• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1076-1077
• /
• 2015

본 연구에서는 최근 널리 사용되고 있는 QFP(Quad Flat Package)의 소형화 및 대량 생산 Quad chip 공정에서 최종 외관 불량 검사를 위한 기존의 2D 영상 검사 시스템에 3D 영상 검사 시스템을 추가하여 광학 장치를 설계하고 이에 따른 영상처리 알고리즘을 개발하였다. 개발된 검사 장치는 실제 LQFP/TQFP에 생산 공정에 적용되어 불량을 검사에 적용하였으며, 10 회 반복 측정 시 최대 오차는 $1.34{\mu}m$와 측정 오차의 표준편차가 $0.715{\mu}m$으로 요구하는 3차원 불량 검사를 만족할 만한 성능을 보였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.280-282
• /
• 2009

본 논문에서는 굴곡에 의한 조도량의 차이와 명암도 차이를 퍼지 기법에 적용하여 개선된 반도체 불량 검출 방법을 제안한다. 제안된 방법은 먼저 회전각과 양선형 보관법을 이용하여 반도체 영상의 각도를 보정하는 전처리 과정 수행한다. 그리고 굴곡에 대한 조도량의 차이와 패턴 매칭를 이용하여 얻어진 오류 영역의 명암도 차이를 퍼지 소속 함수에 적용하여 결과 값을 추론한다. 최종적으로 비퍼지화된 결과 값을 적용하여 반도체의 초기 불량을 검출한다. 본 논문에서 제안한 방법을 실제 사용되는 반도체 정면 영상과 측면 영상 30쌍을 대상으로 실험한 결과, 기존의 방법에 비해서 반도체의 초기 불량 판단에 효과적인 것을 확인하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.25-30
• /
• 2018

Pattern classification system is often an important component of intelligent systems. In this paper, we present a pattern classification system consisted of the feature selection module, knowledge base construction module and decision module. We introduce a feature impact evaluation selection method based on fuzzy cluster analysis considering computational approach and generalization capability of given data characteristics. A fuzzy neural network, OFUN-NET based on unsupervised learning data mining technique produces knowledge base for representative clusters. 240 blemish pattern images are prepared and applied to the proposed system. Experimental results show the feasibility of the proposed classification system as an automating defect inspection tool.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.881-883
• /
• 2022

고객사에 납품하는 웨이퍼의 안정적인 공급을 위한 웨이퍼 엣지의 결함 검출 장비다. 본 연구에서는 OpenCV와 임베디드 시스템, 머신러닝, 전자 회로 그리고 센서/카메라 기술을 핵심 기술로 R&D 한다. 고객사에서 불량 웨이퍼 발생에 대응하기 위한 장비의 데이터를 생산하여 고객과의 신뢰도 향상 및 유지를 할 수 있다. 그리고 결함이 특정 공정 지점에서 발생하는지 탐색할 수 있다.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.32 no.4
• /
• pp.410-417
• /
• 2012

Currently domestic thermal and nuclear power plants are comprised of many type's condenser and steam generator tubes to produce the electricity of good quality. There are some methods to inspect these tubes in the event that several defects were discovered in these facilities. Among many non-destructive methods, we used guided wave to inspect the soundness of tubes, because this method is very fast to detect the defect and very simple to install the equipment and also, can inspect up to the long range at a fixed point. Also, this method has a drawback that does not detect a very small size defect. So, we made an effort to overcome this drawback through the experimentation and signal analysis according to the size and shape of the defect through the manufacture of various artificial cracks capable to generate within the small size tube in the study and we anticipate that these detect limits can be overcome along with the development of the signal processing and manufacturing technology of the sensor for the inspection.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.5
• /
• pp.211-220
• /
• 2022

Recently, research on smart factories triggered by the 4th industrial revolution is being actively conducted. Accordingly, the manufacturing industry is conducting various studies to improve productivity and quality based on deep learning technology with robust performance. This paper is a study on the method of detecting tire surface defects in the visual inspection stage of the tire manufacturing process, and introduces a tire surface defect detection method using a depth image acquired through a 3D camera. The tire surface depth image dealt with in this study has the problem of low contrast caused by the shallow depth of the tire surface and the difference in the reference depth value due to the data acquisition environment. And due to the nature of the manufacturing industry, algorithms with performance that can be processed in real time along with detection performance is required. Therefore, in this paper, we studied a method to normalize the depth image through relatively simple methods so that the tire surface defect detection algorithm does not consist of a complex algorithm pipeline. and conducted a comparative experiment between the general normalization method and the normalization method suggested in this paper using YOLO V3, which could satisfy both detection performance and speed. As a result of the experiment, it is confirmed that the normalization method proposed in this paper improved performance by about 7% based on mAP 0.5, and the method proposed in this paper is effective.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.27-35
• /
• 2023

