• Journal of Biosystems Engineering
• /
• 제27권4호
• /
• pp.341-348
• /
• 2002

Quality evaluation of red ginsengs is determined by outer shape and inner qualities. Especially, the inner qualities are main grading criteria. Currently, red ginsengs are classified into 3-grades; heaven, earth and good. The best heaven grade must not include inner holes and sponge tissues. This study was conducted to develop a red ginseng sorting system using x-ray image processing technique. Because of lens characteristic, gray values of the central region in the x-ray image are higher and gradually decreased towards the edge regions. This difference of gray values gives trouble in segmentation and detection of inner holes in red ginseng image, so preprocessing technique is necessary. The preprocessing was done by subtracting source image from an empty background image. But, simple subtraction was not quite appropriate because of too small contrast between inner holes and sound part. Scaled subtraction images were obtained by multiplying all gray values by some numbers. However this method could not help to set threshold value because the gray values of root part are generally lower than body part when red ginseng is exposed to the x-ray. To determine threshold value for detecting inner holes, an algorithm was developed by increasing overall gray values of less clear images.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2011년도 제40회 동계학술대회 초록집
• /
• pp.472-472
• /
• 2011

The capacitive coupled plasma (CCP) has been extensively used in the semiconductor industry because it has not only good uniformity, but also low electron temperature. But CCP source has some problems, such as difficulty in varying the ion bombardment energy separately, low plasma density, and high processing pressure, etc. In this reason, dual frequency CCP has been investigated with a separate substrate biasing to control the plasma parameters and to obtain high etch rate with high etch selectivity. Especially, in this study, we studied on the etching of $SiO_2$ by using the pulse-time modulation in the dual-frequency CCP source composed of 60 MHz/ 2 MHz rf power. By using the combination of high /low rf powers, the differences in the gas dissociation, plasma density, and etch characteristics were investigated. Also, as the size of the semiconductor device is decreased to nano-scale, the etching of contact hole which has nano-scale higher aspect ratio is required. For the nano-scale contact hole etching by using continuous plasma, several etch problems such as bowing, sidewall taper, twist, mask faceting, erosion, distortions etc. occurs. To resolve these problems, etching in low process pressure, more sidewall passivation by using fluorocarbon-based plasma with high carbon ratio, low temperature processing, charge effect breaking, power modulation are needed. Therefore, in this study, to resolve these problems, we used the pulse-time modulated dual-frequency CCP system. Pulse plasma is generated by periodical turning the RF power On and Off state. We measured the etch rate, etch selectivity and etch profile by using a step profilometer and SEM. Also the X-ray photoelectron spectroscopic analysis on the surfaces etched by different duty ratio conditions correlate with the results above.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1999년도 가을 학술발표논문집 Vol.26 No.2 (2)
• /
• pp.377-379
• /
• 1999

본 연구에서는 축구 경기 장면 분석시스템의 일부인 경기 장면에서 선수의 위치와 운동장 모델에서의 선수의 위치 간의 매핑에 관해 기술한다. 이 때 두 좌표계 간의 사상(mapping)시 발생되는 hole에 관한 문제를 주로 사용되는 보간법이 아니라 역 방향 매핑을 이용하여 해결하고, 또한 운동장 모델의 전체 영역이 아니라 원 영상에서 얻어진 정보를 이용하여 최소한의 영역만을 변환함으로써 계산의 효율성을 높였다. 본 연구에서는 PC상에서 초당 3~5 Frame로 입력된 280$\times$640 해상도의 100여개의 RGB 영상을 대상하여 실험하였다. 축구 경기장면을 분석한 결과 대부분의 경우 두 좌표계간의 매핑이 올바로 이루어 졌다.

• 한국정밀공학회지
• /
• 제23권11호
• /
• pp.58-67
• /
• 2006

Due to the variety of signal processing and complicated mathematical analysis, it is not easy to accomplish 3D bin-picking with non-contact sensor. To solve this difficulties the reliable signal processing algorithm and a good sensing device has been recommended. In this research, 3D laser scanner and CCD camera is applied as a sensing device respectively. With these sensor we develop a two-step bin-picking method and reliable algorithm for the recognition of 3D bin object. In the proposed bin-picking, the problem is reduced to 2D intial recognition with CCD camera at first, and then 3D pose detection with a laser scanner. To get a good movement in the robot base frame, the hand eye calibration between robot's end effector and sensing device should be also carried out. In this paper, we examine auto-calibration technique in the sensor calibration step. A new thinning algorithm and constrained hough transform is also studied for the robustness in the real environment usage. From the experimental results, we could see the robust bin-picking operation under the non-aligned 3D hole object.

