• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.285-286
• /
• 2021

본 논문에서는 변환 학습을 기반으로 한 다중 클래스 이미지 장면 분류 방법을 제안하도록 한다. ImageNet 대형 이미지 데이터 세트에서 사전 훈련 된 네트워크 모델을 사용하여 다중 클래스의 자연 장면 이미지를 분류하였다. 실험에서 최적화 된 ResNet 모델은 Kaggle의 Intel Image Classification 데이터 세트에 분류되어 우수한 결과를 얻었다.

• The Journal of the Korea Contents Association
• /
• v.11 no.4
• /
• pp.92-100
• /
• 2011

In this paper, we propose a novel blind digital watermarking method based on a cellular automata transform (CAT). CAT is based on dynamic systems known as cellular automata(CA) and uses transform bases which are differently defined by a rule number, the number of neighbors, the number of cells, and an initial state, etc. The proposed CAT based method is compared with a blind watermarking method based on DWT which is commonly used for a domain transform in signal processing. We analyse properties on changes of DWT coefficients and CAT coefficients under various attacks and determine optimal parameters for a watermarking method robust to attacks. The simulations show that the watermarked images with high PSNR and MSSIM look visually identical to originals and are robust against most of typical image processing attacks. Moreover, the proposed CAT based watermarking method is superior to the DWT based one in robustness to most of typical image processing attacks including JPEG compression, median and average filtering, scaling, cropping, and histogram equalization.

• The Journal of the Korea Contents Association
• /
• v.15 no.12
• /
• pp.8-17
• /
• 2015

This paper presents a sound into color signal conversion using a colored-hearing synesthesia. The aim of the present paper is to implement a real-time conversion system which focuses on both hearing and sight which account for a great part of bodily senses. The proposed method of the real-time conversion of color into sound, in this paper, was simple and intuitive where scale, octave and velocity were extracted from MIDI input signals, which were converted into hue, intensity and saturation, respectively, as basic elements of HSI color model. In experiments, we implemented both the hardware system for delivering MIDI signals to PC and the VC++ based software system for monitoring both input and output signals, so we made certain that the conversion was correctly performed by the proposed method.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.555-559
• /
• 2006

This paper will intend to contribute to extracting and storing various form of information on user interests by using structural rules user makes and XML-based word document converting techniques. The system named PPE consists of three essential element. One is converting element which converts word documents like HWP, DOC into XML documents, another is extracting element to prepare structural rules and extract concerned information from XML document by structural rules, and the other is storing element to make final XML document or store it into database system. For word document converting, we developed OCX based word converting daemon. Helping user to extracting information, we developed script language having native function/variable processing engine extended from XSLT. This system can be used in the area of constructing word document contents DB or providing various information service based on RAW word documents. We really applied it to project management system and project result management system.

• The Journal of the Korea Contents Association
• /
• v.7 no.4
• /
• pp.1-9
• /
• 2007

Cellular operators develop high speed mobile internet and multi-function cellular phones to activate new business model based on mobile internet services. Domestic cellular operators evolve their mobile networks from cdma2000-1x and EvDo to HSDPA to activate high speed mobile internet services. They also develop mobile internet platforms such as WIPI, BREW, and J2ME on multi-function cellular phones having multimedia solutions such as camera, MP3, MPEG, 3D game engine, DMB, PAN such as bluetooth, IrDA, W-LAN, and location information using GPS. But, content providers have problems of redevelopment of the same mobile internet application on different mobile internet platforms provided by cellular operators. In this paper, we develop conversion solutions for interoperability of mobile internet applications on WIPI and BREW using an one-pass compiler. We confirm the performance the proposed conversion solutions for the API conversion rate, the converted file size, and the full conversion time using the popular mobile games which are the killer applications on WiPI and BREW.

• The KIPS Transactions:PartB
• /
• v.17B no.2
• /
• pp.169-176
• /
• 2010

This paper proposes color transformation of images based on user's preference. Traditional color transformation method transforms only hue based on existing templates that define range of harmonious hue. It does not change saturation and intensity. Users would appreciate the resulting images that adjusted unnatural hue of images. Since color is closely related to peoples' emotion, we can enhance interaction of emotion-based contents and technologies. Therefore, in this paper, we define the range of color of each emotion for the transformation of color and perform the transformation of hue, saturation and intensity. However, the relationship of color and emotion depends on the culture and environment. To reflect these characteristics in color transformation, we propose the transformation of color that is based on user's preference and as a result, people would be more satisfied. We adopt interactive genetic algorithm to learn about user's preference. We surveyed the subject to analyze user's satisfaction about transformed images that are based on preference, and we found that people prefer transformed images to original images. Therefore, we conclude that people are more satisfied with the transformation of the templates which reflected user's preference than the one that did not.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.07a
• /
• pp.124-126
• /
• 2012

