• 한국인터넷방송통신학회논문지
• /
• 제24권2호
• /
• pp.147-153
• /
• 2024

X-ray는 주로 식품 검사, 의료 진단, 산업제품의 비파괴 검사에 사용되어 왔다. 최근에는 전기자동차의 핵심 부품인 배터리 검사에도 X-ray가 많이 활용되고 있다. 빠르게 증가하는 배터리 검사 수요를 고려할 때 수동 검사보다는 딥러닝 기반의 자동 배터리 검사의 도입이 필요하다. 딥러닝 성능은 기본적으로 학습 데이터 크기에 의존하는데 X-ray 응용 분야에서는 다량의 불량 데이터 확보가 쉽지 않다. 본 논문에서는 한정된 불량 데이터로부터 데이터 증강을 통해 학습, 검증, 테스트 데이터를 확보하고 학습 데이터 크기에 따른 X-ray 배터리 검사의 성능 변화를 확인하였다. 본 논문 결과를 통해 검사 성능 개선 프로세스에 대한 이해를 높일 수 있다.

• 한국추진공학회지
• /
• 제16권2호
• /
• pp.65-72
• /
• 2012

초음파 결함 분석 프로그램은 초음파 반사법을 기반으로 초음파 신호처리 기법을 적용하여 개발되었고, FRP 층간분리 및 FRP/내열고무 미접착 결함을 정량적으로 측정할 수 있었다. 복합재 연소관에서 검출된 결함은 절단하여 전산화 단층촬영 및 영상 현미경으로 분석하였고, 결함 분석 프로그램의 결과와 일치하였다. 본 논문은 복합재 연소관의 초음파시험 데이터를 C-Scan 영상으로 변환하여 결함을 분석할 수 있는 프로그램 개발 과정을 기술하였다.

• 한국안전학회지
• /
• 제30권5호
• /
• pp.52-58
• /
• 2015

Although lots of safety inspection and precision safety diagnosis have been conducted on concrete retaining wall, there is no comprehensive analysis on the basis of the accumulated data associated with the statistic. Especially, the concentrated management is necessary on the evaluation items that cause critical damages for the efficient performance. In this regard, this study conducted a correlation analysis between the 18 condition evaluation items and defect index for the concrete retaining wall as well as how each item affects the final defect index as much as in the manual. As a result, correlation coefficient between sliding and overturning was 0.601, which means that they have a strong correlation, and the most influential item on defect index is the condition of drainage that scored the 0.750 correlation coefficient. In addition, as a result of regression analysis, the condition of drainage with the 0.683 correlation coefficient has a strong correlation with the defect index. If the condition evaluation items are integrated or readjusted based on the results of the statistical analysis in this study, the more efficient and accurate maintenance will be possible.

• 산업경영시스템학회지
• /
• 제45권4호
• /
• pp.157-166
• /
• 2022

In the case of a die-casting process, defects that are difficult to confirm by visual inspection, such as shrinkage bubbles, may occur due to an error in maintaining a vacuum state. Since these casting defects are discovered during post-processing operations such as heat treatment or finishing work, they cannot be taken in advance at the casting time, which can cause a large number of defects. In this study, we propose an approach that can predict the occurrence of casting defects by defect type using machine learning technology based on casting parameter data collected from equipment in the die casting process in real time. Die-casting parameter data can basically be collected through the casting equipment controller. In order to perform classification analysis for predicting defects by defect type, labeling of casting parameters must be performed. In this study, first, the defective data set is separated by performing the primary clustering based on the total defect rate obtained during the post-processing. Second, the secondary cluster analysis is performed using the defect rate by type for the separated defect data set, and the labeling task is performed by defect type using the cluster analysis result. Finally, a classification learning model is created by collecting the entire labeled data set, and a real-time monitoring system for defect prediction using LabView and Python was implemented. When a defect is predicted, notification is performed so that the operator can cope with it, such as displaying on the monitoring screen and alarm notification.

• 동력기계공학회지
• /
• 제8권3호
• /
• pp.52-57
• /
• 2004

A new non-destructive inspection technique has been developed. One characteristic of the technique is that defects are visualized by laser ray. Magnetic domains and domain walls of a magneto-optical sensor(MO sensor) are varied by the magnetic flux leaked by defects, and the variations are observed by the reflected light of the laser ray. The information of defect can remotely be inspected by this technique in a real time. This paper describes the results estimated on the 2-dimensional surface defects and opposite-side defects in a ferromagnetic material and the natural surface defect in a clutch disk wheel. The light region of a visible image and the magnitude of a reflected light increases as the input current of the magnetizer increases. The natural surface defect, that has not the width of crack's open mouth, can be also visualized like as 2-dimensional artificial defects.

