• The KIPS Transactions:PartB
• /
• v.17B no.3
• /
• pp.197-206
• /
• 2010

High-quality and low-price digital printing devices are nowadays abused to print or forge official documents and bills. Identifying color laser printers will be a step for media forensics. This paper presents a new method to identify color laser printers with printed color images. Since different printer companies use different manufactural systems, printed documents from different printers have little difference in visual. Analyzing this artifact, we can identify the color laser printers. First, high-frequency components of images are extracted from original images with discrete wavelet transform. After calculating the gray-level co-occurrence matrix of the components, we extract some statistical features. Then, these features are applied to train and classify the support vector machine for identifying the color laser printer. In the experiment, total 2,597 images of 7 printers (HP, Canon, Xerox DCC400, Xerox DCC450, Xerox DCC5560, Xerox DCC6540, Konica), are tested to classify the color laser printer. The results prove that the presented identification method performs well with 96.9% accuracy.

• Journal of the Microelectronics and Packaging Society
• /
• v.27 no.3
• /
• pp.89-93
• /
• 2020

Complex warpage behavior of the electronic packages causes internal stress so many kinds of mechanical failure occur such as delamination or crack. Efforts to predict the warpage behavior accurately in order to prevent the decrease in yield have been approached from various aspects. For warpage prediction, silicon is generally treated as a homogeneous material, therefore it is described as showing no warpage behavior due to thermal loading. However, it was reported that warpage is actually caused by residual stress accumulated during grinding and polishing in order to make silicon wafer thinner, which make silicon wafer inhomogeneous through thickness direction. In this paper, warpage behavior of the single-side polished wafer at solder reflow temperature, the highest temperature in packaging processes, was measured using 3D digital image correlation (DIC) method. Mechanism was verified by measuring coefficient of thermal expansion (CTE) of both mirror-polished surface and rough surface.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.49 no.2
• /
• pp.177-182
• /
• 2023

Since facial skin pigmentation is one of the visual characteristics of skin aging, it is important to evaluate skin pigmentation in the cosmetics and aesthetic fields. Several groups have investigated and developed the image analysis methods for skin pigmentation and some of the groups reported the age-related changes of the number and size of facial pigmented spots. However, they didn't show the changes of the number and size of pigment spots by defined size, and there is no report for Korean female regarding pigmentation. A total of 194 Korean females aged 20 ~ 79 (48.97 ± 17.11 years) were analyzed for the number, size, and intensity of pigmented spots using various filters such as large high-pass filter and median filter in their digital facial images. There were significant correlations between age and total pigmented spot number (No.), size, and intensity (I) (pearson's correlation coefficient r = 0.688, r = 0.645, r = -0.563), and significant correlations were also observed between the number and intensity of pigmented spots of different sizes. According to the ANOVA results, there were significant differences in the percentage of spot size of 2 ~ 4 mm² and > 20 mm² between 20's and 70's. In other words, with aging, pigmentation increases in the facial skin, and the large increase in pigmentation is particularly noticeable in Korean women.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.23 no.2
• /
• pp.71-86
• /
• 2021

This study was conducted to find information on the university's image from words related to 'Korea National College of Agriculture and Fisheries (KNCAF)' in Naver News. For this purpose, word frequency analysis, TF-IDF evaluation and semantic network analysis were performed using web crawling technology. In word frequency analysis, 'agriculture', 'education', 'support', 'farmer', 'youth', 'university', 'business', 'rural', 'CEO' were important words. In the TF-IDF evaluation, the key words were 'farmer', 'dron', 'agricultural and livestock food department', 'Jeonbuk', 'young farmer', 'agriculture', 'Chonju', 'university', 'device', 'spreading'. In the semantic network analysis, the Bigrams showed high correlations in the order of 'youth' - 'farmer', 'digital' - 'agriculture', 'farming' - 'settlement', 'agriculture' - 'rural', 'digital' - 'turnover'. As a result of evaluating the importance of keywords as five central index, 'agriculture' ranked first. And the keywords in the second place of the centrality index were 'farmers' (Cc, Cb), 'education' (Cd, Cp) and 'future' (Ce). The sperman's rank correlation coefficient by centrality index showed the most similar rank between Degree centrality and Pagerank centrality. The KNCAF articles of Naver News were used as important words such as 'agriculture', 'education', 'support', 'farmer', 'youth' in terms of word frequency. However, in the evaluation including document frequency, the words such as 'farmer', 'dron', 'Ministry of Agriculture, Food and Rural Affairs', 'Jeonbuk', and 'young farmers' were found to be key words. The centrality analysis considering the network connectivity between words was suitable for evaluation by Cd and Cp. And the words with strong centrality were 'agriculture', 'education', 'future', 'farmer', 'digital', 'support', 'utilization'.

• Tunnel and Underground Space
• /
• v.33 no.6
• /
• pp.574-593
• /
• 2023

In this study, the effectiveness and applicability of a newly designed Compact CNS shear box for conducting direct shear tests on jointed rock specimens were investigated. CNS joint shear tests were conducted on jointed rocks with Artificially generated roughness while varying the fracture surface roughness coefficient and initial normal stress conditions. In addition, displacement data were validated by Digital image correlation analysis, fracture patterns were observed, and comparative analysis was conducted with previously studied shear behavior prediction models. Furthermore, the accuracy of the displacement data was confirmed through DIC analysis, the fracture patterns were observed, and the shear properties obtained from the tests were compared with existing models that predict shear behavior. The findings exhibited a strong correlation with specific established empirical models for predicting shear behavior. Furthermore, the potential linkage between the characteristics of shear behavior and fracture patterns was deliberated. In conclusion, the CNS shear box was shown to be applicable and effective in providing data on the shear characteristics of the joint.