• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.2
• /
• pp.162-171
• /
• 1991

This paper presents an algorithm for compressing image data by separating the image into two parts. I.e. edge image containing high-frequency components and non-edge image containing low-frequency components of image. The edge image is extracted by using 8 level compass gradient masks and the non-edge image is obtained by removing the edge image from the original image. The edge image is coded by Huffman run-length code and the non edge image is transformed first by DCT and the transformed images is coded next by a quantized bit allocation table. For an example image. GIRL. the proposed algorithm shows bit rate of 0.52 bpp with PSNR of 36dB.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.5
• /
• pp.191-199
• /
• 1993

Investigation is given to the detailed procedure of a computer assisted automatic technique for ortho-image generation from digital stereo image data of close-range photographs scanned by the CCD camera scanner. After rectification of geometric scanning errors, the bundle adjustment technique was used to determine the exterior orientation parameters of terrestrial camera. An automatic correlation matching technique was applied to search for the conjugate pixels in digital stereo pairs. And the 3-dimensional coordinates of the corresponding pixels were calculated by the space intersection method. For the generation of ortho-image from the calculated coordinates and right image data values, inverse-weighted-distance average method was used. And the accuracy of the resulting ortho-image was checked by comparing its image coordinates with there corresponding ground coordinates for the check points.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.2 no.4
• /
• pp.619-626
• /
• 1998

This paper presents the features for quantitative evaluation of medical images based on the technique of image processing. In consideration of advice from medical doctor, 1 derive three features seemed to be strongly correlated with the degree of disease's advance. From comparison between each feature and evaluation value by medical doctor, our research shows that three features can be a useful aids in quantitative evaluation of lung's disease on chest X-ray images.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.2
• /
• pp.113-117
• /
• 1998

The image recognition, with various technics, recently presented an effective recognition scheme both for image and character. The analyzing process, however, is highly complicated so that this does not utilized fully. This paper examines the validity of the image recognition through morphology, which is simple and a theory of sets. The morphology applies erosion and dilation, opening and closing, and structuring element.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1996.10a
• /
• pp.61-65
• /
• 1996

본 논문은, 공간영역에서 화상을 압축할 수 있는 Quad-tree 부호화법을 분석하고, 보다 화질 및 압축율을 개선하기 위하여 적응 불록분할 및 병합 알고리듭을 제안하엿다. 화상은 에지부분을 제외하고는 인접한 화소들간에 데이터의 용장도가 높으므로 이 영역을 하나의 대표값으로 설정하여 그 값과 그 블록의 위치좌표를 부호화할 수 있다. Quad-tree 분할은 초기의 병합을 제외하고 순차적으로 분할과정만 반복처리하지만 본 알고리듬에서는 단위블록(3$\times$3 호소) 의 평균잘류에너지(MRE)를 이용하여 블록의 분할과 병합을 반복처리한다. 시뮬레이션결과, 본 알고리듭은 압축율 1bit/pixel에서 기존의 Quad-tree 방법보다 PSNR에서 1.0dB의 개선이 있었으며, 화상의 블록화 현상도 전혀 나타나지 않았다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.255-258
• /
• 2000

정보 산업의 발달에 의해 여러 가지 형태의 정보 전달이 가능하게 되었으며 최근에 칼라 화상의 정보 전달에 있어서 빠른 전송과 압축기법이 필요하게 되었다. 본 논문에서는 칼라 화상을 압축하는데 있어 개선된 SOM 알고리즘을 이용하여 칼라 이미지에 대한 벡터 양자화 기법을 제안한다. 제안된 방법은 기존의 LBG 알고리즘을 이용한 벡터 양자화 기법에 비해 블록화 현상을 줄일 수 있었으며 이미지 전체에 대해 블록의 수만큼 계속해서 반복하지 않고 동적으로 코드북을 생성시킴으로써 실행 시간도 줄일 수 있었다. 또한 웨이블릿을 칼라 화상에 적용시켜 화상의 특징을 더욱더 두드러지게 함으로써 개선된 SOM을 적용시 재생의 효과를 높일 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.761-763
• /
• 2022

코로나19 이후 화상회의 빈도가 높아지면서 줌 피로라는 신조어가 등장할 만큼 상대방을 가까이 마주하며 회의를 진행하는 것이 사람들의 피로도를 상승시키고 있다. 본 논문에서는 얼굴 합성과 이미지 애니메이션을 이용한 아바타를 통해 사용자가 화상회의에 참가할 수 있는 시스템을 제안한다. 사용자와 닮은 개성 있는 캐릭터는 실시간으로 사용자의 표정 및 움직임을 반영하여 화상회의에 적용될 수 있고 채팅과 커뮤니티에서 캐릭터의 이모티콘으로 감정을 표현할 수 있다.

• Journal of the Korea Concrete Institute
• /
• v.21 no.2
• /
• pp.121-127
• /
• 2009

The failure behavior of RC structure was exceedingly affected by the size and the local strain distribution of the failure zone due to the strain localization behavior on the tension softening materials. However, it is very difficult to quantify and assess the local strain occurring in the failure zone by the conventional test method. In this study, image processing technology, which is available to measure the strain up to the complete failure of RC structures, was used to estimate the local strain distribution and the size of failure zone. In order to verify the reliability and validity for the image processing technology, the strain transition acquired by the image processing technology was compared with strain values measured by the concrete gauge on the uniaxial compressive specimens. Based on the verification of image processing technology for the uniaxial compressive specimens, the size and the local strain distribution of the failure zone of deep beam was measured using the image processing technology. With the results of test, the principal tensile/compressive strain contours were drawn. Using the strain contours, the size of the failure zone and the local strain distribution on the failure of the deep beam was evaluated. The results of strain contour showed that image processing technology is available to assess the failure behavior of deep beam and obtain the local strain values on the domain of the post-peak failure comparatively.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.4
• /
• pp.1132-1140
• /
• 1997

In many vision tasks of the major obstacles is the specular highkight of smoth objects, which causes a misinterpretation of objects.This paper presents an dffcient algorithm for highight detection and object reconstruction, blsed on the theory of photometric stereo in which the location of highilight changes as the position of illumination source changes.Two images, referred to as base image and reference image.are sequentially taken with two different positionhs of the two images.The difference image is thresholded to detct the specular spike of the highlight.Then the specu-lar lobe around the specular spike is detected to reconstruct the object.The proposed algorithm can be applied to metals and dielectrics, regardlless of the surface chracteristics.This method can also be aplied to the case when the background is brighter than the object.

• The KIPS Transactions:PartD
• /
• v.11D no.7 s.96
• /
• pp.1499-1506
• /
• 2004

In this paper an analysis of current video chatting situation and connection status is performed and a technical method to block the obscene contents is proposed. Methods to shut down socially degenerate Internet sites are well known, which are essentially directory or vocabulary based. For the video chatting sites. however, dynamic selectivity is required as the chatting groups in a sites are all different and even in a single group the contents of chatting may turn socially objectionable as time passes. The proposed method is based on the analysis of chatting contents and selectively blocks the sound and/or video data streams. To prove the effectiveness of the method the experiments are performed and the results are shown very satisfactioy.