• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.12
• /
• pp.2241-2248
• /
• 2007

This paper proposes the web-based virtual collaboration among experts. The proposed system supports the presentation tool using Synchronous Multimedia Integration Language(SMIL) which is the easy and efficient way of adding multimedia to presentations. The presentation gives essential information to the participants before actual discuss. The participants use texts in discussing over the presented medical image. The spatial elements such as point or line, and some type of marker with their relative participants' comments can be set or removed dynamically to represent areas of interest in digital images. XML files are used for recording experts' opinions as well as the spatial elements that are associated with digital images during the discussion and stored for future reference. The participants can also set and reset a polygon in the image to select the interested area and refer to the stored relating information.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.3
• /
• pp.171-179
• /
• 2020

Recently, studies have shown that convolution neural networks are achieving the best performance in image classification, object detection, and image generation. Vision based defect inspection which is more economical than other defect inspection, is a very important for a factory automation. Although supervised anomaly detection algorithm has far exceeded the performance of traditional machine learning based method, it is inefficient for real industrial field due to its tedious annotation work, In this paper, we propose ADGAN, a unsupervised anomaly detection architecture using the variational autoencoder and the generative adversarial network which give great results in image generation task, and demonstrate whether the proposed network architecture identifies anomalous images well on MNIST benchmark dataset as well as our own welding defect dataset.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2009.04a
• /
• pp.149-154
• /
• 2009

For the past few decades, residential improvement projects have been mainly carried out after demolition of the original dwelling place. So far, interests and opinions of the existing residents have largely been ignored during the projects. However, citizen participation in local redevelopment has recently been regarded as essential part as progress in democracy and diversified public interests have offered more importance to citizen participation in the implementation of public policies. While the importance of resident participation has been increasingly emphasized in principle, there still has been more to do in its application in reality. We should develop the experience of collecting community opinion to make them reflected in public policy, if we are to achieve the resident and citizen-centered society. The purpose of this study is to develop an image map tool that can be applied to "Maul-Mandulgi" projects as a visualized method to facilitate the exchange of opinions and work toward agreements. The tool is supposed to assist the public discussion by visualizing the policies and reducing the possibility of misunderstanding, so that residents can properly respond to them. In addition, this study will verify the effectiveness of the tool in the application to local community workshops.

• Journal of the Korea Society of Computer and Information
• /
• v.7 no.3
• /
• pp.1-8
• /
• 2002

Due to the fast development in computer application technologies, a video is now being more widely used than ever in many areas. The current information analyzing systems are basically built to process text-based data. Thus, it has little bits Problems when it needs to correctly represent the ambiguity of a video, when it has to process a large amount of comments. or when it lacks the objectivity that the jobs require. We would like to purpose the method that is capable of analyze a large amount of video efficiently. To extract the color, we translate the color from RGB to HSI and use the information that matches with the representative colors. To extract the shape information, we use improved moment invariants(IMI) so that we can solve many problems of histogram intersection.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2014.02a
• /
• pp.313-313
• /
• 2014

GaN계 물질 기반의 광 반도체는 조명 및 디스플레이 관련 차세대 광원으로 많은 관심을 받고 있고, 효율 증대를 위한 에피, 소자 구조 및 패키지 등의 많은 연구가 진행되고 있다. 특히, 투명 전극을 이용한 광 추출 효율의 증가에 대한 연구는 전체 외부양자효율을 증가시키는 중요한 기술로 각광을 받고 있다. 이러한 투명전극은 가시광 영역의 빛을 투과하면서도 전기 전도성을 갖는 기능성 박막 전극으로 산화인듐주석이 널리 사용되고 있으나 인듐 가격의 상승과 산화인듐주석 전극 자체의 크랙 특성으로 인하여 많은 문제점이 지적되고 있다. 이러한 문제를 극복하기 위하여 GaN계 발광 다이오드에 있어서 산화인듐주석 투명 전극의 대체 물질들에 대한 많은 연구들이 활발하게 이루어 지고 있다. 특히, 투명전극 층으로 사용되는 산화인듐주석 대체 박막으로 산화아연에 대한 연구가 각광을 받고 있는 실정이다. 또한, 발광 다이오드의 효율 증가를 위해 발광소자에 표면 요철 구조 형성과 나노구조체 형성 등 박막 표면의 구조 변화를 통한 광추출효율 향상에 대한 많은 연구가 진행되고 있다. 본 연구에서는 산화아연 박막을 투명전극으로 사용하였으며 광추출효율 향상을 위해 산화아연 투명전극에 패터닝을 형성하고, 그 위에 산화아연 나노막대를 형성하여 기존에 사용하던 산화아연 투명전극보다 우수한 추출효율 및 전류 퍼짐 향상 구조를 제안하고 이에 따른 LED 소자의 광추출효율 향상을 연구하였다. 금속유기화학증착법을 이용하여 c-면 사파이어 기판에 n-GaN, 5주기의 InGaN/GaN 다중양자우물 구조 및 p-GaN의 간단한 LED구조를 성장한 후, p-GaN층 상부에 원자층 증착법을 이용하여 투명전극인 산화아연 박막을 60 nm 두께로 증착하였다. 산화아연 투명전극만 증착한 LED-A와 이후 0.1% HCl을 이용한 습식식각을 통하여 산화아연 투명전극에 육각형 모양의 패턴을 형성한 LED-B, 그리고 LED-B위에 전기화학증착법을 이용하여 $1.0{\mu}m$의 산화아연 나노 막대를 증착한 LED-C를 제작하였다. LED-A, -B 및 -C에 대한 표면 구조는 SEM이미지를 통하여 확인한 바 산화아연의 육각 패턴과 그 상부에 산화아연의 나노막대가 잘 형성된 것을 확인하였다. I-L 분석으로부터 패턴이 형성되지 않은 산화아연 투명전극으로만 구성된 LED-A에 비하여 산화아연 투명 전극에 육각 패턴을 형성한 LED-B의 전계 발광 세기가 더욱 큰 것을 확인하였다. 또한, 육각 패턴에 산화아연 나노막대를 성장시켜 융합구조를 형성한 LED-C에서는 LED-B와 -A보다 더 큰 전계 발광세기를 확인할 수 있었다. 특히, 인가 전류가 고전류로 갈수록 LED-C의 발광세기가 더욱 강해지는 것으로 효율저하현상 또한 나노융합구조의 LED-C에서 확인할 수 있었다. 이는 기존 산화아연 투명전극에 육각형의 패턴 및 나노막대융합구조를 형성할 경우 전류퍼짐현상을 극대화 할 뿐 아니라, 추가적인 광추출효율 향상 효과에 의해 질화갈륨 기반LED 소자의 광효율이 증가된 것으로 판단된다.