• 비파괴검사학회지
• /
• 제13권2호
• /
• pp.11-19
• /
• 1993

원자력발전소의 주요 압력용기 용접부에 대한 초음파검사시 결함의 특성과 형태에 대한 정확한 분류는 원자력 발전소의 안전성을 확보하기 위한 결함평가에 중요한 요소이다. 본 연구에서 초음파검사에서 얻어진 결함신호를 digital signal processing 기법으로 처리하여 결함의 특성과 형태를 구분할 수 있는 feature vector를 추출하고 결함의 형태를 형상 인식법을 사용하여 분류 하였다. Training specimen(slit, hole)의 신호와 testing specimen(crack, slag)의 신호를 구분하기 위한 실험에서 사용된 통계적 pattern recognition algorithm은 minimum distance classifier와 maximum likelihood classifier이다. 이러한 형상 classifier를 이용하여 결함의 특성을 정량적으로 분류하여 결함 평가 능력을 향상시켰다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.1021-1024
• /
• 2001

Out of all metal-cutting processes, the hole-making process is the most widely used. It is estimated to be more than 30% of the total metal-cutting process. It is therefore desirable to monitor and detect drill wear during the hole-drilling process. In this paper, the vision system of the sensing methods of drill flank wear on the basis of image processing is used to detect the wear pattern by non-contact and direct method and get the reliable wear information about drill. In image processing of acquired image, median filter is applied for noise removal. The vision flank wear area of the drill was measured. Backpropagation neural networks (BPns) were used for no-line detection of drill wear. The neural network consisted of three layers: input, hidden and output. The input vectors comprised of spindle rotational speed, feed rates, vision flank wear, thrust and torque signals. The output was the drill wear state which was either usable or failure. Drilling experiments with various spindle rotational speed and feed rates were carried out. The learning process was peformed effectively by utilizing backpropagation. The detection of the abnormal states using BPNs achieved 96.4% reliability even when the spindle rotational speed and feedrate were changed.

• 한국자동차공학회논문집
• /
• 제20권5호
• /
• pp.44-52
• /
• 2012

Fuel cell electric vehicles have some noise problems due to its air processing unit which is required to feed the ambient air into the fuel cell stack. Discrete-frequency noises are radiated from a centrifugal blower due to rotor-stator interaction. Their fundamental frequency is the blade passing frequency, which is determined by the number of rotor blades and their rotating speed. To reduce such noises, multi-chamber perforated muffler has been designed. In this paper, in order to improve the transmission loss of a perforated muffler, the relationship between the impedance model of a perforated hole and its noise reduction performance is studied, and the applicability of a short-length perforated muffler to air processing unit of fuel cell system is described using acoustic simulation results and experimental data. The acoustic velocity vector across the neck of a perforated hole is very important design factor to optimize the transmission of an intake muffler. The suggested short-length perforated muffler is effective on discrete-frequency noises while keeping the volume of intake muffler minimized.

• 한국정보처리학회논문지
• /
• 제6권4호
• /
• pp.1090-1097
• /
• 1999

열악한 환경엣 동작되는 멀티컴퓨터는 요소결함(component faults)이 존재하는 상황에서도 정상적 동작을 보장할 수 있도록 설계되어야 한다. 이를 위한 하나의 방법으로 결함포용 라우팅(fault-tolerant routing) 기법이 고려될 수 있다. 본 논문에서는 n-차원 메쉬 네트워크를 기본 토폴로지로 선택하여 이러한 네트워크의 임의의 장소에서 링크결함이 발생했을 경우에도 메시지들을 목적지로 전달시킬 수 있는 결함포용 라우팅 알고리즘을 제안한다. 제안된 결함포용 라우팅 알고리즘은 기본적으로 WH(WormHole)라우팅 방식을 채택하며, deadlock-free를 실현하기 위하여 한 개의 물리적 채널을 공유하는 복수 개의 가상채널들(virtual channels)을 사용한다. 결론적으로 컴퓨터 시뮬레이션을 통해 제안된 알고리즘이 널리 알려진 X-Y 라우팅 알고리즘보다 향상된 성능을 갖는다는 사실을 입증한다.

• 한국멀티미디어학회논문지
• /
• 제18권9호
• /
• pp.997-1007
• /
• 2015

In this paper, we propose a method for automatically analyzing the bone formation in a mouse model of frontal bone defect. We perforate two holes of 0.8mm diameter in the frontal bone and observe the bone formation process using a micro CT. Because the conventional analysis software of the micro CT does not support automatic analysis of the bone formation status, we have to use a manual analysis method. However the manual analysis is very cumbersome and requires a lot of time, we propose an automatic analysis method. It rotates the image around three axes directions so that the mouse's skull come into regular position. It calculates the cumulative image of the voxel values for the perforated bone surface. It estimates the hole location by finding the darkest point in the cumulative image. The proposed method was applied to 24 CT images of saline administration group and PTH administration group and hole location was estimated. BV/TV index was calculated for the estimated hole to evaluate the bone formation status. Experimental results showed that bone formation process is more active in PTH administration group. The method proposed in this paper could replace successfully the cumbersome and time consuming manual job.

• 한국분무공학회지
• /
• 제4권1호
• /
• pp.45-54
• /
• 1999

The spray characteristics of a direct injection multi-hole diesel nozzle having the 2-spring nozzle holder were investigated by using the image processing system and a PDPA(phase Bowler particle analyzer) system. The spray tip penetration, the spray angle, and the droplet diameter and velocity with the variation of the pump speed, injection quantity were measured. From, the experiments, we know that there are small droplets which are not to be detected with spray image around the leading edge of the spray. In order to represent the mean characteristics of the intermittent spray very well, it is very important to set the time windows accurately. From the measurements along the axis of the spray, close to the nozzle, the initially injected droplets are overtaken by droplets that follow them. And also there are the maximum axial mean velocity and SMD at the following part of the leading edge of the spray.