• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.1
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.8
• /
• pp.39-47
• /
• 1999

This paper describes on-line retrainable RBF network in order to control the coiling temperature for a hot coil strip at Pohang Iron & Steel Company(POSCO). The proposed neural network can be used for improving conventional rule-based lookup table, which generates a heat transmission coefficient. To cope with time-varying characteristics of hot coil process, additional synaptic weights for on-line retraining purposes are introduced to hidden-to-output weights of conventional RBF network. Those weights are locally adjusted to newly incoming test data while preserving old information trained with off-line past data. Hence the effect of catastrophic interference can be greatly alleviated with the proposed network. In addition, rejection scheme is introduced for reliability concerns. From the experimental results applied to the actual process, it is noticed that overall control performance represents about 2.2% increase compared to the conventional one.

• Journal of Biomedical Engineering Research
• /
• v.30 no.6
• /
• pp.503-509
• /
• 2009

The development of technology has led to ubiquitous health care service, which enables many patients to receive medical services anytime and anywhere. For the ubiquitous health care environment, real-time measurement of biomedical signals is very important, and the medical instruments must be small and portable or wearable. So, such devices have been developed to measure biomedical signals. In this study, we develop the biomedical monitoring device which is sensing the PPG signal, one of the useful signal in the field of ubiquitous healthcare. We design a watch-like biomedical signal monitoring system without a finger probe to prevent the user's inconvenience. This system obtains the PPG from the radial artery using a sensor in the wrist band. But, new device developed in this paper is easy to get the motion artifacts. So, we proposed new algorithm removing the motion artifacts from the PPG signal. The method detects motion artifacts by changing the degree of brightness of the light source. If the brightness of the light source is reduced, the PPG pulses will disappear. When the PPG pulses have disappeared completely, the remaining signal is not the signal that results from the changing blood flow. We believe that this signal is the motion artifact and call it the noise reference signal. The motion artifacts are removed by subtracting the noise reference signal from the input signal. We apply this algorithm to the system, so we can stabilize the biomedical monitoring system we designed.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.19 no.6
• /
• pp.1673-1675
• /
• 2005

The pulse diagnosis is an important and universal method in Oriental medicine. Nevertheless, because of characteristic that depends on subjective sense of Oriental medicine doctor (OMD), it is not recognized by objective basis. The Korean Institute of Oriental Medicine(KIOM) and Daeyo Medi. Co. Ltd. developed the 3-D Mac using arrey piezoresistive sensors and multi-axial robot. 133 healthy subjects participated in this study, 75 males and 58 females, between 20 and 70 years of age. All subjects were relaxed in a supine position on a comfortable chair for twenty minutes before the measurement was taken. The measured position is the radial artery of subject's left wrist and the position is called Chon, Kwan and Chuck in Oriental medicine. To detect floating and sinking pulse, we established coefficient of floating and sinking(CFS). CFS means relative position of maximum pulse pressure in PH curve. The lower CFS value means that the pulse has floating tendency. There was significant diffence between CFS and diagnosis of floating-sinking pulse by OMD(p=0.020). CFS value of over 40's group was significantly larger than those of 20's and 30's(p=0.000). There was no significant difference between male and female(p=0.061).

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.2
• /
• pp.173-180
• /
• 2012

The most recent, 80 mega pixels digital camera appeared through the development of digital technology, and nonmetric digital cameras have been using in various field of photogrammetry. In this study, we experimented lens calibration using aerial photographs and ground control points. The aerial photographs were taken a non-metric digital camera which is CMOS(Complementary Metal Oxide Semiconductor) 21.1 mega pixels sensor and 35mm lens at a helicopter. And the ground control points were selected on the 1:1,000 plotting origin data. As a result, we calculated focal length, PPA(Principal Point of Autocollimation) and symmetric radial distortion coefficients from the lens. Also, RMSE(root mean square error) and maximum residual of the ground control points from the aerial triangulation were compared before and after calibration. And we found that the accuracy of the after calibration was improved very significantly.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.19 no.3
• /
• pp.29-37
• /
• 2011

The modal characteristics of a compressor wheel of an automotive turbocharger have been investigated using an experimental method based on an acoustic frequency response function, p/f(${\omega}$), where p is sound pressure radiated from a structure, and f is impact force. First, a well-defined annular disc with narrow radial slots was examined to check whether the vibro-acoustic test could precisely determine natural quencies and vibration modes of structures showing that the vibro-acoustic test proposed in this paper was comparable to the conventional modal test with an accelerometer and the numerical analysis. The conventional method has been found to be inappropriate for compressor wheel because of additional mass due to the accelerometer and additional damping from the accelerometer cable alter the dynamic responses of the wheel blades. odal characteristics of the wheel have been defined using vibro-acoustic test and verified with the results from another conventional method using a laser vibrometer. Natural quencies and mode shapes of a turbocharger wheel, which can't be precisely obtained with onventional method, could be defined accurately without the additional effects from sensor and cable. Proposed method can be applied to small structures where conventional sensors and cables could generate troubles.

