• Journal of the Korean Society of Clothing and Textiles
• /
• v.22 no.8
• /
• pp.1090-1098
• /
• 1998

The purpose of this study was to classify size specifications of the Korean adult male for the men's ready-made garments, especially jacket and dress shirts. By the stratified sampling method, data were collected by the real anthropometric measurement and by the photo-graphic sources. Sample size was 532 subjects as the sample and their age range was from 28 to 35 years old. 66 variables from the direct anthropometric data in total were applied to analyze. ANOVA in SPSS WIN package was applied to the data and the expected frequency distribution of 10.000 men was calculated by the extraction of density function. This study was performed to classify size specifications by the control dimensions and at the same interval of KS-K. The drop values of 15, 12 and 9 have the high coverage rate of 26.00%, 24.29% and 21.06% respectively and are composed of the majority of 71.35% of the subjects. According to the drop values, size specifications and distribution of control and reference dimensions are predicted. About 52.12% of the expected frequency distribution without stature were covered by 12 size specifications.

• Journal of KIISE
• /
• v.42 no.4
• /
• pp.483-486
• /
• 2015

Security systems that use face recognition are vulnerable to spoofing attacks where unauthorized individuals use a photo or video of authorized users. In this work, we propose a method to detect a face spoofing attack with a video of an authorized person. The proposed method uses three sequential frames in the video to extract features by using Fourier Transform and Dense-SIFT filter. Then, classification is completed with a Support Vector Machine (SVM). Experimental results with a database of 200 valid and 200 spoof video clips showed 99% detection accuracy. The proposed method uses simplified features that require fewer memory and computational overhead while showing a high spoofing detection accuracy.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.35-36
• /
• 2006

In Nd:YAG laser welding, evaluation methods of welding flaw are various. But, the method due to fume shape is difficult to classification of welding flaw. The Nd:YAG laser process is known to have high speed and deep penetration capability to become one of the most advanced welding technologies. At the present time, some methods are studied for measurement of fume shape by using high-speed camera and photo diode. This paper describes the machining characteristics of SM45C carbon steel welding by use of an Nd:YAG laser. In spite of its good mechanical characteristics, SM45C carbon steel has a high carbon contents and suffers a limitation in the industrial application due to the poor welding properties. In this study, fume shape was measured by infrared thermal camera that is non-contact/non-destructive thermal measurement equipment through change of laser generating power, speed, focus. Weld was performed on bead-on method. Measurement results are compared as two equipments. Here, two results are composed of measurement results of fume quantities due to fume shape by infrared thermal camera and inspection results of weld bead include weld flaws by ultrasonic inspector.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.30 no.2
• /
• pp.226-235
• /
• 2020

Objectives: This study was to investigate the signs and notations of skin absorption, carcinogenicity, germ cell mutagenicity, and reproductive toxicity in the occupational exposure limits of Korea and of other advanced countries. Methods: Information on occupational exposure limits in Korea, the USA, the UK, Germany, and Japan was investigated through the Internet, and items marked as carcinogenicity and skin absorption were compared by country. Results: Legal occupational exposure limits have been greatly simplified. However, in the case of HSE WEL, skin absorption, carcinogenicity classification, sensitization, and in the case of DFG MAK, skin absorption, carcinogenicity, pregnancy risk group, germ cell mutagenicity, airway and skin sensitization, photo contact sensitization, and vapor pressure were provided. Conclusions: It is desirable to indicate the carcinogenicity and skin absorption within permissible limits, and to include information on critical effects in chemical substance exposure limits to uphold the right to know of industrial hygienists and workers in Korea. It is also necessary to clarify the precautions, limitations and protections for skin absorption.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.35 no.10
• /
• pp.1252-1263
• /
• 2011

This study examined the current state of snowboarding apparel. We investigated the preferred design, the required functions, and inconvenient factors in snowboard apparel through interviews with snowboard pro-players and questionnaires with functional apparel consumers. The research was conducted as follows. In order to raise problems through interviews with pro-snowboarders and grasp the individual traits of consumers, a survey was conducted with male and female consumers in their 20s-30s who enjoy snowboarding and those who had purchased specialized brand snowboarding apparel more than once. After the survey with consumers, this study set the classification standard for snowboard maniacs according to snowboarding frequency and classified the snowboarders into two groups. Both groups carry MP3 players most frequently in ordinary times and they preferred notable and brilliant colors and partially-used patterns (printed patterns). Through the investigation of the mobile functions that the snowboarders wanted for snowboarding apparel, it was found that the most preferred functions were those of listening to music and photo/video image-taking.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.892-911
• /
• 2019

Biometric recognition systems have been widely used for information security. Among the most popular biometric traits, there are fingerprint and face due to their high recognition accuracies. However, the security system that uses face recognition as the login method are vulnerable to face-spoofing attacks, from using printed photo or video of the valid user. In this study, we propose a fast and robust method to detect face-spoofing attacks based on the analysis of spatial frequency differences between the real and fake videos. We found that the effect of a spoofing attack stands out more prominently in certain regions of the 2D Fourier spectra and, therefore, it is adequate to use the information about those regions to classify the input video or image as real or fake. We adopt a divide-conquer-aggregate approach, where we first divide the frequency domain image into local blocks, classify each local block independently, and then aggregate all the classification results by the weighted-sum approach. The effectiveness of the methodology is demonstrated using two different publicly available databases, namely: 1) Replay Attack Database and 2) CASIA-Face Anti-Spoofing Database. Experimental results show that the proposed method provides state-of-the-art performance by processing fewer frames of each video.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.11-14
• /
• 2020

