• Annual Conference of KIPS
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.33-34
• /
• 2023

성장하는 빅 데이터 시장과 빅 데이터 수의 기하급수적인 증가는 기존 컴퓨팅 환경에서 데이터 처리의 어려움을 야기한다. 특히 이미지 데이터 처리 속도는 데이터양이 많을수록 현저하게 느려진다. 이에 본 논문에서는 Apache Spark와 OpenCV를 활용한 분산 클러스터 컴퓨팅 환경의 대용량 이미지 머신러닝 시스템을 제안한다. 제안하는 시스템은 Apache Spark를 통해 분산 클러스터를 구성하며, OpenCV의 이미지 처리 알고리즘과 Spark MLlib의 머신러닝 알고리즘을 활용하여 작업을 수행한다. 제안하는 시스템을 통해 본 논문은 대용량 이미지 데이터 처리 및 머신러닝 작업 속도 향상 방법을 제시한다.

• Journal of the Korean Society of Clothing and Textiles
• /
• 제26권3_4호
• /
• pp.402-411
• /
• 2002

The objective of this study was to investigate sensibility images of Korean traditional motifs cognized by college students in U.S.A. The subjects consisted of 217 male and 351 female undergraduate students. The experimental materials used in this study were 48 stimuli and questionnaires, composed of 7-point semantic differential scales of 15 bipolar adjectives. Twelve motifs selected from 3 groups of Korean motifs were used as motif stimuli. Twelve repeated patterns were constructed from them to be applied on a CAD-simulated dress. The data were analyzed by factor analysis, ANOVA, Duncan's multiple range test. The major finding were as follows: 1. Four dimensions were emerged accounting for the dimensional structure of the images of Korean traditional motifs. These dimensions were ‘quality’, ‘simplicity’, ‘cheerfulnees’, and ‘modernity’. Among them, ‘quality’and ‘simplicity’were the major dimensions. 2. Category, interpretation type, composition type, and application object had significant effects on the images of above-mentioned dimensions. The interpretation type had a significant effect on ‘quality’image, the composition type on ‘cheerful’image, and the application object on ‘modernity’image.

• Annual Conference of KIPS
• /
• 한국정보처리학회 2010년도 추계학술발표대회
• /
• pp.609-612
• /
• 2010

최근 스마트 폰을 이용한 사용자들이 생성하는 사진 데이터의 양이 급속히 증가하였다. 폭발적인 사진 데이터 양의 증가는 사용자가 원하는 사진에 대한 접근을 어렵게 하였다. 때문에 본 연구에서는 사진의 접근 및 관리의 효율을 높이기 위한 폭소노미를 통한 태그 추천 시스템을 안드로이드 모바일 플랫폼과 서버의 연계로 구현하였다. 구현된 애플리케이션은 25,000 장의 사진을 기반으로 하는 폭소노미를 통해 태그 추천을 하며, 태그 추천에 평균적으로 5.5 초의 시간이 걸렸다.

• Annual Conference of KIPS
• /
• 한국정보처리학회 2007년도 춘계학술발표대회
• /
• pp.301-304
• /
• 2007

자동으로 이미지 안에 존재하는 객체들을 인식하는 문제는 내용 기반 이미지 검색이나 로봇 비전과 같은 다양한 분야들에서 매우 중요한 문제이다. 이 문제를 해결하기 위하여 본 논문에서는 객체의 주요 색상 정보들을 이용하여 실시간으로 이미지 안의 객체들을 인식하는 알고리즘을 제안한다. 본 논문에서 제안하는 방법의 전체적인 구조는 다음과 같다. 처음에 MPEG-7 색상 정보 기술자들 중 하나인 주요 색상 정보 기술자를 이용하여 객체의 주요 색상 정보들을 추출한다. 이 때 이 정보는 가우시안 색상 모델링을 통하여 빛이나 그림자와 같은 외부 환경 조건에 좀 더 강인한 색상 정보로 변환된다. 다음으로 변환된 색상 정보들을 기반으로 주요 객체와 입력 이미지와의 픽셀 값차이를 계산하고, 임계값 이상의 값을 가지는 픽셀들을 제거한다. 마지막으로 입력 이미지에서 제거되지 않은 픽셀들을 기반으로 하나의 영역을 생성한다. 결론으로서, 본 논문에서는 제안된 방법에 대한 실험 평가들을 수행 및 분석하고 몇몇 한계점들에 대해서 알아본다. 또한 이 문제들을 해결하기 위한 앞으로의 연구 계획에 대해서 기술한다.

