• Journal of Surface Science and Engineering
• /
• v.50 no.3
• /
• pp.170-176
• /
• 2017

EPIG (Electroless Pd/immersion Au) process was studied to replace ENIG (electroless Ni/immersion Au) and ENEPIG (electroless Ni/electroless Pd/immersion Au) processes for bump surface treatment used in high reliable flip chip packages. The palladium and gold layers formed by EPIG process were uniform with thickness of 125 nm and 34.5 nm, respectively. EPAG (Electroless Pd/autocatalytic Au) also produced even layers of palladium and gold with the thickness of 115 nm and 100 nm. TEM results exhibited that the gold layer in EPIG surface had crystalline structure while the palladium layer was amorphous one. After annealing at 250 nm, XPS analysis indicated that the palladium layer with thickness more than 22~33 nm could act as a diffusion barrier of copper interconnects. As a result of comparing the chip shear strength obtained from ENIG and EPIG surfaces, it was confirmed that the bonding strength was similar each other as 12.337 kg and 12.330 kg, respectively.

• Journal of the Korean Society for information Management
• /
• v.35 no.1
• /
• pp.13-32
• /
• 2018

The purpose of this study is to analyze the association among the subject areas of big data research papers. The subject group of the units of analysis was extracted by applying co-citation networks, and the rules of association were analyzed using Apriori algorithm of R program, and visualized using the arulesViz package of R program. As a result of the study, 22 subject areas were extracted and these subjects were divided into three clusters. As a result of analyzing the association type of the subject, it was classified into 'professional type', 'general type', 'expanded type' depending on the complexity of association. The professional type included library and information science and journalism. The general type included politics & diplomacy, trade, and tourism. The expanded types included other humanities, general social sciences, and general tourism. This association networks show a tendency to cite other subject areas that are relevant when citing a subject field, and the library should consider services that use the association for academic information services.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.4
• /
• pp.215-228
• /
• 2019

HDF5 data format is used to effectively store and distribute large volume of Geostationary Ocean Color Imager(GOCI) satellite data. The Korea Ocean Satellite Center has developed and provided a GOCI Data Processing System(GDPS) for general users who are not familiar with HDF5 format. Nevertheless, it is not easy to merge and process Hierarchical Data Format version5(HDF5) data that requires an understanding of satellite data characteristics, needs to learn how to use GDPS, and stores location and attribute information separately. Therefore, the open source R and rhdf5, data.table, and matrixStats packages were used to develop algorithm that could easily utilize satellite data in HDF5 format without the need for the process of using GDPS.

• Proceedings of the KWS Conference
• /
• 2009.11a
• /
• pp.76-76
• /
• 2009

눈부신 전자산업의 발달로 대부분의 전자제품이 다기능/경박단소화 되고 있어, 고밀도 실장 기술인 양면 표면실장과 고집적 패키징 기술인 패키지 적층 공정의 적용이 점차 확대되고 있다. 따라서 양면 표면실장 및 패키지 적층 공정에 사용되는 저온 접합용 무연 솔더 즉, $183^{\circ}C$(Sn-37Pb 공정 솔더 융점) 이하의 융점을 가지는 저온 무연 솔더에 대한 관심이 높아지고 있다. 한편, 미세피치 적용 분야에 있어 ACF/P를 이용한 COG 접속 분야 외에도 최근 저온 접합용 무연 솔더를 이용한 접속 분야가 각광을 받고 있다. 따라서, 접속피치 미세화에 대응하기 위해 스크린 인쇄성을 향상시킬 수 있는 저온 무연 솔더 paste 제조 및 공정 기술의 개발이 필요한 실정이다. 현재 대표적인 저온 무연 솔더 조성은 Sn-Bi계($138^{\circ}C$ 융점)와 Sn-In계($120^{\circ}C$ 융점)이다. 하지만, 이들 조성의 신뢰성 등에 있어 개선의 여지가 있으므로 이를 해결하기 위한 무연솔더 조성의 개발이 필요하다. 이와 같은 관점에서, 본 연구는 $137^{\circ}C$의 융점을 갖는 Sn-57.6Ag-0.4Ag 저온 무연 솔더 paste를 $217^{\circ}C$의 융점을 갖는 Sn-3.0Ag-0.5Cu 솔더 paste와 비교하여 인쇄성, reflow 특성, void inspection, 미세조직 관찰 및 underfill 적용 등의 실험을 실시하였다.

