• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1413-1416
• /
• 2005

An infrared temperature sensor module developed for the detection of defects in a plate was modified to use in a cylinder. A set of optical fiber leads and a mechanism maintaining sensor-object distance constant were utilized for the modification of the IR sensor module. The detection performance was experimentally investigated, and the measured temperature was also compared with computed temperature distribution. The experimental outcome indicates that the detection of a simulated defect is readily available. The temperature distribution is better for defect detection than that with the previous device. In addition, the measured distribution is comparable to the calculated one using a heat conduction equation. The developed device of defect detection is suitable to be utilized in chemical processes where most of vessels and piping systems are in the shape of a cylinder.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2007년도 추계학술대회 논문집
• /
• pp.913-916
• /
• 2007

In this paper, the possibility of time reversal phenomenon was investigated in damage detection of structure. In conventional lamb wave techniques, damage is identified by comparing the measured data (baseline signals) and the current data. But this method can lead to high false signal in the intact condition of structures due to environmental conditions of the structures. So in this studying, we investigate the possibility of damage detection in the aluminum plate using the time reversal phenomenon of lamb waves.

• 한국방재안전학회논문집
• /
• 제11권2호
• /
• pp.53-60
• /
• 2018

본 연구에서는 강판에 발생한 국부적인 손상의 진단을 위해 강자성의 연속체 구조물에 적합한 비파괴진단 기법인 누설자속 기법을 적용하였다. 강판시편의 두께 변화에 따른 누설자속 기반 진단 기법의 민감도를 분석하기 위해 각각 다른 두께를 가지는 몇 가지의 강판시편을 준비하였고, 각각의 시편에 3가지 깊이의 인공결함을 가공하였다. 홀센서와 자화요크를 이용하여 일정한 자화밀도를 가지는 누설자속 센서헤드를 제작하여 강판시편으로부터 누설자속 신호를 계측하였다. 센서헤드로부터 수집된 자속신호의 품질향상을 위해 일련의 신호처리 과정을 거쳤으며, 각 손상지점으로부터 계측된 누설자속 신호의 확인을 통해 국부손상 감지의 가능성을 확인하였다. 강판두께에 따른 누설자속 기법의 손상감지 민감도의 분석을 위해 각각 다른 두께의 강판시편으로부터 검출된 MFL 신호에서 P-P value를 정량적으로 추출하였고 그 값을 비교 분석 하였다.

• 전기학회논문지
• /
• 제65권9호
• /
• pp.1531-1540
• /
• 2016

In this paper, we propose vehicle license plate recognition system based on Radial Basis Function Neural Networks (RBFNNs) with the use of local binarization functions and canny edge algorithm. In order to detect the area of license plate and also recognize license plate numbers, binary images are generated by using local binarization methods, which consider local brightness, and canny edge detection. The generated binary images provide information related to the size and the position of license plate. Additionally, image warping is used to compensate the distortion of images obtained from the side. After extracting license plate numbers, the dimensionality of number images is reduced through Principal Component Analysis (PCA) and is used as input variables to RBFNNs. Particle Swarm Optimization (PSO) algorithm is used to optimize a number of essential parameters needed to improve the accuracy of RBFNNs. Those optimized parameters include the number of clusters and the fuzzification coefficient used in the FCM algorithm, and the orders of polynomial of networks. Image data sets are obtained by changing the distance between stationary vehicle and camera and then used to evaluate the performance of the proposed system.

• 제어로봇시스템학회논문지
• /
• 제20권7호
• /
• pp.718-723
• /
• 2014

In this paper, we developed the inspection system of electric vehicle battery plate using image processing. Four cameras are used for acquiring the principal parts of the plate, and several steps of image processing for extracting significant dimensions of the plate such as widths and lengths. As a preceding step, calibration of four cameras is carried for compensating distorted images using dot-arrayed sheet. Coordinate systems for four cameras are defined where one coordinate system is assigned to the reference coordinate system to which the others are relatively described. Line information of the edge in the windowed image is extracted using elaborate edge-detection algorithm, and finally the intersection points between lines are extracted to calculate widths and lengths of the plate from which the error status of the battery plate is decided.

• 대한토목학회논문집
• /
• 제26권4D호
• /
• pp.549-551
• /
• 2006

차량번호판인식시스템은 무인 카메라 등의 영상장치를 통하여 입력된 차량 이미지로부터, 차량번호판 정보를 읽어내는 시스템이다. 이러한 차량번호판인식시스템의 응용 시스템 중 과속차량 단속과 같은 일부 응용 시스템은 번호판의 글자나 숫자를 다른 글자나 숫자로 잘못 인식할 경우 심각한 문제를 발생시킬 수 있다. 이러한 문제를 피하기 위해 우리는 인식 결과에 대한 신뢰도가 낮은 경우 인식을 포기 또는 위임하는 '신중한 분류기(Cautious Classifier)'를 이용하여 인식시스템을 구성하였다. 또한 학습 예제의 가중치를 조정하는 방법을 사용하여 이러한 신중한 분류기의 성능을 향상시켰다.

