• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.6
• /
• pp.692-704
• /
• 2002

In this paper, we present am image retrieval method based on color-spatial features using quadtree and texture information extracted from object MBRs in an image. Tile proposed method consists of creating a DC image from an original image, changing a color coordinate system, and decomposing regions using quadtree. As such, conditions are present to decompose the DC image, then the system extracts representative colors from each region. And, image segmentation is used to search for object MBRs, including object themselves, object included in the background, or certain background region, then the wavelet coefficients are calculated to provide texture information. Experiments were conducted using the proposed similarity method based on color-spatial and texture features. Our method was able to refute the amount of feature vector storage by about 53%, but was similar to the original image as regards precision and recall. Furthermore, to make up for the deficiency in using only color-spatial features, texture information was added and the results showed images that included objects from the query images.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.2
• /
• pp.143-148
• /
• 2012

There are many researches on segmentation of sidescan sonar image to recognize or classify the underwater objects. Although existing algorithms's performance is good in detecting object's shadow and reducing the underwater noise, the computing time is very low. In this paper we try to separate shadow from background and segment the underwater image by using morphological method using background's noise distribution characteristics and object's shadow charateristics. This algorithm is useful when the average of background is lower than the average of the shadow, because this is adjusted from the background's chracteristics. Results shows that the algorithm works fine in multiple object environments and the computing time is reduced to 1 second.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.6
• /
• pp.19-27
• /
• 2010

An region of interest processing technique is to handle preferentially some part of an image dynamically according to region of interest of the users in JPEG2000 image. A small image is not important, but in a big image the specified region that the user indicated has to be handled preferentially because it takes long time to display the whole image. If the user indicates a region of the outline image, the browser masks the region and sends the mask information to the source that transmitted the image. The server which got the mask information preferentially sends the code blocks matching the masks. Here, quickly generating mask information is important, so, in this paper using predefined 48 mask patterns, selecting one of the patterns according to the distribution of ROI(Region-of-Interest) and background, we remarkably reduced the time computing the mask region. Blocks that the patterns are applied are the blocks mixed of ROI and background in a block. If a whole block is an ROI or a background, these patterns are not applied. As results, comparing to the method that precisely handles ROI and background, the quality is unsatisfactory but the processing time remarkably reduced.

• Journal of the Korea Convergence Society
• /
• v.10 no.3
• /
• pp.15-22
• /
• 2019

In this paper, we proposed a Hangeul detection method using structural features of histogram, consonant, and vowel to solve the problem of Hangul which is separated and detected consonant and vowel The proposed method removes background by using DoG (Difference of Gaussian) to remove unnecessary noise in Hangul detection process. In the image with the background removed, we converted it to a binarized image using a cumulative histogram. Then, the horizontal position histogram was used to find the position of the character string, and character combination was performed using the vertical histogram in the found character image. However, words with a consonant vowel such as '가', '라' and '귀' are combined using a structural characteristic of characters because they are difficult to combine into one character. In this experiment, an image composed of alphabets with various backgrounds, an image composed of Korean characters, and an image mixed with alphabets and Hangul were tested. The detection rate of the proposed method is about 2% lower than that of the K-means and MSER character detection method, but it is about 5% higher than that of the character detection method including Hangul.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.3C
• /
• pp.92-101
• /
• 2005

It is difficult to detect the accurate detection which leads the camera it moves follows in change of the noise or illumination and Also, it could be recognized with backgound if the object doesn't move during hours. In this paper, the proposed method is updating changed background image as much as N＊M pixel mask as time goes on after get a difference between imput image and first background image. And checking image pixel can efficiently detect moving by computing fixed distance pixel instead of operate all pixel. Also, set up minimum area of object to use boundary point of object abstracted through checking image pixel and motion detect of object. Therefore motion detection is available as is fast and correct without doing checking image pixel every Dame. From experiment, the designed and implemented system showed high precision ratio in performance assessment more than 90 percents.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.5
• /
• pp.31-38
• /
• 2024

In this paper, we propose an ASAP(AI-driven Service for Advertisement Production) application that provides a generative AI-based automatic advertising image production service. This application utilizes GPT-3.5 Turbo Instruct to generate suitable background mood and promotional copy based on user-entered keywords. It utilizes OpenAI's DALL·E 3 model and Stability AI's SDXL model to generate background images and text images based on these inputs. Furthermore, OCR technology is employed to improve the accuracy of text images, and all generated outputs are synthesized to create the final advertisement. Additionally, using the PILLOW and OpenCV libraries, text boxes are implemented to insert details such as phone numbers and business hours at the edges of promotional materials. This application offers small business owners who face difficulties in advertising production a simple and cost-effective solution.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1997.11a
• /
• pp.321-339
• /
• 1997

