• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.1237-1239
• /
• 2011

본 논문에서는 다양한 상황에서의 정보 접근성을 향상시키기 위해 이미지를 검색 수단으로 사용하는 검색 시스템을 제안한다. 본 콘텐츠는 안드로이드 플랫폼 기반의 스마트폰에서 이미지를 얻어 텍스트를 추출하고, 이를 검색 엔진의 키워드로 입력하여 그 결과를 출력하는 과정을 거친다. 또한, 검색 결과를 스마트폰의 내장 데이터베이스에 저장하고, 이를 관리하여 추후에 재사용할 수 있도록 한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.218-220
• /
• 2021

본 논문에서는 CAR(content adaptive resampler)로 축소된 저해상도 이미지를 직접 다른 모델에 여러가지 방식으로 훈련을 시켜 성능을 개선시키고자 하였다. 본 논문에서는 단일 영상 super resolution 에 관하여 여러 기술이 존재하는 상황에 더 나은 기술을 테스트하려 하고 그를 위해 과거의 모델들에 대한 이해가 필요하여 이를 구현하였다. 현재 가장 뛰어난 성능을 보이고 있는 모델 중의 하나인 CAR 에서 복원 전 이미지를 사용하여 훈련을 시키면 더 나은 성능의 모델을 만들 수 있을 것이라고 가정하고 다양한 훈련을 통해 성능을 개선시키고자 하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.376-378
• /
• 2022

Artificial intelligence technology is widely used in various fields such as artificial intelligence speakers, artificial intelligence chatbots, and autonomous vehicles. Among these AI application fields, the image processing field shows various uses such as detecting objects or recognizing objects using artificial intelligence. In this paper, data synthesized by a virtual human is used as a method to analyze images taken in a specific space.

• Journal of Platform Technology
• /
• v.12 no.1
• /
• pp.128-140
• /
• 2024

'Deepfake' refers to a video synthesis technique that utilizes various artificial intelligence technologies to create highly realistic fake content, causing serious confusion to individuals and society by being used for generating fake news, fraud, malicious impersonation, and more. To address this issue, there is a need for methods to detect malicious images generated by deepfake accurately. In this paper, we extract and analyze saliency features from deepfake and real images, and detect candidate synthesis regions on the images, and finally construct an automatic deepfake detection model by focusing on the extracted features. The proposed saliency feature-based model can be universally applied in situations where deepfake detection is required, such as synthesized images and videos. To demonstrate the performance of our approach, we conducted several experiments that have shown the effectiveness of the deepfake detection task.

• Korean Journal of Environment and Ecology
• /
• v.23 no.4
• /
• pp.309-316
• /
• 2009

The usefulness of media representations for assessing normative standards depends in part on how accurately media reflect "on-the-ground" resource conditions. This study compared encounter standards based on still and moving pictures to assess movement effects. The study location was the Jungmoeri area of Mudeungsan Provincial Park (MPP) in Korea. A total of 50 college students participated in a laboratory experiment where they evaluated still and moving pictures constructed using Photoshop and Flash computer programs. For the maximum acceptable number (MAN), however, there was no significant difference of ratings between still and moving pictures, and the overall encounter norm curves were nearly identical. There were some "method findings" for ordering effects and percent of people moving, but for a resource manager developing standards there was no advantage to the more complex logistics of using moving pictures to assess this particular impact. The trade-offs of using more sophisticated media are discussed, and more research is needed to further explore factors such as movement of sound in evaluation of other resource conditions.

• Journal of Digital Convergence
• /
• v.10 no.6
• /
• pp.297-302
• /
• 2012

In this paper, we present a object detection and tracking method based on image context analysis. It is robust from the image variations such as complicated background, dynamic movement of the object. Image context analysis is carried out using the hybrid network of k-means and RBF. The proposed object detection employs context-driven adaptive Bayesian framework to relive the effect due to uneven object images. The proposed method used feature vector generator using 2D Haar wavelet transform and the Bayesian discriminant method in order to enhance the speed of learning. The system took less time to learn, and learning in a wide variety of data showed consistent results. After we developed the proposed method was applied to real-world environment. As a result, in the case of the object to detect pass outside expected area or other changes in the uncertain reaction showed that stable. The experimental results show that the proposed approach can achieve superior performance using various data sets to previously methods.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.127-133
• /
• 2020

This paper deals with the development of a smart mirror to support changing hair styles. A key function of the service is the ability to synthesize the image into the user's face when the user chooses a desired hair image and virtually styling the hair. To check the effectiveness of the hair image synthesis function, the success rate measurement experiment of Haar-cascade algorithm's facial recognition was conducted. Experiments have confirmed that the facial recognition succeeds with a 95 percent probability, with both eyes and eyebrows visible to the subjects. It is the highest success rate. It confirmed that if either of the eyebrows of the subjects are not visible or one eyeball is covered, the success rate of facial recognition is 50% and 0% respectively.

• The Journal of the Korea Contents Association
• /
• v.14 no.2
• /
• pp.60-71
• /
• 2014

Images of TV that crossed the reality & the virtual reality have been discussed for a long time, Then, With what feature can TV make messages of the reality & the virtual reality maintain to achieve such a target, and in which situation will the feature be exposed? One who answered first about these questions is a German critic, G$\ddot{u}$nter Anders. Anders called the reality confused with the reality and the confused world of the virtual reality Phantom, and thought human beings would live in the reproduced world created by media machine in the long run as a world that the Phantom created would be gloomy and confusing. The media, themselves, become images, and an image creates another image. Through this process, human beings became unaware of which image the reality was indeed. As TV often created these situations, we have been already seduced into the deep Phantom world before discussing right or wrong of the situation. Of course, the media reality & the reality can't be strictly distinguished. Because the means that help to form judgement of viewers is the media. The subject of practical judgement is the media reality not human beings.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.433-436
• /
• 2009

유비쿼터스 컴퓨팅 환경으로 진화하면서 사용자의 상황정보를 이해하고 사용자의 상황에 적합한 서비스를 제공하고자 하는 연구들이 다양하게 진행되고 있다. 이러한 상황정보 서비스에 사용자의 의도를 반영함으로써 정확한 서비스 제공을 가능하게 하면서도 사용자가 서비스의 내용을 알 수 있도록 하기 위하여 Public Code라는 개념을 정의한다. 이를 이용하여 사용자의 의도를 해석하여 서비스를 시작하고, 의미 있는 서비스 제공을 위하여 사용자의 상황정보를 조합하여 이용함으로써 사용자 입장에서 직관적이면서도 편리한 서비스를 제공하기 위한 서비스를 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.620-621
• /
• 2013

본 논문에서는 스마트기기의 영상, 음성, 이미지 등의 다양한 멀티미디어 데이터 수집 기능을 활용하여 응급상황 발생 시 현장상황의 정확한 정보를 유관기관에 제공함으로써 효율적인 응급상황 대처가 가능하고, 신고자 인증 모듈을 활용하여 허위신고 문제를 개선한 응급상황 대처를 위한 스마트 세이프티 시스템을 설계하였다.