• Journal of the Microelectronics and Packaging Society
• /
• v.26 no.4
• /
• pp.23-31
• /
• 2019

In this study, we conducted numerical analyses using the Taguchi method and finite element method to calculate the thermal deformation of a printed circuit board and the effect of design factors on the thermal deformation. Analysis results showed that the thermal deformation of the panel had the strongest effect on the thermal deformation and shape of the strip and unit. In particular, the deformation in the z direction was larger than that in the xy-plane direction. The effect of design factors and the design conditions for reducing the thermal deformation of the panel and strip changed at the unit level. Therefore, it is recommended that panel-level thermal deformation must be controlled to reduce the final thermal deformation at the unit level because the thermal deformation of the strip strongly affects that of the unit.

• Korean Journal of Optics and Photonics
• /
• v.25 no.2
• /
• pp.78-84
• /
• 2014

In this paper secondary optics for an LED lamp with a narrow beam angle of $15^{\circ}$ were optimized by using a two-reflector system, to reduce both its size and the occurrence of satellite rings. The conic constant and the curvature of the primary reflector were determined by considering the relation of the source size to the beam angle, and the optimal position and radius of the secondary reflector were found for reducing the occurrence of satellite rings. Luminous flux efficiency was about 80%.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.4
• /
• pp.69-78
• /
• 2023

With the rapid growth of artificial intelligence, the demand for semiconductors is enormously increasing everywhere. To ensure the manufacturing quality and quantity simultaneously, the importance of automatic defect detection during the packaging process has been re-visited by adapting various deep learning-based methodologies into automatic packaging defect inspection. Deep learning (DL) models require a large amount of data for training, but due to the nature of the semiconductor industry where security is important, sharing and labeling of relevant data is challenging, making it difficult for model training. In this study, we propose a new framework for securing sufficient data for DL models with fewer computing resources through a divide-and-conquer approach. The proposed method divides high-resolution images into pre-defined sub-regions and assigns conditional labels to each region, then trains individual sub-regions and boundaries with boundary loss inducing the globally coherent and seamless images. Afterwards, full-size image is reconstructed by combining divided sub-regions. The experimental results show that the images obtained through this research have high efficiency, consistency, quality, and generality.

• The Journal of the Korea Contents Association
• /
• v.20 no.3
• /
• pp.223-234
• /
• 2020

This research is the association rule analysis using Apriori algorithm of data mining focusing on 32 predictive key words extracted from Hong (2019) affecting successful aging in Korea. And, to examine rules and patterns of those key words or predictive variables, this research used support, confidence, and lift. The data was analyzed with the R version 3. 5. 1 program, and visualized using arulesViz package and visNetwork. It was found that the variables highly associated with successful aging in Korea were 'hobby', 'volunteer service', 'preparation', and 'exercise'. This research concludes that, the variable which needs to be considered first of all for successful aging in Korea is 'hobby', followed by 'volunteer service', 'preparation', and 'exercise'.

• The Journal of the Korea Contents Association
• /
• v.18 no.2
• /
• pp.616-625
• /
• 2018

The purpose of this study was to analyze the relationship between the BMI and suicidal ideation using Korean Health Panel data from 2009 to 2013 to identify risk factors for suicide. We conducted a logistic regression analysis using the R statistical package to analyze the relationship between the BMI and the suicidal ideation. The results of this study show that all models with BMI had a statistically significant as a significant variable. It was found that the obese group was more likely to suicide ideation than the other groups. Based on these results, we try to find social implications for suicide prevention and intervention.

• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.345-362
• /
• 2017

Various data visualization methods are utilized to analyze a huge amount of data. Among various methods, a three-dimensional image requires the rotation of the image to show a stereo image on a two-dimensional screen. This study discusses two methods of batch method and real-time method, which make it possible to construct of stereo images to improve the restriction of the three-dimensional image display with virtual reality. This investigation can be useful to better explore a three-dimensional data structure.