• Journal of the Korean Society for Nondestructive Testing
• /
• v.17 no.3
• /
• pp.162-173
• /
• 1997

Automatic on-line surface inspection systems have been applied for monitoring a quality of steel strip surfaces. One of the important issues in this application is the performance of on-line defect classifiers. Rule-based classification table methods which are conventionally used for this purpose have been suffered from their low performances. In this work, probabilistic neural networks and the enhanced classification tables which are newly proposed here are applied as alternative on-line classifiers to identify types of surface defects on cold rolled strips. Probabilistic neural networks have shown very excellent performance for classification of surface defects.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.577-586
• /
• 2022

Detection and classification of steel surface defects are critical for product quality control in the steel industry. However, due to its low accuracy and slow speed, the traditional approach cannot be effectively used in a production line. The current, widely used algorithm (based on deep learning) has an accuracy problem, and there are still rooms for development. This paper proposes a method of steel surface defect detection combining EfficientNetV2 for image classification and YOLOv5 as an object detector. Shorter training time and high accuracy are advantages of this model. Firstly, the image input into EfficientNetV2 model classifies defect classes and predicts probability of having defects. If the probability of having a defect is less than 0.25, the algorithm directly recognizes that the sample has no defects. Otherwise, the samples are further input into YOLOv5 to accomplish the defect detection process on the metal surface. Experiments show that proposed model has good performance on the NEU dataset with an accuracy of 98.3%. Simultaneously, the average training speed is shorter than other models.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.1
• /
• pp.107-116
• /
• 2024

Copper Filter Dryer (CFD) are responsible for removing impurities from the circulation of refrigerant in refrigeration and cooling systems to maintain clean refrigerant, and defects in CFD can lead to product defects such as leakage and reduced lifespan in refrigeration and cooling systems, making quality assurance essential. In the quality inspection stage, human inspection and defect judgment methods are traditionally used, but these methods are subjective and inaccurate. In this paper, YOLOv7 object detection algorithm was used to detect defects occurring during the CFD Shaft pipe and welding process to replace the existing quality inspection, and the detection performance of F1-Score 0.954 and 0.895 was confirmed. In addition, the cause of defects occurring during the welding process was analyzed by analyzing the sensor data corresponding to the Timestamp of the defect image. This paper proposes a method for manufacturing quality assurance and improvement by detecting defects that occur during CFD process and analyzing their causes.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.25 no.2
• /
• pp.127-132
• /
• 2005

The optical glass lens is required high dimensional precision such as the lack of defect. In this paper, we examined the detectable defect by using the resonant ultrasound spectroscopy(RUS). The RUS is the measurement system which is to excite the specimen and to inspect the differences of resonant frequency pattern between acceptable specimen and specimen which has some defects. In this paper, for nondestructive evaluation by using RUS, we measured the resonant frequency of each specimen which is spherical and aspherical glass lens. With the results, we knew the polishing processing degree of spherical glass lens by the measured resonant frequency and could evaluate the characteristic of aspherical glass lens about some flaws.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.369-372
• /
• 2010

평판 디스플레이(flat panel display)의 크기가 커짐에 따라 다양한 기판을 이용한 제조 방법이 개발되고 있다. 디스플레이 제조 공정 중 기판의 결함을 찾아서 분류하는 검사 시스템은 최종 제품의 품질을 결정하는 매우 중요한 부분이다. 본 연구는 머신비전 기술을 이용하여 불투명하고 반사율이 높은 기판 표면의 결함을 찾아내고, 이 결함을 스크래치(scratch), 흑결함(dark defect), 백결함(white defect)으로 분류하는 장치를 구현하는데 목적이 있다. 이를 구현하기 위해 본 논문에서는 정밀 스테이지(stage)와 라인 카메라(line CCD camera)을 이용한 광학계를 활용하여 검사 시스템을 구현하였다. 구축된 시스템을 이용하여 취득한 이미지를 12 개의 영역으로 등분하여 각각의 국부 영역에 대한 문턱값 연산(thresholding)을 적용함으로써 조명의 불균일을 의한 검출 에러율을 획기적으로 낮추었다. 간단한 컴퓨터비전 알고리듬의 채용으로도 검사 시스템의 구현이 가능함을 보였다.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.146-146
• /
• 2017