The importance of data in the development of deep learning is very high. Data with high morphological features are usually utilized in the domains where careful lens calibrations are needed by a human to capture those data. Synthesis of high morphological data for that domain can be a great asset to improve the classification accuracy of systems in the field. Unsupervised learning can be employed for this task. Generating photo-realistic objects of interest has been massively studied after Generative Adversarial Network (GAN) was introduced. In this paper, we propose Morpho-GAN, a method that unifies several GAN techniques to generate quality data of high morphology. Our method introduces a new suitable training objective in the discriminator of GAN to synthesize images that follow the distribution of the original dataset. The results demonstrate that the proposed method can generate plausible data as good as other modern baseline models while taking a less complex during training.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.211-213
• /
• 2022

Images in the textbooks influence the learning process. Students often see pictures before reading the text and these pictures can enhance the power of imagination of the students. The findings of some researches show that the images in textbooks can increase students' creativity. However, when learning major subjects, reading a textbook or looking at a picture alone may not be enough to understand the topics and completely realize the concepts. Studies show that viewers remember 95% of a message when watching a video than reading a text. If we can combine textbooks and videos, this teaching method is fantastic. The "TEXT + IMAGE + VIDEO (Animation)" concept could be more beneficial than ordinary ones. We tried to give our solution by using machine learning Image Classification. This paper covers the features, approaches and detailed objectives of our project. For now, we have developed the prototype of this project as a web app and it only works when accessed via smartphone. Once you have accessed the web app through your smartphone, the web app asks for access to use the camera. Suppose you bring your smartphone's camera closer to the picture in the textbook. It will then display the video related to the photo below.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.951-952
• /
• 2023

소위 '인생네컷'이라 불리는 셀프사진관은 MZ 세대의 새로운 놀이 문화로 떠오르며 사용자 수가 나날이 증가하고 있다. 그러나 짧은 시간 내에 다양한 포즈를 취해야 하는 셀프사진관 특성상 촬영이 낯선 사람에게는 여전히 진입장벽이 존재한다. 더불어 매번 비슷한 포즈와 사진 결과물에 기존 사용자는 점차 흥미를 잃어가는 문제점도 발생하고 있다. 이에 본 연구에서는 셀프사진관 사용자의 관계를 분류하는 모델을 개발하여 관계에 따른 적합하고 다양한 포즈를 추천하는 프레임워크를 제안한다. 사용자의 관계를 'couple', 'family', 'female_friend', 'female_solo', 'male_friend', 'male_solo' 총 6 개로 구분하였고 실제 현장과 유사하도록 단색 배경의 이미지를 우선으로 학습 데이터를 수집하여 모델의 성능을 높였다. 모델 학습 단계에서는 모델의 성능을 높이기 위해 여러 CNN 기반의 모델을 전이학습하여 각각의 정확도를 비교하였다. 결과적으로 195 장의 test_set 에서 accuracy 0.91 의 성능 평가를 얻었다. 본 연구는 객체 인식보다 객체 간의 관계를 학습시켜 관계성을 추론하고자 하는 것을 목적으로, 연구 결과가 희박한 관계 분류에 대한 주제를 직접 연구하여 추후의 방향성이나 방법론과 같은 초석을 제안할 수 있다. 또한 관계 분류 모델을 CCTV 에 활용하여 미아 방지 혹은 추적과 구조 등에 활용하여 국가 치안을 한층 높이는 데 기대할 수 있다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.491-492
• /
• 2024

맞벌이 가정이 증가함에 따라 영유아, 장애인, 노인 등의 사회적 약자를 낮시간 동안 보육/보호하는 데이케어 센터의 수요가 증가하고 있다. 데이케어 센터는 센터 경쟁력 확보 및 보호자 만족도 제고를 위해서 피보호자의 일상 사진을 제공하는 곳이 대부분이다. 하지만 데이케어 센터의 직원이 다수의 사람에 대한 사진을 촬영 및 선별해서 메시지를 전송하는 일은 데이케어 센터 본연의 업무를 방해할 수 있다. 따라서 본 논문에서는 사진 선별을 업무 부담을 완화시키는데 도움을 줄 수 있는 얼굴 특징 기반 사진 자동분류하는 시스템을 개발한다. 제안한 방법에서는 얼굴 특징 추출 기법과 클러스터링 알고리즘인 DBSCAN을 이용하여 얼굴기준 사진 분류시스템을 설계하엿다. 특히, OpenCV와 face recognition 라이브러리를 이용하여 카메라로 촬영된 사진 속의 얼굴 객체를 인식하고 얼굴사진을 저정한 후 얼굴의 특징을 추출한다.