인터넷의 활용과 컴퓨터 기술환경의 새로운 패러다임 변화는 디지털도서관의 등장을 가져오게 하였으며, 동시에 이미 온라인 서비스를 실시해오던 여러 분야의 정보센터들로 하여금 어떠한 유형의 데이터도 저장, 관리, 제공하며 정보라는 상품을 생산해 내는 정보시장(Information Marketplace)의 역할 자로서 요구하고 있다. 본 논문에서는 1980년대부터 진행되어온 도서관 서지자료의 Database화 작업과 병행하여 도서관 이용자에게 보다 광범위하고 효용성 있는 정보를 제공하기 위한 디지털도서관의 이론적 배경에 대하여 기술하고, 디지털도서관의 주요 소장정보인 원문이미지정보의 처리기술에 대한 이론을 정리하며, 현재국내에서 추진중인 원문이미지정보시스템의 분석을 통하여 향후 디지털도서관 구축시 비교분석 모델을 제시해보고자 한다.

• Proceedings of the Korea Contents Association Conference
• /
• 2013.05a
• /
• pp.161-162
• /
• 2013

본 논문에서는 Mental ray 렌더러를 통해 렌더링 된 이미지의 색감과 사실적인 빛의 감쇠효과를 위해 사용되는 Color Management를 활용한 Linear Work Flow 작업과정을 제시하고 그 이론적인 배경을 설명하고자 한다. 디지털로 사용되는 대부분의 이미지 포맷은 모든 모니터에서 동일한 색감을 표현하기 위해 sRGB라는 표준화된 형식을 사용하고 있다. 하지만 Mental ray는 기본적으로 감마 커브가 1.0으로 이루어진 이미지를 렌더링한다. sRGB가 가지고 있는 2.2의 감마커브와 차이가 발생함에 따라 사용자가 원하는 색감을 얻어내기 어렵다. 반면에 Color Management를 활용해 렌더링 된 이미지의 감마 커브를 수정해주면 3D 소프트웨어로 사용 된 텍스쳐의 원본 색감을 렌더링 후에도 유지할 수 있다. 또한 라이팅의 감쇠(Decay)와 산란에도 영향을 주어 보다 풍부한 색감의 이미지를 얻게 된다. 본 연구에서는 Autodesk Maya의 Mental Ray를 이용해 실험을 하고 Color Management를 활용한 이미지와 그렇지 않은 이미지를 비교 분석한다. 연구의 범위는 제품 렌더링을 위한 스튜디오 라이팅 기법으로 한정하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.359-361
• /
• 2000

본 논문에서는 비디오 자막 이미지를 인식하기 위해 필요한 영상 향상의 단계로서 다중 결합을 적용한다. 또한 다중 결합을 위한 동일한 자막의 판단 및 결합된 결과를 재평가하기 위한 방법을 제안한다. 입력된 칼라 이미지로부터 RLS(Run Length Smearing)가 적용된 에지 이미지를 얻고, 수직 및 수평 히스토그램 분포를 이용하여 자막과 자막 영역에 대한 정보를 추출한다. 프레임 내의 자막 영역의 중첩 정도를 이용하여 동일 자막을 판단하고, 동일한 자막을 갖는 프레임들끼리 다중 결합을 수행함으로써 향상된 이미지를 얻는다. 끝으로 결합된 영상에 대한 평가를 수행하여 잘못 결합된 이미지들로 인한 오류를 해결하고 재평가한다. 제안한 방법을 통해, 배경 부분의 잡영이 완화된 자막 이미지를 추출하여 인식의 정확성과 신뢰성을 높일 수 있었다. 또한 동일한 자막의 시작 프레임과 끝 프레임의 위치 파악은 디지털 비디오의 색인 및 검색에 효과적으로 이용될 수 있을 것이다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06d
• /
• pp.444-448
• /
• 2008

The ubiquitous smart home is the home of the future that takes advantage of context information from the human and the home environment and provides an automatic home service for the human. Human location and motion are the most important contexts in the ubiquitous smart home. We present a real-time human tracker that predicts human location and motion for the ubiquitous smart home. We used four network cameras for real-time human tracking. This paper explains the real-time human tracker's architecture, and presents an algorithm with the details of two functions (prediction of human location and motion) in the real-time human tracker. The human location uses three kinds of background images (IMAGE1: empty room image, IMAGE2:image with furniture and home appliances in the home, IMAGE3: image with IMAGE2 and the human). The real-time human tracker decides whether the human is included with which furniture (or home appliance) through an analysis of three images, and predicts human motion using a support vector machine. A performance experiment of the human's location, which uses three images, took an average of 0.037 seconds. The SVM's feature of human's motion recognition is decided from pixel number by array line of the moving object. We evaluated each motion 1000 times. The average accuracy of all the motions was found to be 86.5%.