• 한국환경보건학회지
• /
• 제36권2호
• /
• pp.141-147
• /
• 2010

Indicator bacteria of fecal pollution were enumerated and compared by various detection methods for influent and final effluent of a sewage treatment plant. Total coliforms were enumerated by four methods including most probable numbers, chromogenic enzyme substrate test, membrane filtration, and plate counts and were about $10^4$ for influent and $10^2{\sim}10^3\;CFU/ml$ for final effluent. Fecal coliforms ranged between $10^3$ and $10^4$ for influent and $10^2\;CFU/ml$ for effluent by chromogenic enzyme substrate test and membrane filtration. Fecal streptococci counts were 1-log less than fecal coliforms counts, $10^2{\sim}10^3$ for influent and $10^1\;CFU/ml$ for effluent. Total coliforms numbers by plate count both in influent and in effluent showed 1-log higher than by the other three methods. Statistical analysis revealed that numbers of total coliforms by plate count in final effluent had the highest average of correlation (r=0.778, p<0.01) compared with those by the other three methods. In addition, total coliforms numbers by plate count showed most significant correlation (r=0.835, p<0.01) with those by chromogenic test which is well-known as its highest recovery efficiency. These results suggest that the plate count would be the optimum detection method for total coliforms in wastewater treatment plants which are the only microbiological standard of final effluent from wastewater treatment plants in the Republic of Korea, considering economic aspects and difficulties in laboratories.

• 한국인터넷방송통신학회논문지
• /
• 제13권1호
• /
• pp.55-62
• /
• 2013

동판 생산량의 증가와 수요의 활성화로 더욱 동판에 대한 자동 검사 시스템이 필요하게 되었다. 본 연구에서는 동판의 3차원 표면형상 및 결함 검출을 하기 위한 3차원 영상 분석 시스템 및 GUI를 개발했다. 2차원 영상을 통해 분석을 할 수 있으나 오류가 많이 발생하기 쉽고, 작업자가 분석하기에는 무리가 따르기 때문에 3차원 영상으로 분석하여 살펴보고 자동으로 판정을 내리므로 작업자가 사용하기 쉽다. 동판 제작 공정에서 발생되는 검사 방법에서 사람에 의한 육안 검사가 주로 행해지고 있는데, 여기서 자동 검사를 통해 정확한 검사율과 비용 발생을 감소를 할 수 있다. 동판에 대한 결함을 정의하고, 동판 결함 검사 측정을 위한 시스템을 개발한다. 그리고 분석 알고리즘과 3차원 영상 분석 프로그램을 개발하여 동판에 결함을 자동 검출한다.

• Smart Structures and Systems
• /
• 제11권1호
• /
• pp.35-51
• /
• 2013

Over the last two decades Wavelet Transformation (WT) method has been widely utilized for the damage identification of structures. The main objective of this paper is to discuss and present some of common shortcomings and limitations of mathematical software, as well as other precautionary measures that need to be considered when using them for wavelet analysis applications. Due to popular usage of MATLABMATLAB(R) comparing to other mathematical tools among researchers for data processing of structural responses through WT analysis, this software was chosen for specific study. To the best of the authors' knowledge, these limitations and observations have not been previously identified or discussed in the literature. In this work, a square plate with a severe damage, in form of a crack, parallel to the left edge of the plate is selected for a pilot study. The steady state harmonic response is used for measuring the deflection shape across the line parallel to one edge and perpendicular to the damage. Several criteria and cases such as the smallest size damage that can be detected, correlation between the crack width and the number of sampling points, and the influence of the damage thickness on the accuracy of the result are investigated.

• 전기학회논문지
• /
• 제65권1호
• /
• pp.142-149
• /
• 2016

This paper proposes a method to detect LP(License Plate) of vehicles in indoor and outdoor parking lots. In restricted environment, there are many conventional methods for detecting LP. But, it is difficult to detect LP in natural and complex scenes with background clutters because several patterns similar with text or LP always exist in complicated backgrounds. To verify the performance of LP text detection in natural images, we apply MB-LGP feature by combining with ensemble machine learning algorithm in purpose of selecting optimal features of small number in huge pool. The feature selection is performed by adaptive boosting algorithm that shows great performance in minimum false positive detection ratio and in computing time when combined with cascade approach. MSER is used to provide initial text regions of vehicle LP. Throughout the experiment using real images, the proposed method functions robustly extracting LP in natural scene as well as the controlled environment.