현대의 스마트폰 및 태블릿pc등을 가능하게 만든 집적 기술 중의 하나는 3차원 집적 회로(3D-IC)와 같은 패키징 기술이다. 이러한 첨단 3차원 집적 기술은 메모리집적을 통한 대용량 메모리 모듈 개발뿐만 아니라, 메모리와 프로세서의 집적, high-end FPGA, Back side imaging (BSI) 센서 모듈, MEMS 센서와 ASIC 집적, High Bright (HB) LED 모듈 등에 적용되고 있다. 3D-IC의 3차원 모듈 제작 시에는 기존에 발생하지 않았던 여러 가지 파괴 모드들이 발생하고 있는데 Thermal/Photonic Emission 장비 등 기존의 2차원 결함분리 (Fault Isolation) 기술로는 첨단의 3차원 적층 제품들에서 발생하는 불량을 비파괴적으로 혹은 3차원적으로 분리하는 것이 불가능하므로, 비파괴 3차원 결함 분리 기술은 향후 선행 제품 적기 개발에 매우 필수적인 기술이다. 본 연구는 3D-IC 반도체의 비파괴적 내부결함 검사를 위하여 가시광선-근적외선 대역(351nm~1770nm)의 InGaAs (Indium Galium Arsenide) 계열 영상검출기 (imaging detector)를 사용하여 분광 시스템 광학 설계를 통한 초분광 영상 기반 검출 모듈을 제작하였다. 제작된 초분광 영상 기반 검출 모듈을 이용하여 구리 회로 위에 실리콘 웨이퍼가 3단 적층 된 반도체 더미 샘플의 초분광 영상을 촬영하였으며, 촬영된 초분광 영상에 대하여 Chemometrics model 기반의 분석기술을 적용하여 실리콘 웨이퍼 내부의 집적 구조에 대한 검사가 가능함을 확인하였다.

• Journal of Korea Multimedia Society
• /
• v.3 no.2
• /
• pp.112-120
• /
• 2000

In this paper, we implement a real-time textile dimension inspection system. It can detect various types of real defects which determine the quality of fabric product, defect positions of textile, classify the distortion angel of moving textile and the density. In the implemented system, we measure the density of textile using zone-crossing method with optical lens to solve the noise and real-time problems. And we compensate distortion angel of textile with the classification of distortion types using gaussian gradient and mean gradient features. And also, it detecs real defects of textile and its positions using gray level co-occurrence matrix features. The implemented texile demension inspection systemcan inspect textile dimensions such as density, distortion angle, defect of textile and defect position at real-time. In the implemented proposed texitile dimension inspection system, It is possible to calculate density and detect default of textile at real-time dimension inspection system, it is possible to calculate density and detect default of textile at textile states throughout at all the significant working process such as dyeing, manufacturing, and other texitle processing.

• Fire Science and Engineering
• /
• v.26 no.4
• /
• pp.31-41
• /
• 2012

This study collected 5,100 cases of gas accident occurred in Korea for 14 years from 1995 to 2008, established Database and based on it, analyzed them by detailed forms and reasons. As the result of analyzing the whole city gas accidents with Poisson analysis, the item of "Careless work-Explosion-Pipeline' showed the highest rate of accidents for the next 5 years. And, "Joint Losening and corrosion-Release-Pipeline" showed the lowest rate of accident. In addition, for the result of analyzing only accidents related to LPG vaporizer, "LPG-Vaporizer-Fire" showed the highest rate of accident and "LPG-Vaporizer-Products Faults" showed the lowest rate of accident. Also, as the result of comparing and analyzing foreign LPG accident accompanied by Jet fire, facility's defect which is liquid outflow cut-off device and heat exchanger's defect were analyzed as the main reason causing jet fire, like the case of Korea, but the number of accidents for the next 5 years was the highest in "LPG-Mechanical-Jet fire" and "LPG-Mechanical-Vapor Cloud" showed the highest rate of accidents. By grafting Poisson distribution theory onto gas accident expecting program of the future, it's expected to suggest consistent standard and be used as the scale which can be used in actual field.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.33 no.1
• /
• pp.7-13
• /
• 2013

The surface microcrack over a few tens of micrometers is one of severe problems of a steel wire rod to lead to the failure of the final products, so the method to evaluate crack depth has been required to develop. This work investigates the feasibility of electromagnetic acoustic resonance (EMAR) for this problem. EMAR is the method for measurement of resonant features using electromagnetic acoustic transducer (EMAT). Generally, EMAR is sensitive to small variation of the structures and easy to apply it to the industrial field because of the feature of noncontact measurement. Through several EMAR experiments, the change of the resonant frequencies and attenuation in reverberation has been observed. The results confirms that the surface cracks of around 100 micrometer depth can be detected successfully with the present method.

• The Journal of the Korea Contents Association
• /
• v.6 no.11
• /
• pp.266-275
• /
• 2006

Mostly mechanical products are connected by several components instead of single accessory in product process. Although majority of assembly process is automated, the fault analysis is not automated because it needs expert knowledge in various fields to support inclusive decision-marking. This paper proposes an assembly fault analysis support system that uses image regions which can be easily accessed and understood by experts of various fields. An assembly fault analysis support system helps effective fault analysis from assembly by integrating image regions, product design information, and fault detection information. The proposed method enables fault information access from multimedia information by segmenting product images. After product images are segmented by labeling, design information and fault information are integrated in extended Attributed Relational Graph.