• 제목/요약/키워드: image understanding

검색결과 1,068건 처리시간 0.029초

터보 코드의 적용을 통한 워터마크 신호 보호 및 원 영상의 화질 보존 (Protection of Watermark Signals and Preservation of Original Images by Turbo Encoding)

  • 조동욱;배영래
    • 한국멀티미디어학회:학술대회논문집
    • /
    • 한국멀티미디어학회 2001년도 추계학술발표논문집
    • /
    • pp.869-872
    • /
    • 2001
  • This paper proposes on the implementation of efficient image transmission system by generation and protection of watermark signals. For this, the image structure understanding is performed for improving the image quality and the generation of watermark signals. Then, the histogram is constructed and the watermark signals are selected from this. At this stage, by embedding of the coefficients of curve fittness into the lower 4 its of the image data pixels, image quality degradation due to the embedding of watermark signals are prevented. Finally, turbo encoding is performed to solve the problem of watermark signal losses that may occur on channels when transmitting. Especially, a new interleaving method called semi-random interleaver is proposed.

  • PDF

사진 속 피사체의 법선 벡터 예측에 기반한 햅틱 상호 작용 (Haptic Interaction with Objects Displayed in a Picture based on Surface Normal Estimation)

  • 김승찬;권동수
    • 로봇학회논문지
    • /
    • 제8권3호
    • /
    • pp.179-185
    • /
    • 2013
  • In this paper we propose a haptic interaction system that physically represents the underlying geometry of objects displayed in a 2D picture, i.e., a digital image. To obtain the object's geometry displayed in the picture, we estimate the physical transformation between the object plane and the image plane based on homographic information. We then calculate the rotated surface normal vector of the object's face and place it on the corresponding part in the 2D image. The purpose of this setup is to create a force that can be rendered along with the image without distorting the visual information. We evaluated the proposed haptic rendering system using a set of pictures of objects with different orientations. The experimental results show that the participants reliably identified the geometric configuration by touching the object in the picture. We conclude this paper with a set of applications.

영상 처리 방법을 이용한 트리 길이와 열화면적 계측 (Measurement of Electrical Treeing Length and Area by Use of Image Processing)

  • 백관현;김재환
    • 한국조명전기설비학회지:조명전기설비
    • /
    • 제8권1호
    • /
    • pp.57-62
    • /
    • 1994
  • 영상처리 방식을 이용하여 유기 절연재료의 트리잉 열화를 측정하였다. 전통적으로 목측(가시측정)에 의한 트리현상 변화를 관찰해 왔으나 트리형상 특징과 변화를 이해하는데 정확한 측정에 어려움을 만들었다. 영상처리 시스템을 이용한 트리 측정장치를 이용하여 트리진전길이와 열화면적들의 고유특성을 영상처리 시스템을 이용하여 목측치와 거의 대등한 결과가 나오는 트리진전 영상처리 자동계측 시스템을 구현하였다.

  • PDF

어도비포토샵 프로그램을 이용한 anaglyphic 입체영상 제작법 (Construction of Anaglyphic Stereo Pair Image using Adobe $Photoshop^{(R)}$ Program)

  • 김지웅;이세정;유임주
    • Applied Microscopy
    • /
    • 제37권2호
    • /
    • pp.143-146
    • /
    • 2007
  • The objects of the nature have three dimensional (3-D) parameters. The 3-D profiles are embedded on the photographs and microscopic images. To understand 3-D configuration, stereo pair image with thick section is frequently employed. The perception of 3-D images is possible with the aid of stereoscopic glasses, although the expert can perceive 3-D images without the glasses. Anaglyphic stereo images are constructed by various softwares from commercial and freeware. Here we would like to present an easy anaglyphs construction method with Adobe $Photoshop^{(R)}$ based on tilting paired images from high voltage electron microscope. The anaglyphic stereo images constructed revealed the same 3-D perception with conventional stereoscopy. We could zoom in/out the anaglyph image digitally to investigate the detail configuration by real time. This method is expected to contribute to understanding complex structures 3 dimensionally.

