• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.242-245
• /
• 2006

This paper addresses an image fingerprinting scheme for the print-to-capture model performed by a photo printer and digital camera. When capturing an image by a digital camera, various kinds of distortions such as noise, geometrical distortions, and lens distortions are applied slightly and simultaneously. In this paper, we consider several steps to extract fingerprints from the distorted image in print-and capture scenario. To embed ID into an image as a fingerprint, multi-bits embedding is applied. We embed 64 bits ID information as a fingerprint into spatial domain of color images. In order to restore a captured image from distortions a noise reduction filter is performed and a rectilinear tiling pattern is used as a template. To make the template a multi-bits fingerprint is embedded repeatedly like a tiling pattern into the spatial domain of the image. We show that the extracting is successful from the image captured by a digital camera through the experiment.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.7
• /
• pp.677-684
• /
• 2016

With the increase of multimedia information such as image, researches on extracting high-level semantic information from low-level visual information has been realized, and in order to automatically generate this kind of information. Various technologies have been developed. Generally, image retrieval is widely preceded by comparing colors and shapes among images. In some cases, images with similar color, shape and even meaning are hard to retrieve. In this article, in order to retrieve the object in an image, technical value of middle level is converted into meaning value of middle level. Furthermore, to enhance accuracy of segmentation, K-means algorithm is engaged to compute k values for various images. Thus, object retrieval can be achieved by segmented low-level feature and relationship of meaning is derived from ontology. The method mentioned in this paper is supposed to be an effective approach to retrieve images as required by users.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.632-639
• /
• 2021

In this paper, we proposed a reversible data hiding technique that greatly enhances the security of confidential data by encrypting confidential data and then spatially encrypting the encrypted confidential data and hiding it in the cover image. When a result image is generated by hiding the encrypted confidential data in the cover image using a spatial encryption technique, the quality of the result image is very good, and the original cover image and the result image cannot be visually distinguished. Since the encrypted confidential data is spatially encrypted and concealed, it is not possible to know where the encrypted confidential data is concealed in the result image, and the encrypted confidential data cannot be extracted from the result image. Even if the encrypted confidential data is extracted, the original confidential data is not known because the confidential data is encrypted. Therefore, if confidential data is concealed in images using the proposed technique, the security of confidential data is greatly improved. The proposed technique can be effectively used in medical and military applications.

• Annual Conference of KIPS
• /
• 2020.11a
• /
• pp.937-940
• /
• 2020

본 논문은 AI 기술을 기반으로 텍스트 스크립트를 자동으로 인식하고 영상 합성 기술을 응용하여 텍스트 정보를 시각화하는 AI 아나운서 소프트웨어 연구에 대하여 기술한다. 기존의 AI 기반 영상 정보 전달 서비스인 AI 앵커는 텍스트를 인식하여 영상을 합성하는데 오랜 시간이 필요하였으며, 특정 인물 이미지로만 영상 합성이 가능했기 때문에 그 용도가 제한적이었다. 본 연구에서 제안하는 방법은 Tacotron 으로 새로운 음성을 학습 및 합성하여, LRW 데이터셋으로 학습된 모델을 사용하여 자연스러운 영상 합성 체계를 구축한다. 단순한 얼굴 이미지의 합성을 개선하고 다채로운 이미지 제작을 위한 과정을 간략화하여 다양한 비대면 영상 정보 제공 환경을 구성할 수 있을 것으로 기대된다.

• Annual Conference of KIPS
• /
• 2024.05a
• /
• pp.586-589
• /
• 2024

본 실험 연구에서는 주의 메커니즘과 컨볼루션 신경망을 결합하여 모델을 개선하는 방법을 탐색하는 딥 러닝 기술을 소개한다. 이 기술은 지도 학습 방식을 위해 공개 데이터 세트의 쓰레기 분류 데이터를 사용하고, Grad-CAM 기술과 채널 주의 메커니즘 SE 를 적용하여 모델의 분류 의사 결정 과정을 더 잘 이해하기 위해 히트 맵을 생성한다. Grad-CAM 기술을 사용하여 히트 맵을 생성하면 분류 중에 모델이 집중하는 영역을 시각화할 수 있다. 이는 모델의 분류 결정을 설명하는 방법을 제공하여 다양한 이미지 카테고리에 대한 모델 결정의 기초를 더 잘 이해할 수 있다. 실험 결과는 전통적인 합성곱 신경망과 비교하여 제안한 방법이 쓰레기 분류 작업에서 더나은 성능을 달성한다는 것을 보여준다. 주의 메커니즘과 히트맵 해석을 결합함으로써 우리 모델은분류 정확도를 향상시킬 수 있다. 이는 실제 응용 분야의 이미지 분류 작업에 큰 의미가 있으며 해석 가능성에 대한 딥 러닝 연구 진행을 촉진하는 데 도움이 된다.