• Journal of Sensor Science and Technology
• /
• v.6 no.5
• /
• pp.392-399
• /
• 1997

In this study, the effect of the sintering temperature on the dielectric and piezoelectric properties of 0.5 wt% $MnO_{2}$-doped $0.125Pb(Mg_{1/2}Nb_{2/3})O_{3}-0.435PbTiO_{3}-0.44PbZrO_{3}$ ceramics were investigated aiming at ultrasonic motor applications. From experimental result, it was found that the optimal sintering temperature condition was at $1270^{\circ}C$. The sample sintered at $1270^{\circ}C$ had density of $7.72\;g/cm^{3}$, dielectric constant of 570, dielectric loss of 0.82%, remanent polarization of $19.18{\mu}C/cm^{2}$, coercive field of 9.63 kV/cm, electromechanical coupling factor of radial mode of 55.1%, mechanical quality factor of 886. Temperature and frequency dependence of dielectric constant and dielectric loss of the sintered sample at $1270^{\circ}C$ was also investigated.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.2
• /
• pp.187-193
• /
• 2015

In this study, regressive models were employed to predict the blood velocity by using independent variables which measured by 3 sensors, such as electrocardiogram (ECG), photo-plethysmogram (PPG), pulse diagnosis sensor, without ultrasound device at high cost. In experiment, the high predictable model was induced to estimate the blood velocity correctly by comparing correlation values and significance probabilities between independent variables and blood velocities. Results showed that the model induced by two or three independent variables had a higher predictability than those by a single independent variable.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 1999.10a
• /
• pp.52-57
• /
• 1999

Korea Earthquake Monitoring System(KEMS) in the Korea Institute of Geology Mining and Materials(KIGAM) as detected more than 1000 events since the end of 1998. But not all events are interpreted as earthquakes because many events are concentrated on daytime. It strongly implies that in addition to earthquake these events include artificial effects such as industrial blasting. Before the determination of eathquake charactertistics in the korean peninsula it is necessary to discriminate the detected events as earthquakes or artificial events. For the discriminant study KIGAM and SMU(Southern Methodist University) installed a triangular four-element 1-km aperture seismo-acoustic array at Chul-Won area northeast of Seoul Korea. Each array element includes a GS-13 seismometer in the bottom of borehole and a Validyne DP250-14 microbarometer sensor mounted inside of the borehole 1,2 meter deep connected to a 11 arm radial array of 10m porous soaker hoses. This array introduce the use of 2.4-GHz radios for inter-array self-contained solar-charged power system and GPS time-keeping system. A 24-bit digital data acquisition system performs 40 SPS in the infrasound and seismometer data. Velocity and direction of wind and temperature are also measured at hub site and included to the data stresam. This seismo-acoustic array will be used to identify and locate associated with industrial blasting and these identified and located events will be applied to form a ground truth database useful to assist the other development of discriminant studies.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.9
• /
• pp.925-932
• /
• 2014

If a rotor possesses a high gyroscopic coupling or the running speed is high, the dynamical changes in the rotor become prominent. When active magnetic bearings are used to support such rotors, it is necessary for the bearing controller to take these dynamical changes into consideration. Independent-axis controllers, which are the most commonly used, modulate the bearing force solely based on the sensor output of the same axis. However, this type of controller has difficulties in overcoming the dynamical changes. On the other hand, mixed-axis controllers transform the sensor output into components corresponding to the vibrational modes. A separate controller can then be designed for each vibrational mode. In this way, the controller can be designed based on the dynamics of the rotor. In this paper, we describe a design process for a mixed-axis controller that uses a detailed mathematical model of the system. The performance of the controller is evaluated based on the ISO sensitivity requirements and unbalance response, while considering the change in the system dynamics due to the running speed.