• Journal of Korea Multimedia Society
• /
• v.8 no.4
• /
• pp.450-460
• /
• 2005

Today, multimedia information is used on the internet and various social areas by rapid development of computer and communication technology. Therefor, the usage is growing dramatically. Multimedia information analysis system is basically based on text. So, there are many difficult problems like expressing ambiguity of multimedia information, excessive burden of works in appending notes and a lack of objectivity. In this study, we suggest a method which uses color and shape information of multimedia image partitions efficiently analyze a large amount of multimedia information. Partitions use field growth and union method. To extract color information, we use distinctive information which matches with a representative color from converting process from RGB(Red Green Blue) to HSI(Hue Saturation Intensity). Also, we use IMI(Improved Moment Invariants) which target to only outline pixels of an object and execute computing as shape information.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.239-242
• /
• 2000

급속한 컴퓨터 및 통신 기술의 발전으로 인해 동영상 정보는 인터넷 및 사회전반의 다양한 분야들에서 활용되고, 기하급수적으로 증가되고 있으나, 기존의 정보 분석 시스템은 기본적으로 텍스트를 기반으로 한 것이기에, 동영상 정보가 가지는 애매성을 표현하기 곤란하며, 주석 작성에 따르는 과다한 작업부담 및 객관성 결여들의 문제점을 가지고 있다. 이에, 상기에 기술한 문제점을 해결하기 위하여 대용량의 동영상 정보를 효율적으로 분석 할 수 있는 동영상 분석 알고리즘을 제안하고자 한다. 본 논문의 주된 연구과제는 컷 검출을 위한 통합 특징량 적용법과 색상특징, 모양특징을 사용하여 이미지 내의 특징 정보를 분석하는 것이다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.637-642
• /
• 2020

패션산업은 매년 1 조원씩 성장(연평균 2.1%)하며 많은 연구자들의 관심을 받고 있다. 전통적인 패션산업은 점차 디지털화되어 선진적인 컴퓨터 비전 기술을 적용해 소비자들에게 더 좋은 쇼핑 서비스를 제공하고 있다. 본 논문에서는 2014 년부터 2019 년 사이에 구축된 대표적인 패션 데이터셋을 연도별로 정리하고 각 데이터셋에 포함된 주석(annotation)의 특징을 정리했다. 또한 데이터셋이 패션 상품 검출(Fashion detection), 패션 이미지 생성(Fashion image generation), 가상 피팅(Virtual try-on) 그리고 패션 의류 분할(Fashion Clothing segmentation) 등 연구에서의 활용될 수 있는 여부에 대해 분석했다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.227-233
• /
• 2019

Recently, along with the recent development of deep learning technique, neural networks are achieving success in computer vision filed. Convolutional neural network have shown outstanding performance in not only for a simple image classification task, but also for tasks with high difficulty such as object segmentation and detection. However many such deep learning models are based on supervised-learning, which requires more annotation labels than image-level label. Especially image semantic segmentation model requires pixel-level annotations for training, which is very. To solve these problems, this paper proposes a weakly-supervised semantic segmentation method which requires only image level label to train network. Existing weakly-supervised learning methods have limitations in detecting only specific area of object. In this paper, on the other hand, we use multi-classifier deep learning architecture so that our model recognizes more different parts of objects. The proposed method is evaluated using VOC 2012 validation dataset.

• Journal of radiological science and technology
• /
• v.28 no.1
• /
• pp.19-24
• /
• 2005

The purpose of this study was to design and implement a useful annotation-based Radiological image retrieval system to accurately determine on education and image information for Radiological technologists. For better retrieval performance based on large image databases, we presented an indexing technique that integrated $B^+-tree$ proposed by Bayer for indexing simple attributes and inverted file structure for text medical keywords acquired from additional description information about Radiological images. In our results, we implemented proposed retrieval system with Delphi under Windows XP environment. End users, Radiological technologists, are able to store simple attributes information such as doctor name, operator name, body parts, disease and so on, additional text-based description information, and Radiological image itself as well as to retrieve wanted results by using simple attributes and text keywords from large image databases by graphic user interface. Consequently proposed system can be used for effective clinical decision on Radiological image, reduction of education time by organizing the knowledge, and well organized education in the clinical fields. In addition, It can be expected to develop as decision support system by constructing web-based integrated imaging system included general image and special contrast image for the future.