• 전기학회논문지
• /
• 제56권5호
• /
• pp.876-883
• /
• 2007

The magnetic flux leakage(MFL) type nondestructive testing(NDT) method is widely used to detect corrosion, defects and mechanical deformation of the underground gas pipelines. The object pipeline is magnetically saturated by the magnetic system with permanent magnet and yokes. Hall sensors detect the leakage fields in the region of the defect. The defects are sometimes occurred in group. The accuracy of the detecting signals in this defect cluster become lowered because of the complexity of the defect cluster. In this paper, the effects of the multi -defects are analyzed. The detecting signals are computed by 3-dimensional finite element method and compared with real measurement. The results say that, rather than the size of the defects, the effects of the relative position of the multi-defects are very important on the detecting signals.

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2001년도 추계학술발표논문집
• /
• pp.210-214
• /
• 2001

PCB관련 제품의 최종 제작단계에서 defect 검사 과정은 제품의 질을 유지하기 위해 필수적인 단계이다. PCB 자동화 검사 시스템은 사람에 의해 이루어지는 품질검사에서 발견되는 비용을 절감하고, 신뢰성있는 제작 프로세스를 유지하기 위해 적극적으로 개발되고 있다. 이 논문에서는 PCB 필름의 defect를 검사하기 위하여 적응적 템플렛 기반 검사 방법을 제시하고자 한다. 고정된 템플릿은 구현하기 편리하고 속도면에서 이점을 발휘할 수 있으나, 강력한 센서의 선택에 제약이 있을 환경 하에서 100%에 근접하는 오류검출률 defect detection rate이 요구되는 고정된 템플릿을 제작하는 것에 문제가 있을 수 있다. 여기서는 템플릿 모델에 유연성을 부여하기 위하여 템플릿의 이미지를 목표 이미지들의 상태에 따라 템플릿을 적응적으로 구축하여 검사과정에 동적으로 적용하는 기법을 개발하고자 한다.

• 전자공학회논문지A
• /
• 제30A권2호
• /
• pp.46-52
• /
• 1993

The estimation of yield for semiconductor devices requires not only establishment of critical area but also a new parameter of process defect density that contains inspection mean defect density related cleanness of manufacure process line, minimum feature size and the total number of mask process. We estimate the repaired yield of memory devide, leads the semiconductor technique, repaired by redundancy scheme in relation with defect density distribution function, and we confirm the repaired yield for different devices as this model. This shows the possibility of the yield estimation as statistical analysis for the condition of device related cleanness of manufacture process line, design and manufacture process.

• 비파괴검사학회지
• /
• 제35권6호
• /
• pp.407-413
• /
• 2015

Ultrasonic infrared thermography is an active thermography methods. In this method, mechanical energy is introduced to a structure, it is converted into heat energy at the defects, and an infrared camera detects the heat for inspection. The heat generation mechanisms are dependent on many factors such as structure characteristics, defect type, excitation method and contact condition, which make it difficult to predict heat distribution in ultrasonic infrared thermography. In this paper, a method to simulate frictional heating, known to be one of the main heat generation mechanisms at the closed defects in metal structures, is proposed for ultrasonic infrared thermography. This method uses linear vibration analysis results without considering the contact boundary condition at the defect so that it is intuitive and simple to implement. Its advantages and disadvantages are also discussed. The simulation results show good agreement with the modal analysis and experiment result.

• 전자공학회논문지S
• /
• 제35S권4호
• /
• pp.76-84
• /
• 1998

The requirements of the accuracy, the high speed and the stability are very important factors in the defect-detection sytem for the texture. In this paper, we describe a novel scheme of the defect detection using a statistical behavior of defect patterns. Some prior knowledge as to the characteristics of flaws is that the defects are consistently distributed in the space and the noise are randomly generated. An empirical knowledge is adapted for the binarization and the determination process of defects in textured image. Since the process of the determination exclude the segmentations or delineation steps, we are able to meet the speed requirements. We show the validity of the scheme through the simulation of textured images.