• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 1998.11a
• /
• pp.178-182
• /
• 1998

본 연구는 의미변별법을 사용하여 우리말 형용사의 의미구조를 파악하였다. 우리말. 형용사의 대표적인 반대어 40쌍이 선정되었으며, 요인분석 결과 5개의 요인이 추출되었다. 의미변별법을 색 이미지에 적용시진 결과도 비슷한 요인들을 추출 했으나, 전체 변량을 설명하는 양에 있어서 차이를 나타냈다. 본 연구 결과는 형용사를 사용하는 감성연구의 기반기술을 제공할 것이라 예측된다.

• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.1778-1782
• /
• 2004

A new algorithm, CBIR (Correlation-Based Image Registration) was proposed to improve the resolution of image registration for PSP (Pressure-Sensitive Paint). The local displacement vectors were obtained by finding the displacement which maximizes the cross-correlation between two interrogation windows of 'wind-off' and 'wind-on' images. A recursive multigrid processing was employed to increase the non-linear spatial resolutions. The variations of image were precisely measured without identifying the control points.

• Journal of Science Education
• /
• v.44 no.2
• /
• pp.214-224
• /
• 2020

Physics image and interest are factors that influence physics learning. Freshmen enter an engineering college under various learning conditions when they were in high school. Understanding physics image and interest according to characteristics of freshmen will help college physics education. The purpose of this study is to investigate the physics image and interest of freshmen in an engineering college according to their gender and physics course completion in high school and discuss the educational implications of college students on physics learning. The subjects of the study are 664 first grade students in engineering college. We analyzed physics image and interest of students according to gender and physics course completion in high school. Physics image is analyzed using semantic analysis. As a result of the analysis, the physics image is different according to the physics course completion. Interest in Physics depends on gender and physics course completion. Finally, we discuss the educational implications of college physics learning for engineering students.

• Journal of Korea Multimedia Society
• /
• v.15 no.11
• /
• pp.1292-1304
• /
• 2012

Recently, digital images are extremely forged by editors or advertisers. Also, amateurs can modify images throughout easy editing programs. In this study, we propose identification and analytical methods for the modified images to figure out those problems. In modified image analysis, we classify two parts; a filtering change and a geometric change. Those changes have an algorithm based on interpolation so that we propose the algorithm which is able to analyze a trace on a modified area. With this algorithm, we implement a detection map of interpolation using minimum filter, laplacian algorithm, and maximum filter. We apply the proposed algorithm to modified image and are able to analyze its modified trace using the detection map.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1191-1194
• /
• 2022

인터넷의 발전으로 수많은 이미지와 비디오를 손쉽게 이용할 수 있게 되었다. 이미지와 비디오 데이터의 양이 기하급수적으로 증가함에 따라, JPEG, HEVC, VVC 등 이미지와 비디오를 효율적으로 저장하기 위한 부호화 기술들이 등장했다. 최근에는 인공신경망을 활용한 학습 기반 모델이 발전함에 따라, 이를 활용한 이미지 및 비디오 압축 기술에 관한 연구가 빠르게 진행되고 있다. NNIC (Neural Network based Image Coding)는 이러한 학습 가능한 인공신경망 기반 이미지 부호화 기술을 의미한다. 본 논문에서는 NNIC 모델과 인공신경망 기반의 초해상화(Super Resolution) 모델을 합동훈련하여 기존 NNIC 모델보다 더 높은 성능을 보일 수 있는 방법을 제시한다. 먼저 NNIC 인코더(Encoder)에 이미지를 입력하기 전 다운 스케일링(Down Scaling)으로 쌍삼차보간법을 사용하여 이미지의 화소를 줄인 후 부호화(Encoding)한다. NNIC 디코더(Decoder)를 통해 부호화된 이미지를 복호화(Decoding)하고 업 스케일링으로 초해상화를 통해 복호화된 이미지를 원본 이미지로 복원한다. 이때 NNIC 모델과 초해상화 모델을 합동훈련한다. 결과적으로 낮은 비트량에서 더 높은 성능을 볼 수 있는 가능성을 보았다. 또한 합동훈련을 함으로써 전체 성능의 향상을 보아 학습 시간을 늘리고, 압축 잡음을 위한 초해상화 모델을 사용한다면 기존의 NNIC 보다 나은 성능을 보일 수 있는 가능성을 시사한다.

