measurement systems to confirm that the new momentum measurement system shows superior reliability and accuracy. Base on this results, we will produce the enhanced momentum measurement system, compact and possible to shoe-mount, in future.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.489-490
• /
• 2023

본 논문에서는 신체 영상 인식 기술을 이용한 운동 자세 교정 시스템을 제안하고 개발하였다. 구글에서 제공하는 미디어파이프 포즈(MediaPipe Pose) 오픈소스를 사용하여 웹캠으로 사용자의 운동 동작을 실시간으로 인식하여, 인식된 신체 구조의 33개의 관절 위치로 Pose Landmark를 사용하여 사용자의 운동 자세에 대한 횟수 카운트, 운동 동작의 정확도 측정을 할 수 있게 하여 혼자 운동하거나 처음 운동하는 사람들에게 운동의 접근성을 높이고, 올바른 자세로 운동을 하도록 유도할 수 있다.