• Journal of the Semiconductor & Display Technology
• /
• v.15 no.3
• /
• pp.18-23
• /
• 2016

QFN (Quad Flat No-leads Package) is one of the SMD (Surface Mount Device). Since there is no lead in QFN, there are many defects on solder. Therefore, we propose an efficient mechanism for QFN solder defect detection at this paper. For this, we employ Convolutional Neural Network (CNN) of the Machine Learning algorithm. QFN solder's color multi-layer images are used to train CNN. Since these images are 3-channel color images, they have a problem with applying to CNN. To solve this problem, we used each 1-channel grayscale image (Red, Green, Blue) that was separated from 3-channel color images. We were able to detect QFN solder defects by using this CNN. In this paper, it is shown that the CNN is superior to the conventional multi-layer neural networks in detecting QFN solder defects. Later, further research is needed to detect other QFN.

• Journal of Magnetics
• /
• v.22 no.1
• /
• pp.34-42
• /
• 2017

Material defects in the Steam Generator Tubes (SGT) of sodium cooled fast breeder reactor (PFBR) can lead to leakage of water into sodium. The water and sodium reaction will lead to major accidents. Therefore, the examination of steam generator tubes for the early detection of defects is an important requirement for safety and economic considerations. In this work, the Magnetic Flux Leakage (MFL) based Non Destructive Testing (NDT) technique is used to perform the defect detection process. The rectangular notch defects on the outer surface of steam generator tubes are modeled using COMSOL multiphysics 4.3a software. The obtained MFL images are de-noised to improve the integrity of flaw related information. Grey Level Co-occurrence Matrix (GLCM) features are extracted from MFL images and taken as input parameter to train the neural network. A comparative study on characterization have been carried out using feed-forward back propagation (FFBP) and cascade-forward back propagation (CFBP) algorithms. The results of both algorithms are evaluated with Mean Square Error (MSE) as a prediction performance measure. The average percentage error for length, depth and width are also computed. The result shows that the feed-forward back propagation network model performs better in characterizing the defects.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.45-46
• /
• 2021

본 논문에서는 제조산업의 제품 품질검사의 자동화를 위한 딥러닝 기법을 제안하고 모델의 성능 최적화를 위한 특징 추출 필터의 크기를 비교한다. 이미지 특징을 자동 추출할 수 있는 CNN을 사용하여 전문인력 없이 제품의 표면 결함을 검출하고 제품의 적합성을 판단할 수 있는 이미지 처리 알고리즘을 구축하고 산업 현장에 적용하기 위한 검증 지표로 검출 정확도와 연산속도를 측정하여 결함 검출 알고리즘의 성능을 확인한다. 또한 연산량에 따른 성능 비교를 위해 필터의 크기에 따른 CNN의 성능을 비교하여 결함 검출 알고리즘의 성능을 최적화한다. 본 논문에서는 커널의 크기를 다르게 적용했을 때 빠른 연산으로 높은 정확도의 검출 결과를 얻었다.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.11
• /
• pp.76-83
• /
• 2007

The dark-field laser scattering system has been developed to inspect surface defects in infrared cut-off filters and then laser scattering characteristics against the defects are investigated. The qualitative analysis for the reliable and accurate detection performance is described through the correlation between incident angles of a laser and viewing ones of a camera. In this paper, reliable and important information with laser scattering is given for the surface defect inspection of IR filters. Its performance has been verified through various experiments.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2008.11a
• /
• pp.379-380
• /
• 2008

• Journal of Magnetics
• /
• v.20 no.3
• /
• pp.312-316
• /
• 2015

Unlike conventional Eddy Current Test (ECT), Pulsed Eddy Current (PEC) uses a multiple-frequency current pulse through the excitation coil. In the present study, the detection of subsurface cracks using a specially designed probe that allows the detection of a deeper crack with a relatively small current density has been attempted using the PEC technique. The tested sample is a piece of 304 stainless steel (SS304) with a thickness of 30mm. Small electrical discharge machining (EDM) notches were put in the test sample at different depths from the surface to simulate the subsurface cracks in a pipe. The designed PEC probe consists of an excitation coil and a Hall sensor and can detect a subsurface crack as narrow and shallow as 0.2 mm wide and 2 mm deep. The maximum distance between the probe and the defect is 28 mm. The peak amplitude of the detected pulse is used to evaluate the cracks under the sample surface. In time domain analysis, the greater the crack depth the greater the peak amplitude of the detected pulse. The experimental results indicated that the proposed system has the potential to detect the subsurface cracks in stainless steel plates.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.57 no.1
• /
• pp.56-59
• /
• 2008

We have designed 4 types(void in insulation paper, protrusion electrode, floating electrode, surface discharge) of partial discharge(PD) defect to simulate typical faults found in oil filled power transformers. Ultra-high frequency(UHF) radiation due to PD was measured using a UHF measuring system and a conventional PD measuring system, simultaneously. Electromagnetic radiation spectra of these defects show UHF radiation up to about 1.5-2 GHz range. The phase resolved partial discharge(PRPD) patterns of UHF radiation from the PD defects were also measured and the pattern reveals distinct feature for each defect types. The UHF measuring could be used to detect PDs in oil filled transformers and analysis of the PRPD pattern should provide useful information on origin of PD signal.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.849-852
• /
• 2021

This paper presents a vision based tire mold inspection system where mold defects are inspected and the sizes of specific parts of the mold are measured. There are a lot of challenging issues as letters and pictures of intaglio are engraved on a bright surface of the tire mold. To solve the issues, we carefully selected a line-scan camera and a line light. In addition, we used PLC to control the mechanical parts. The developed system provides inspection of misspelled and deformed letters as well as a variety of the functions such as size measurement of engraved regions and inspection report file creation.

• Journal of Power System Engineering
• /
• v.17 no.2
• /
• pp.56-62
• /
• 2013

A technique is presented that uses a circular waveguide for the measurement of the bulk shear (S-wave) velocities of unconsolidated, saturated media, with particular application to near surface soils. The technique requires the measurement of the attenuation characteristics of the fundamental torsional mode that propagate along an embedded pipe, from which the acoustic properties of the surrounding medium are inferred. From the dispersion curve analysis, the feasibility of using fundamental torsional mode which is non-dispersive and have constant attenuation over all frequency range is discussed. The principles behind the technique are discussed and the results of an experimental laboratory validation are presented. The experimental data are best fitted for the different depths of wetted sand and the shear velocities are evaluated as a function of depths. Also the characteristics of the reflected signal from the defects are examined and the reflection coefficients are calculated for identifying the relation between defect sizes and the magnitude of the reflected signal.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.367-370
• /
• 2016

QFN(Quad Flat No-leads package) is one of the SMD(Surface Mount Device). Since there is no lead in QFN, there are many defects on solder. Therefore, we propose an efficient mechanism for QFN solder defect detection at this paper. For this, we employ Convolutional Neural Network(CNN) of the Machine Learning algorithm. QFN solder's color multi-layer images are used to train CNN. Since these images are 3-channel color images, they have a problem with applying to CNN. To solve this problem, we used each 1-channel grayscale image(Red, Blue, Green) that was separated from 3-channel color images. We were able to detect QFN solder defects by using this CNN. Later, further research is needed to detect other QFN.