3차원 동영상은 수많은 응용분야에서 차세대 멀티미디어 콘텐츠로 큰 주목을 받고 있지만 2차원에서 3차원 콘텐츠로의 과도기인 현 시점에서 3차원 콘텐츠의 부족현상은 관련 산업분야의 큰 문제점으로 떠오르고 있다. 2D-to-3D 변환 기술은 높은 비용 없이 기존의 방대한 2차원 콘텐츠를 3차원 콘텐츠로 재사용하는 것이 가능하기 때문에 3차원 콘텐츠의 부족 문제를 해결할 수 있는 기술로 큰 관심을 끌고 있다. 본 논문에서는 블록 단위 초점 정보의 보간을 통한 새로운 초점 기반 3차원 변환 기법을 제안한다. 기존의 화소 단위 초점 측정치를 그대로 깊이 정보로 이용하는 경우나 분할 영역 단위 초점 측정치를 깊이 정보로 활용하는 경우는 이웃 화소간 깊이정보의 불연속성이 발생할 수 있지만 제안방식은 블록 초점 측청치의 보간수행으로 인해 이웃 화소 사이에서 뿐만 아니라 블록경계 혹은 영역경계에서도 특별할 스무딩필터 처리 없이도 화소간 깊이정보의 불연속성이 발생하지 않게 된다. 실험결과를 통해 제안한 방식이 기존의 방식들 보다 더 자연스러운 깊이 정보 추출 성능을 보여준다는 것을 알 수 있다.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.648-653
• /
• 2008

With development of mobile communications terminal and expansion of mobile Internet, a large number of users can use mobile web contents at anytime and anywhere and will demand service of greater contents. Due to such demand, many studies are being carried out on transcoding contents so that wired web contents can be used in mobile web. However, transcoding and creation of mobile web contents involve difficulties because specifications available from telecommunications companies and mobile terminals have not been standardized. Especially, in order for serving dynamic object of wired web page contents to dynamically change according to time or user, it is required not only to program scripts to suit each terminal, but also to transcode the resources used in advance. For solution to this problem, this study uses the hierarchy structure of DOM (document object model) to display structural characteristics of wired web page. In other words, this study proposes the following technique Wired web pages are analyzed and the results are established as data structure. Then, dynamic object is extracted and the domain is indexed so that, when serving mobile web page, information can be extracted at the indexed position to create mobile web contents for service on real-time basis.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.219-220
• /
• 2022

3D 그래픽스 환경에서 입체적인 물체가 2D 모니터에 보여지는 과정에는 여러 단계의 공간 변환과 행렬 계산을 거치게 된다. 이러한 공간 변환은 각 단계가 어떤 의미인지 쉽게 이해하기에 어려운 면이 있다. 본 논문에서는 유니티 3D 엔진에서 작동하는 셰이더를 통해 각 단계의 공간을 시각화 하여 학생들이 보다 쉽게 공간 변환을 이해하는 시각화 방법을 제시한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.39-41
• /
• 2011

본 논문에서는 깊이 영상의 해상도와 품질을 향상시키는 방법을 제안한다. 일반적으로 2D-plus-Depth 구조의 3D 콘텐츠에서는 깊이 영상의 품질이 매우 중요하다. 최근 들어 Time-of-Flight (TOF) 방식의 깊이 센서가 깊이 영상 획득에 많이 사용되고 있는데 TOF 깊이 센서가 제공하는 깊이 영상은 저해상도이기 때문에 고해상도 3D 콘텐츠를 제작하기 위해서는 깊이 영상의 해상도를 상향 변환하는 것이 필수적이다. 또한 고품질의 깊이 영상을 얻기 위해서는 물체 간의 경계를 정교하게 보존하는 것이 중요하다. 최근에는 깊이 영상의 해상도 상향 변환을 위해서 Joint Bilateral Upsampling(JBU) 방식이 많이 사용되고 있다. 본 논문은 깊이 영상의 해상도를 높임에 있어서 우선 보간법을 수행하여 영상의 상향 변환 시에 생긴 빈 홀들의 값을 채워준 후 Bilateral Filtering을 수행함으로써 성능을 높인다. 일반적으로 영상을 상향 변환을 할 때 다양한 방법들이 있는데 본 논문에서는 Nearest Neighborhood(NN), Gaussian과 경계 보존 보간법, 경계 보존 보간법과 Fast Curvature Based Interpolation(FCBI)를 결합한 보간법을 사용하였다. 실험 결과 제안 방법이 기존 방법보다 우수한 성능을 가짐을 보여준다. 또한 경계 보존 보간법과 FCBI를 결합한 보간법을 이용해서 상향 변환을 수행한 결과가 다른 보간법들에 의한 결과보다 우수하다는 점을 알 수 있다.