• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.579-580
• /
• 2023

본 논문에서는 동물의 십자인대 질환의 진단지표인 경골고원각도(TPA)를 자동으로 측정하는 딥러닝 소프트웨어 기법을 제안한다. 동물 X-ray 영상에서 나타나는 피사체의 위치와 형태에 대한 다양한 변이는 TPA(Tibial Plateau Angle) 지표 산출에 필요한 특징점 검출과정에서 학습 효율을 현저하게 저하시킨다. 이에 본 연구에서는 YOLO(You Only Look Once) 기반 모델을 사용하여 일차적으로 경골영역의 분할 단계를 수행하고, 이어서 경골 상단부의 과간융기와 복사뼈의 중심점을 찾는 과정을 Resnet 기반의 특징점 추출 모듈로서 구현함으로써 학습의 효율과 지표 검출의 정확도를 향상시켰다. 총 201 개의 실제 X-ray 영상을 사용하여 학습 속도와 영역 분할 및 특징점 추출의 정확도 측면을 고려함으로 제안된 이론의 타당성을 실험적으로 평가하였다.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.13 no.1
• /
• pp.42-55
• /
• 2008

To evaluate the accuracy of currents measured by HF radar in the coastal sea off Keum River estuary, we compared the facing radial vectors of two HF radars, and HF radar-derived currents with in-situ measurement currents. Principal component analysis was used to extract regression line and RMS deviation in the comparison. When two facing radar's radial vectors at the mid-point of baseline are compared, RMS deviation is 4.4 cm/s in winter and 5.4 cm/s in summer. When GDOP(Geometric Dilution of Precision) effect is corrected from the RMS deviations that is analyzed from the comparison between HF radar-derived and current-metermeasured currents, the error of velocity combined by HF radar-derived current is less than 5.1 cm/s in the stations having moderate GDOP values. These two results obtained from different method suggest that the lower limit of HF radar-derived current's accuracy is 5.4 cm/s in our study area. As mentioned in previous researches, RMS deviations become large in the stations located near the islands and increase as a function of mean distance from the radar site due to decrease of signal-to-noise level and the intersect angle of radial vectors. We found that an uncertain error bound of HF radar-derived current can be produced from the separation process of RMS deviations using GDOP value if GDOP value for each component is very close and RMS deviations obtained from current component comparison are also close. When the current measured in the stations having moderate GDOP values is separated into tidal and subtidal current, characteristics of tidal current ellipses analyzed from HF radar-derived current show a good agreement with those from current-meter-measured current, and time variation of subtidal current showed a response reflecting physical process driven by wind and density field.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.12 s.354
• /
• pp.55-64
• /
• 2006

This work proposes a calibration-free 14b 70MS/s 0.13um CMOS ADC for high-performance integrated systems such as WLAN and high-definition video systems simultaneously requiring high resolution, low power, and small size at high speed. The proposed ADC employs signal insensitive 3-D fully symmetric layout techniques in two MDACs for high matching accuracy without any calibration. A three-stage pipeline architecture minimizes power consumption and chip area at the target resolution and sampling rate. The input SHA with a controlled trans-conductance ratio of two amplifier stages simultaneously achieves high gain and high phase margin with gate-bootstrapped sampling switches for 14b input accuracy at the Nyquist frequency. A back-end sub-ranging flash ADC with open-loop offset cancellation and interpolation achieves 6b accuracy at 70MS/s. Low-noise current and voltage references are employed on chip with optional off-chip reference voltages. The prototype ADC implemented in a 0.13um CMOS is based on a 0.35um minimum channel length for 2.5V applications. The measured DNL and INL are within 0.65LSB and l.80LSB, respectively. The prototype ADC shows maximum SNDR and SFDR of 66dB and 81dB and a power consumption of 235mW at 70MS/s. The active die area is $3.3mm^2$.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.23 no.5
• /
• pp.439-444
• /
• 2003

Acoustic emission(AE) ran be very effectively applied to locate the damaged area in large structures by detecting the elastic waves generated during the damage process within solids. Source location in the composite structures has been, however, extremely difficult due to the acoustic anisotropy with the velocity dependence on fiber orientations. In this study, it has been shown that a newly proposed method for 2-D source location of anisotropic structures is practically applicable to the real structure. The method employes wave velocities obtained with different velocities from $0^{\circ}\;to\;90^{\circ}$ for a filament wound composite pressure vessel under the air-filled and the water-filled conditions.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.6
• /
• pp.254-261
• /
• 2021