• Korean Journal of Childcare and Education
• /
• 제8권6호
• /
• pp.219-239
• /
• 2012

The purpose of this study was to analyze pre-service teachers' perceptions and change on the images about young children. The subjects of this study were 31 students in the early childhood department at a university. It was a longitudinal study conducted for four years. The data regarding the images that students had about young children were collected by having them answer a question. The question was "Young children are... Because of..." Subjects were asked to answer that question when they were freshmen and then, to answer the same question when they were seniors in order to see the change. A content analysis and frequency were conducted. The study of the results could be summarized as follows. Firstly, from their first year to senior year, the images that pre-service teachers perceived such as, vibrant active beings and beings with various characteristics" did not change. However, the images such as "being who was capable of interacting with environments" had changed. And the new image which viewed a child as a being who should be respected emerged. Secondly, the factors that influenced the image changes were based on the experience of working as assistant teachers at an early childhood setting, and practicum. Therefore, a systematic work on assistant teachers should be done and a concept about child-center education should be built.

• Journal of KIISE:Databases
• /
• 제36권4호
• /
• pp.327-340
• /
• 2009

To achieve the noise reduction effect in boundary image matching, we use the moving average transform of time-series matching. Our motivation is based on an intuition that using the moving average transform we may exploit the noise reduction effect in boundary image matching as in time-series matching. To confirm this simple intuition, we first propose $\kappa$-order image matching, which applies the moving average transform to boundary image matching. A boundary image can be represented as a sequence in the time-series domain, and our $\kappa$-order image matching identifies similar images in this time-series domain by comparing the $\kappa$-moving average transformed sequences. Next, we propose an index-based matching method that efficiently performs $\kappa$-order image matching on a large volume of image databases, and formally prove the correctness of the index-based method. Moreover, we formally analyze the relationship between an order $\kappa$ and its matching result, and present a systematic way of controlling the noise reduction effect by changing the order $\kappa$. Experimental results show that our $\kappa$-order image matching exploits the noise reduction effect, and our index-based matching method outperforms the sequential scan by one or two orders of magnitude.

• Journal of The Korean Association For Science Education
• /
• 제37권6호
• /
• pp.1051-1061
• /
• 2017

An image is a comprehensive result that you have experienced about an object and means the image that you have on the surface of your consciousness. The image of the subject has an important influence on learning the subject. The image analysis of the subjects that the learners have will be good data to decide the direction of teaching and learning. The purpose of this study is to develop and apply measurement tools for physics image and discuss its educational implications. The research method is to develop the measurement tools for the physics image by semantic analysis method and apply it to the secondary pre-service physics teacher. The subjects of the study were 39 first graders, 31 second graders, 37 third graders, and 38 fourth graders at the University of Education, a total of 145 students, 82 of whom were male and 63 were female. The study results show that the image measurement tools for physics consisted of 25 items from five elements: 'interest,' 'feeling,' 'scope,' 'evaluation,' and 'viewpoint.' There were statistically significant differences between the male and female students in applying the measurement tools developed for the physics image of secondary pre-service physics teachers. Male students showed significantly higher statistical significance than female students in the 'interest' and 'feeling' elements of measurement tools for the physics image. In the 'scope' element of measurement tools for the physics image the second grade was statistically higher than the fourth grade. Finally, we discussed educational implications for image analysis of physics and the usefulness of using measurement tools in physics image.