In this paper, we propose a process of increasing productivity by applying a deep learning-based defect detection and classification system to the prepreg fiber manufacturing process, which is in high demand in the field of producing composite materials. In order to apply it to toe prepreg manufacturing equipment that requires a solution due to the occurrence of a large amount of defects in various conditions, the optimal environment was first established by selecting cameras and lights necessary for defect detection and classification model production. In addition, data necessary for the production of multiple classification models were collected and labeled according to normal and defective conditions. The multi-classification model is made based on CNN and applies pre-learning models such as VGGNet, MobileNet, ResNet, etc. to compare performance and identify improvement directions with accuracy and loss graphs. Data augmentation and dropout techniques were applied to identify and improve overfitting problems as major problems. In order to evaluate the performance of the model, a performance evaluation was conducted using the confusion matrix as a performance indicator, and the performance of more than 99% was confirmed. In addition, it checks the classification results for images acquired in real time by applying them to the actual process to check whether the discrimination values are accurately derived.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.1
• /
• pp.61-67
• /
• 2015

Recently, the initiation of outside diameter stress corrosion cracking (ODSCC) at the tube support plate region of domestic steam generators (SG) with Alloy600 HTMA tubes has been increasing. As a result, SGs with Alloy600 HTMA tubes must be replaced early or are scheduled to be replaced prior to their designed lifetime. ODSCC is one of the biggest threats to the integrity of SG tubes. Therefore, the accurate evaluation of tube integrity to determine ODSCC is needed. Eddy current testing (ECT) is conducted periodically, and its results could be input as parameters for evaluating the integrity of SG tubes. The reliability of an ECT inspection system depends on the performance of the inspection technique and abilty of the analyst. The detection probability and ECT sizing error of degradation are considered to be the performance indices of a nondestructive evaluation (NDE) system. This paper introduces an optimized evaluation method for ECT, as well as the sizing error, including the analyst performance. This study was based on the results of a round robin program in which 10 inspection analysts from 5 different companies participated. The analysis of ECT sizing results was performed using a linear regression model relating the true defect size data to the measured ECT size data.

• Design & Manufacturing
• /
• v.14 no.1
• /
• pp.8-14
• /
• 2020

In general, auto parts production assembly line is assembled and produced by automatic mounting by an automated robot. In such a production site, quality problems such as misalignment of parts (doors, trunks, roofs, etc.) to be assembled with the vehicle body or collision between assembly robots and components are often caused. In order to solve such a problem, the quality of parts is manually inspected by using mechanical jig devices outside the automated production line. Automotive inspection technology is the most commonly used field of vision, which includes surface inspection such as mounting hole spacing and defect detection, body panel dents and bends. It is used for guiding, providing location information to the robot controller to adjust the robot's path to improve process productivity and manufacturing flexibility. The most difficult weighing and measuring technology is to calibrate the surface analysis and position and characteristics between parts by storing images of the part to be measured that enters the camera's field of view mounted on the side or top of the part. The problem of the machine vision device applied to the automobile production line is that the lighting conditions inside the factory are severely changed due to various weather changes such as morning-evening, rainy days and sunny days through the exterior window of the assembly production plant. In addition, since the material of the vehicle body parts is a steel sheet, the reflection of light is very severe, which causes a problem in that the quality of the captured image is greatly changed even with a small light change. In this study, the distance between the car body and the door part and the door are acquired by the measuring device combining the laser slit light source and the LED pattern light source. The result is transferred to the joint robot for assembling parts at the optimum position between parts, and the assembly is done at the optimal position by changing the angle and step.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.16 no.3
• /
• pp.162-173
• /
• 1996

Dry magnetic particle inspection(MPI) was performed to detect the surface defects of steel ingot cast billets. Magnetic properties of several materials were characterized by the measurement of the B-H hysteresis curve. The inspection results were evaluated in terms of the magnetizing current, temperature, and the amount of magnetic particles applied to billets. Magnetic flux leakage near the defect site of interest was measured and compared with the results of calculation by the finite element method in the case of direct magnetizing current. Direct and alternating magnetizing currents for materials were deduced by the comparison of the inspections. Results of the magnetic particle inspection by direct magnetizing current were compared with those of finite element method calculations, which were verified by measuring magnetic leakage flux above the surface and the surface defects of the material. For square rods, due to the geometrical effect, the magnetic flux density at the edges along the length of the rods was about 30% of that at the center of rod face for a sufficiently large direct magnetizing current, while it was about 70% for an alternating magnetizing current. Thus, an alternating magnetizing current generates rather uniform magnetic flux density over the rods, except for the region on the face across about 10 mm from the edge. The attraction of the magnetic particle by the magnetic leakage field was nearly independent of the surface temperature of the billets up to $150^{\circ}C$. However, the temperature should have been limited below $60^{\circ}C$ for an effective fixing of gathered magnetic particles to the billet surface using methylene chloride. We also found that the amount of applied magnetic particles tremendously affected the detection capability.