This study was intended to develop an inspection method to detect defects in closed-cell steel members in steel girder bridges. The ultrasonic pulse velocity method was selected as a rapid and effective method to identify thickness changes of steel specimens caused by corrsion. This study developed an algorithm to expedite the process and improve the accuracy in the prediction of steel plate thickness. Also, both static and continuous scanning methods were compared to each other to identify the difference in accuracy, but the results revealed that both methods produce almost the same results. This study also provided the idea to calculate the height of water contained in the closed-cell steel member and results of laboratory experimental results. The water heights which is thicker than the steel plate thickness were detectable and predicted using the idea suggested by this study, but the water heights lower than the steel plate thickness were not possible. However, the results showed whether the steel member contains water or not.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.12
• /
• pp.133-142
• /
• 2021

Travel pattern recognition is widely used in many aspects such as user trajectory query, user behavior prediction, interest recommendation based on user location, user privacy protection and municipal transportation planning. Because the current recognition accuracy cannot meet the application requirements, the study of travel pattern recognition is the focus of trajectory data research. With the popularization of GPS navigation technology and intelligent mobile devices, a large amount of user mobile data information can be obtained from it, and many meaningful researches can be carried out based on this information. In the current travel pattern research method, the feature extraction of trajectory is limited to the basic attributes of trajectory (speed, angle, acceleration, etc.). In this paper, permutation entropy was used as an eigenvalue of trajectory to participate in the research of trajectory classification, and also used as an attribute to measure the complexity of time series. Velocity permutation entropy and angle permutation entropy were used as characteristics of trajectory to participate in the classification of travel patterns, and the accuracy of attribute classification based on permutation entropy used in this paper reached 81.47%.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.11
• /
• pp.7941-7948
• /
• 2015

Rolling noises during train operation are caused by vibration excited from irregularities of surface roughness between wheel and rail. Therefore, a proper measurement and analysis techniques of acoustic roughness between wheel and rail surface are required for transmission, prediction, and analysis of the train rolling noise. However, since current measuring devices and methods use trolley-based manual handling devices, the measurements induce unstable measuring speed and vibrational interface that increases errors and disturbances. In this paper, a new automatic rail surface exploring platform with a speed controller has been developed for improving measurement accuracy and reducing inconsistency of measurements. In addition, we propose a data integration method of the rail surface roughness with multiple homogeneous displacement sensors and verified the accuracy of the integrated data through standard test-bed railway track investigation.

• Aerospace Engineering and Technology
• /
• v.9 no.2
• /
• pp.36-43
• /
• 2010

A GEO satellite launched by Arian 5 ECA launcher will be located in a transfer orbit where it requires several Apogee burn maneuvers to reach the target orbit. To obtain the required performance of Apogee burn maneuvers, a calibration of gyro drift error needs to be performed before each maneuver. In this paper, a unique gyro calibration scheme which is applied to COMS is described and the calibration error budget analysis is performed.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.665-666
• /
• 2023

미세 물질을 포함하고 있는 광산 폐기물의 디워터링 과정(dewatering process)은 작은 입자들의 침전속도가 낮기 때문에 시간이 오래 걸리고 어려운 과정이다. 따라서 광산 폐기물이 안정적으로 침전되었는지 확인하기 위해서 디워터링 과정을 연속적으로 모니터링하는 기술이 필요하다. 이 연구에서는 kaolinite, illite, bentonite 3 종류의 점토를 3 가지 농도(0.1g/L, 1g/L, 5g/L)로 나눠서 초음파 음향 감지를 이용해 후방산란 신호를 측정했다. 그리고 측정된 신호를 활용하여 합성곱 신경망(CNN) 모델을 개발하여 점토의 분류 모델을 만드는 연구를 수행했다. 본 연구에서 보여준 CNN 의 높은 정확도는 광산 폐기물의 디워터링 과정에서 미세 광물과 미세 농도 분류 모니터링에 적합한 저렴하고 측정하기 쉬운 음향 감지의 사용 가능성을 입증했다.

• Korean Journal of Applied Biomechanics
• /
• v.19 no.1
• /
• pp.179-186
• /
• 2009

This works studied to compare gas analyzer with accelerometer and the estimation of energy expenditure based on different attaching position of tri-axial accelerometer such as waist and top of the foot Based on the fact that oxygen intake increases more radically linearly during walking more than 8.0km/hr. 9 male subjects performed walking and running on the treadmill with speed of $1.5{\sim}8.5km$/hr and $4.5{\sim}13.0km$/hr, respectively. Commercially available Nike + iPod Sports kit was used to compare energy expenditure with sensor module attached to their foot. Actual energy expenditure was determined by a continuous direct gas analyzer and two multiple regression equations of walking and running mode for different attaching position were developed. Results showed that estimation accuracy of energy expenditure using waist mounted accelerometer was higher than that of the top of the foot and Nike + iPod Sports kit. Results of energy expenditure based on waist and top of the foot showed that the crossover state of energy expenditure occurred at 7.5km/hr. But Nike + iPod Sports kit could not find intersection of energy expenditure in all nine subjects. Therefore the sensor module attached to the waist and separate multi regression equation by walking and running mode was the best to estimate more accurate prediction.