• Journal of KIISE:Databases
• /
• 제34권3호
• /
• pp.193-205
• /
• 2007

Increasing amount of digital images requires more accurate and faster way of image retrieval. So far, image retrieval method includes content-based retrieval and keyword based retrieval, the former utilizing visual features such as color and brightness and the latter utilizing keywords which describe the image. However, the effectiveness of these methods as to providing the exact images the user wanted has been under question. Hence, many researchers have been working on relevance feedback, a process in which responses from the user are given as a feedback during the retrieval session in order to define user’s need and provide improved result. Yet, the methods which have employed relevance feedback also have drawbacks since several feedbacks are necessary to have appropriate result and the feedback information can not be reused. In this paper, a novel retrieval model has been proposed which annotates an image with a keyword and modifies the confidence level of the keyword in response to the user’s feedback. In the proposed model, not only the images which have received positive feedback but also the other images with the visual features similar to the features used to distinguish the positive image are subjected to confidence modification. This enables modifying large amount of images with only a few feedbacks ultimately leading to faster and more accurate retrieval result. An experiment has been performed to verify the effectiveness of the proposed model and the result has demonstrated rapid increase in recall and precision while receiving the same number of feedbacks.

• Journal of the Korean Geotechnical Society
• /
• 제40권1호
• /
• pp.15-28
• /
• 2024

This paper proposes a method to acquire image data inside tunnel structures and a method to structure the acquired image data. By improving the conditions by which image data are acquired inside the tunnel structure, high-quality image data can be obtained from area type tunnel scanning. To improve the data acquisition conditions, a longitudinal rail of the tunnel can be installed on the tunnel ceiling, and image data of the entire tunnel structure can be acquired by moving the installed rail. This study identified 0.5 mm cracked simulation lines under a distance condition of 20 m at resolutions of 3,840 × 2,160 and 720 × 480 pixels. In addition, the proposed image-data-structuring method could acquire image data in image tile units. Here, the image data of the tunnel can be structured by substituting the application factors (resolution of the acquired image and the tunnel size) into a relationship equation. In an experiment, the image data of a tunnel with a length of 1,000 m and a width of 20 m were obtained with a minimum overlap rate of 0.02% to 8.36% depending on resolution and precision, and the size of the local coordinate system was found to be (14 × 15) to (36 × 34) pixels.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 한국신호처리시스템학회 2005년도 추계학술대회 논문집
• /
• pp.213-216
• /
• 2005

자동차 번호판의 이미지에서 번호의 추출이나, 직물 이미지에서 오염 또는 훼손부분의 추출 또는 방사성 폐기물이나 기타 독극물 보관함의 이미지에서 오염이나 산화에 의한 훼손부위 등과 같은 목표물 이미지 추출은 흑백 이미지에서 명암의 차이를 이용하는 것보다는 컬러 이미지에서 색상의 차이를 이용하는 것이 더 효율적일 때가 많으며, 특히 배경과 목표물의 명암차이가 크지 않은 경우에 효과적이다. 배경과 목표물이 갖는 색상의 차이를 이용하여 분리하기 위해서 적색(R), 녹색(G), 청색(B) 의 RGB 평면 또는 순도(H), 포화도(S), 휘도(I)를 사용하는 HSI 컬러 평면 등이 많이 사용되며, 이 때 배경과 목표물의 색상의 히스토그램을 구해보면 보면 많은 경우 유사한 색 정보가 배경과 목표물에 공통으로 포함되어 분리에 어려움을 겪게 된다. 본 논문에서는 Fisher 선형 분리자(Fisher's linear discriminant)[1] 함수를 이용하여 3차원의 색상 특징 벡터를 1차원 직선에 투사하여 변환된 1차원 공간상에서 복잡성을 줄이고 효과적으로 분류할 수 있는 기법을 제안하였으며, 이를 도축된 식용 가금류의 영상에 적용하고 변질된 부분이 포함되어 식용으로 사용할 수 없는 것들을 효과적으로 분류할 수 있음을 보였다.