• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.249-251
• /
• 2021

In this paper, we introduce the deep-learning system using Tensorflow for recognizing situations that can occur fall situations when the elderly are moving or standing. Fall detection uses the LSTM (long short-term memory) learned using Tensorflow to determine whether it is a fall or not by data measured from wearable accelerator sensor. Learning is carried out for each of the 7 behavioral patterns consisting of 4 types of activity of daily living (ADL) and 3 types of fall. The learning was conducted using the 3-axis acceleration sensor data. As a result of the test, it was found to be compliant except for the GDSVM(Gravity Differential SVM), and it is expected that better results can be expected if the data is mixed and learned.

• Investigative Magnetic Resonance Imaging
• /
• v.10 no.2
• /
• pp.70-80
• /
• 2006

Spin echo gradient moment nulling pulse sequences were designed and implemented on a clinical magnetic resonance imaging system. A new technique was introduced for flow compensation that minimized echo time and effectively suppresses unwanted echoes on the slice selection gradient axis in spin echo sequences. A unified gradient shape was used in all orders of flow compensation up to the third order. A dual-purpose gradient was applied for flow compensation and to reduce unwanted artifacts. The sequences were used to generate images of phantoms and/or human brains. This technique was especially good at reducing eddy currents and artifacts related to imperfection of the refocusing pulse. The developed sequences were found to have shorter echo times and better flow compensation in through-plane flow than those of the previous models that were used by other investigators.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.12 no.3
• /
• pp.221-233
• /
• 2007

In this paper, a simple but effective position profile generation algorithm for single axis high dynamic linear machine drive system is presented. In the recent industrial application fields like as LCD/PDP and semiconductor factory, requirements for the high performance positioning system with optimal position profile generator are highly increased to reduce the overall processing time. There might be various solutions for position profile generating algorithm according to the application type. A square-wave Impact quantity(Jerk) based algorithm with minimal locomotion time is argued in this paper to minimize the total time of one movement under some specific constrains such as maximum speed limit and maximum acceleration limit. In order to reduce the calculation efforts and satisfy the minimal locomotion time condition, the time variants representing each profile sector and a simple condition comparison algorithm are adopted. Also, the actual implementation method for profile generation algorithm and it's real performance results are presented through commercial linear machine drive system.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.4
• /
• pp.510-516
• /
• 2019

In this paper, a snowboard simulator that measures the user's motion and makes the user feel physical changes and enjoy actual snowboarding was developed. The speed and direction of the snowboard are determined by the user's center of gravity. The developed simulator is equipped with four springs on the snowboard plate, so that the slope can change according to the change in the user's weight center and be felt directly. The slope due to the change in the center of gravity of the user is measured using a three-axis acceleration sensor. The friction of the slope generated by the rotation of the snowboard is made possible by the user using the BLDC motor, and the rotation of the snowboard is measured using the hole sensor. For rapid data processing of the simulator, two MCUs are used to transfer the measured data to the PC using the acceleration sensor and motor separately. The developed simulator can experience slopes and friction of the slope directly, and wear measured data and HMD to enjoy more realistic snowboarding.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.3
• /
• pp.585-590
• /
• 2012

In this paper, a system using vibrating wire sensor and zigbee wireless network has been implemented to monitor and manage the structure. The implemented strain controller drives vibrating wire sensor and computes strain from measuring frequency of the output signal. Temperature sensor is included to compensate strain by temperature. Using two axis acceleration sensor of strain controller can measure the direction of strain or deformation. To measure strain more effectively, wired and wireless communication function is included in this device. As results of experiments, it is shown that the developed system can be effectively applied to measure strain of the structure.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.1
• /
• pp.32-38
• /
• 2002

This paper describes a development of a portable physical activity monitoring system using two accelerometers to quantify physical activity. The system hardware consists of two piezoresistive accelerometers, amplifiers with gain of 30, lowpass filters with cut-off frequency of 15Hz, offset control circuits, one-chip microcontroller and flash memory card. In order to evaluate the performance of the system we acquired 3 channel data at 32 sample/sec from body-fixed accelerometers in chest and right upper leg. And then the acquired data were processed by MatLab on personal computer. We tried to distinguish not only fundamental actions which are steady-state activities such as standing, sitting, and lying but also dynamic activities with walking, up a stairway, down a stairway, and running. Five subjects participated the evaluation process which compare the video data with the measured data. As a result, the activity classification rate of 90.6% on average was obtained. Overall results showed that the steady-state activities could be classified from the low component of 3-axis acceleration signal and dynamic activities could be distinguished from frequency analysis using wavelet transform and FFT. Finally, we could find that this system can be applied to acquire and analyze the static and dynamic physical activity data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.2
• /
• pp.237-244
• /
• 2016

There are two destinations to aware the risk of common life. Recognition of the condition of pedestrian's own and the environmental factor awareness both are beneficial for risk awareness. It is good way of advancing the crime prevention effectivity that including IoT technology at the crime prevention research. The purpose of this research is that advanced way of crime prevention with multi-sensor data fusion of the condition of pedestrian and environmental factors. The 3-axis acceleration sensor is available to recognize the gait and the illumination sensor also useful to infer the road state. This research suggest a novel way of assess these factors and the result is the degree of danger.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.7
• /
• pp.1533-1540
• /
• 2012

u-Life care device collect body bio formation, and classify and store them in exercise patterns. Afterwards, the devices send the data through bluetooth wireless communication to the smart phones which set Google Android operation system at regular intervals. The information is checked out through application. u-Life care device calculates calories spent a day after monitoring activity quantity with 3-axis acceleration sensor. The device judges the status of health through body data mining and consults tailored exercise treatment. When sending body data, the device sends them in smart phone through Blue Tooth wireless communication at once. So, as a strong point, the device doesn't need mobile gateway or home gateway used for sending to web server information sensed from exercise life care products.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.462-469
• /
• 2014

As a basic study on both engineering applications and representation methods of synesthesia, this paper aims at building basic system which converts a muscular sense into both visual and auditory elements. As for the building method, data of the muscular sense can be acquired through roll and pitch signals which are calculated from both three-axis acceleration sensor and the two-axis gyro sensor. The roll and pitch signals are then converted into both visual and auditory information as outputs. The roll signals are converted into both intensity elements of the HSI color model and octaves as one of auditory elements. In addition, the pitch signals are converted into both hue elements of the HSI color model and scales as another one of auditory elements. Each of the extracted elements of the HSI color model is converted into each of the three elements of the RGB color model respectively, so that the real-time output color signals can be obtained. Octaves and scales are also converted and synthesized into MIDI signals, so that the real-time sound signals can be obtained as anther one of output signals. In experiments, the results revealed that normal color and sound output signals were successfully obtained from roll and pitch values that represent muscular senses or physical movements, depending on the conversion relationship based on the similarity between color and sound.