A Novel Text to Image Conversion Method Using Word2Vec and Generative Adversarial Networks

  • LIU, XINRUI;Joe, Inwhee
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2019년도 춘계학술발표대회
    • /
    • pp.401-403
    • /
    • 2019
  • In this paper, we propose a generative adversarial networks (GAN) based text-to-image generating method. In many natural language processing tasks, which word expressions are determined by their term frequency -inverse document frequency scores. Word2Vec is a type of neural network model that, in the case of an unlabeled corpus, produces a vector that expresses semantics for words in the corpus and an image is generated by GAN training according to the obtained vector. Thanks to the understanding of the word we can generate higher and more realistic images. Our GAN structure is based on deep convolution neural networks and pixel recurrent neural networks. Comparing the generated image with the real image, we get about 88% similarity on the Oxford-102 flowers dataset.

Is Brand Identity Aligned with Brand Image on Instagram? An Empirics-First Investigation of the Indian Brands

  • Anand V;Daruri Venkata Srinivas Kumar
    • Asia pacific journal of information systems
    • /
    • 제33권3호
    • /
    • pp.768-791
    • /
    • 2023
  • Effective brand management using images has been a challenge for the brand managers. The brand identity-brand image alignment on the social media is an important yet mostly-overlooked phenomenon. We proposed a scalable Google Cloud Vision-based approach for measuring the alignment between brand identity and brand image, and understanding the brand positions. We analyzed 3247 images of 13 leading Indian brands on Instagram. Images containing wordy announcements by the firms are in stark contrast with the relatively more emotive images by the users. It leads to a noticeable disconnect between the brand identity and brand image. Also, the private sector brands do not always outperform the public sector brands in branding efforts. By offering practical guidance on how to measure and reduce the misalignment, this study paved a feasible path towards better visual branding on Instagram.

과학 탐구 수업에서 초등학생들이 바라는 과학 교사의 모습에 대한 요인 분석 (Factor Analysis of Image of Science Teacher Desired by Elementary School Students in Scientific Inquiry-Based Instruction)

  • 채유정;박재용
    • 한국초등과학교육학회지:초등과학교육
    • /
    • 제40권3호
    • /
    • pp.366-389
    • /
    • 2021
  • 이 연구에서는 과학 탐구 수업에서 초등학생들이 바라는 과학 교사의 모습을 구성하는 요인들을 추출한 후, 이들 요인 사이의 구조적 연관성을 분석하였다. 과학 탐구 수업에서 초등학생들이 바라는 과학교사의 모습을 구성하는 요인들을 확인하기 위하여 탐색적 요인 분석(EFA)을 실시하였고, 그 결과를 바탕으로 확인적 요인 분석(CFA)을 실시하였다. 또한, 확인적 요인 분석을 통해 산출된 요인 구조를 토대로 과학 탐구 수업에서 초등학생들이 바라는 과학 교사의 모습을 종합적으로 살펴보았다. EFA 결과, 과학 탐구 수업에서 초등학생들이 바라는 과학 교사의 모습을 이루는 요인으로 5개 요인(교수 전략, 평가 전략, 사전 지도, 학생 이해, 학습 발달)을 추출하였다. CFA 결과, 교수 전략과 학습 발달, 교수 전략과 학생 이해 사이에 상관관계가 비교적 높게 나타났다. 추출된 요인 구조를 바탕으로 과학 탐구 수업에서 과학 교사의 모습에 대한 학생들의 인식을 종합적으로 살펴본 결과, 초등학생들은 과학 교사가 학생들에 대한 이해를 반영하여 학습 발달을 지원하고, 효과적인 수업 전략과 적절한 평가 방법을 활용하는 것을 중요하게 인식하고 있었다.