• Korean Journal of Optics and Photonics
• /
• v.31 no.5
• /
• pp.205-212
• /
• 2020

In the performance of a wind turbine system, the blades play a vital role. However, they are susceptible to damage arising from complex and irregular loading (which may even cause catastrophic collapse), and they are expensive to maintain. Therefore, it is very important both to find defects after blade manufacturing is completed and to find damage after the blade is used for a certain period of time. This study provides a new perspective for the detection of internal defects in glass-fiber- and carbon-fiber-reinforced panels, which are used as the main materials in wind turbine blades. A gap or fracture between fiber-reinforced materials, which may occur during blade manufacturing or operation, is simulated by drilling a hole 5 mm in diameter in the middle layer of the laminated material. Then, a digital-image-correlation (DIC) method is used to detect internal defects in the blade. Tensile load is applied to the fabricated specimen using a tensile tester, and the generated changes are recorded and analyzed with the DIC system. In the glass-fiber-reinforced laminated specimen, internal defects were detected from a strain value of 5% until the end of the experiment, while in the case of the carbon-fiber-reinforced laminated specimen, internal defects were detected from 1% onward. It was proved using the DIC system that the defect was detected as a certain level of strain difference developed around the internal defects, according to the material properties.

• Smart Media Journal
• /
• v.13 no.3
• /
• pp.27-35
• /
• 2024

This study deals with a method of combining image generation using Semi Supervised Learning based on GAN (Generative Adversarial Network) and image classification using ResNet50. Through this, a new approach was proposed to obtain more accurate and diverse results by integrating image generation and classification. The generator and discriminator are trained to distinguish generated images from actual images, and image classification is performed using ResNet50. In the experimental results, it was confirmed that the quality of the generated images changes depending on the epoch, and through this, we aim to improve the accuracy of industrial accident prediction. In addition, we would like to present an efficient method to improve the quality of image generation and increase the accuracy of image classification through the combination of GAN and ResNet50.

• Journal of Korean Society of Forest Science
• /
• v.104 no.4
• /
• pp.685-696
• /
• 2015

This study intends to evaluate visual preference and image for 10 types of forest sceneries which have own distinct characteristics without overlapping each other among 22 types of beautiful forest sceneries based on Korea National Parks. Z scale and paired comparison methods were used to analyze visual preference for forest scenery, and SD scale method was also adopted to assess visual image for forest scenery. Experiments were conducted with 3 different university students groups. The results of this study can be summarized as follows. First, the evaluation results of Z scale (N=70) and paired comparison (N=64) methods show similar visual preference for different types of forest scenery. 'Scenery inside forests' and 'valley and rock' sceneries have the highest level of preference, and 'panorama of ridges', 'ridge of curious rock peaks', and 'waterfall and cliff' also have relatively higher preference level than others. However, 'community of dead trees' has the lowest preference in forest sceneries. Second, the factor analysis outputs of the image scores for beautiful forest scenery types by SD scale (N=66) show 3 factors of 'mysterious charm', 'colorful sense', and 'vividness/variety'. Examined the variables of forest scenery types for each factor, the image of 'mysterious charm' is based on the sceneries of mountain ridges, the image of 'colorful sense' is focused on the sceneries of color-expressed sense of the season, and the image of 'vividness/variety' is based on the sceneries of waterscape and curious rock peaks.

• Journal of The Korean Association of Information Education
• /
• v.20 no.5
• /
• pp.527-534
• /
• 2016

This study is an analysis of Software image using the semantic differential scale with elementary students. We have selected the items in a total of 35 pairs of test about software-related image adjectives and are categorized into 7 main factors, and then analyzed the entire image with the students. The analysis of the differences between the software images of sex, the female students than male students were recognized the software is complex, slow and difficult and do not want to have. The analysis of the self-awareness on the software, the students who know that well recognized for the software select the positive term for the software. The inter-grade analysis are the older grade students were the answer to the objective features of the software like more difficult and complex.

• Journal of the Korean School Mathematics Society
• /
• v.19 no.1
• /
• pp.1-19
• /
• 2016

This study investigated the mathematical notation used today in terms of skip ping the parenthesis. At first we have studied the elementary and secondary curriculum content related to omitted rules. As a result, it is difficult to find explicit evidence to answer that question 'What is the calculation of the $48{\div}2(9+3)$?'. In order to inquire the notation fundamentally, we checked the characteristics on prefix, infix and postfix, and looked into the advantages and disadvantages on infix. At the same time we illuminated the development of mathematical notation from the point of view of skipping the parenthesis. From this investigation, we could check that this interpretation was smooth in the point of view that skipping the parentheses are the image of the function. Through this we proposed some teaching methods including 'teaching mathematical notation based on historic genetic principle', 'reproduction of efforts to overcome the disadvantages of infix and understand the context to choose infix', 'finding the omitted parentheses to identify the fundamental formula' and 'specifying the viewpoint that skipping the multiplication notation can be considered as an image of the function'.