• Management & Information Systems Review
• /
• v.28 no.4
• /
• pp.175-197
• /
• 2009

Numerous researches related to internet shopping malls have been researched by numerous researchers. Such as, establishing criteria to evaluate service quality of electronic commerce, identifying factors affecting successful internet shopping mall operation, and examining relationships among the factors in electronic commerce based shopping mall needed to be focused on. However, most researches that have been undertaken only consider the service quality model or basic functional dimension. In accordance with this indispensability, the integrated structural relationship among variables, which are potentially inherent in customer's perception and affect personnel royalty on internet shopping mall needs to be acknowledged. The purpose of this study is to examine which factors should be able to facilitate performance of internet shopping mall. Based on the relevant literature, it has been empirically analyzed how corporate image, system quality, service quality and delivery quality affect customer satisfaction as well as customer royalty. The research's problem is that it was tested with data collected from 212 respondents. This study developed and empirically analyzed a model representing the relationship by using the Structural Equation Model. The major findings of this study are, firstly, that the higher corporate image is positively affecting the system quality and delivery quality. Secondly, the higher delivery quality is positively affecting the service quality. Thirdly, the higher service quality is positively affecting the customer satisfaction. Finally, the higher customer satisfaction is affecting the customer royalty.

• Archives of design research
• /
• v.17 no.3
• /
• pp.51-60
• /
• 2004

The esthetics of 'grotesque' and its entertaining values have many things in common with the characteristics and properties of animation as a medium of expression. The ways to express grotesque images in animation have expanded its realm of expression with the development of modern image digitalization technologies. The grotesque images that were expressed by the combination of photographs and animations and objet animations have been developing its disharmony in a more realistic manner with full 3D animation technology and realistic motions of virtual characters with motion capture. The subject matter also has become more diverse with the advance of modern scientific civilization. An example is the appearance of Cyborg(cybernetic+organism) characters and the grotesque images in the modern times mainly start to appear in animations dealing with the isolation of human beings among its characteristics. Also, the changes of the expression methods of grotesque images along with the development of visual technology development in the future will transform its methodological forms by changing its expression methods to VR forms and interactive visuals. Thus it is expected that the grotesque images and characteristics will change into a visual genre which emphasizes dehumanization.

• Annual Conference of KIPS
• /
• 2001.04b
• /
• pp.661-664
• /
• 2001

본 논문에서는 이동단말에 최적화된 서비스를 제공하기 위한 이동단말의 정보 이용방법과 무선선로상의 제약 및 무선 단말기의 제한된 자원으로 좀더 효율적인 무선 인터넷 서비스를 제공하기 위한 WBMP(Wireless BitMaP) 변환기의 설계 및 구현에 대하여 기술한다. 이동단말의 정보는 CC/PP(Composite Capabilities/Preferences Profiles) 또는 WSP(Wireless Session Protocol) 헤더로부터 알아낼 수 있으며 이 정보를 이용하여 WBMP 변환기는 기존의 이미지를 이동단말에 가장 적절한 WBMP 이미지로 변환한다. 그러나 컬러 이미지를 변환하는 경우 이동단말이 가지는 제약에 의해 이미지 정보가 손실되거나 왜곡될 수 있으며, 현재 무선환경에서 지원되는 0/1 형식의 WBMP 이미지로는 각 단말의 자원을 최대한 활용할 수 없다. 따라서 향후 그레이-레벨 또는 컬러 이미지로의 변환 기능이 추가 구현되어야 한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.415-417
• /
• 2004

저작권보호를 목적으로 사용하는 워터마킹 기술은 원본 이미지에 특정 워터마크 정보를 삽입하는데 이러한 경우, 원본이미지는 다시 복구할 수 없다. 의료영상 데이터베이스나 군사용 이미지 시스템과 같은 특정된 응용에서는 이미지 열화가 시스템 전체에 미치는 영향이 크기 때문에 이러한 워터마크 알고리즘을 사용할 수 없다. 본 연구에서는 Hadamard Matrix(HM)을 이용하여 디지털 이미지에 워터마크를 삽입하며, 가역적인 연산에 기초하여 원 이미지의 복구가 가능하도록 워터마크 알고리즘을 개발한다. 본 연구에서는 Hadamard Coefficients의 규칙적인 연관성을 이용하여 2D HM을 알고리즘 개발에 적용한다.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.1
• /
• pp.389-393
• /
• 2020

There is a growing interest and importance in local-based products, especially agri-food globally, but systematic research on this issues is still insufficient. Recently, consumer interest in agri-foods is local based product and local brand. The research began on the relationship between local food and local food brands. This study examined the effect of local food products and brand image congruence on brand preference and purchase intention of local food. The results showed that the stronger the image congruence between local food products and brands, the more positively the brand preference. In addition, the stronger the brand preference, the higher the customer's intention to repurchase.