오목렌즈로 관찰되는 상에 대한 초등학교 예비교사의 이해 (Elementary Preservice Teachers' Understanding of the Image Observed in a Diverging Lens)

  • 권경필
    • 한국과학교육학회지
    • /
    • 제35권5호
    • /
    • pp.871-876
    • /
    • 2015
  • 본 연구에서는 오목렌즈로 관찰되는 상에 대한 초등예비교사들의 이해를 분석하였다. 연구를 위해 29명의 3학년 초등예비교사들을 대상으로 2009개정 과학교육과정 '오목렌즈로 물체 관찰하기' 탐구활동을 수행하게 하였으며, 탐구활동에서 예비교사들이 겪는 어려움에 대해 조사하였다. 그 결과는 다음과 같다. 첫째, 예비교사들은 상의 크기 비교에 어려움을 나타냈다. 둘째, 렌즈로부터 멀리 떨어진 물체가 더 작게 관찰되는 원인을 옳게 설명하지 못했다. 셋째, 눈과 렌즈사이의 거리가 멀어짐에 따른 상의 크기 변화를 옳게 설명하지 못했다. 넷째, 오목렌즈에 의한 상의 형성 개념을 5단계로 나누었으며, 대부분의 예비교사들은 1단계에 머물러 있었다. 연구결과로부터 초등 예비교사교육에 대한 제고 방안에 대해 논의하였다.

패션 컬렉션에 나타난 번지는 이미지 문양의 염색기법과 조형미 (Dyeing Technique and Aesthetics of Diffusing Image Pattern in Fashion Collection)

  • 유현정
    • 한국의상디자인학회지
    • /
    • 제14권3호
    • /
    • pp.1-10
    • /
    • 2012
  • This study has purpose to understanding recent trends and image effect of textile in fashion by looking to dyeing technique and aesthetics of diffusing image pattern. Further the study will hope to provide basic data for fashion and textile designer fashion. The methods of this study are academic literatures as well as practical study through case studies about actual works. Dyeing techniques and aesthetics of diffusing image pattern since 2000 fashion collection and their results are as follows. First, Dyeing techniques of diffusing image pattern are wet-dyeing and printing. There are plangi dyeing, salt dyeing, alcohol dyeing and ikat dyeing in the wet-dyeing and spray dyeing, discharge dyeing in the printing. There are tie-dyeing, rumpling-dyeing and plank-dyeing in the plangi. Second, Forming of the diffusing image pattern are due to uneven seeping and the contingency of a chemical reaction by reducing the pressure and intensity. Third, Aesthetics of diffusing image pattern are extracted purity missing childhood, fantasy escaping facts of the situation and abstraction representing informal world from formative characteristics of diffusing image pattern.

  • PDF

Impact of Female Consumer Self-Image on Pursued Fashion Style

  • Yoon, DoohAh;Yu, JongPil
    • 패션비즈니스
    • /
    • 제21권3호
    • /
    • pp.29-42
    • /
    • 2017
  • This study investigates the impact of female consumer self-image on pursued fashion style. A survey was carried out among 717 women between the ages of 20 and 60 living in Seoul, Incheon, and Gyeonggi. Analysis was conducted in the following manner: SPSS 18.0 was used to perform an Exploratory Factor Analysis (descriptive analysis, principal factor analysis, Pearson correlation analysis, frequency analysis, and reliability analysis) and AMOS 21.0 was used to carry out a Confirmatory Factor Analysis. Based on a Structure Equation Model, results show that ideal self-image and realistic self-image, which are factors derived from psychology, affect pursued fashion style. By contrast, social self-image - derived from social contexts - does not. Therefore, the female consumers' self-image influences pursued fashion style; this is opposed to the relationship between the realistic self-image and ideal self-image of women, which is more unconscious and self-satisfying. The presented results indicate that we should respond to changes in the fashion industry and develop a deeper understanding of consumer niches to discover the factors that predict purchasing behavior. This knowledge can then be applied to establish market strategies. This study contributes to the literature by producing preliminary data that can help support such strategy formulation in the fashion and clothing industry.