• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.2
• /
• pp.81-87
• /
• 1999

In this research, we propose a field mismatch compensation method for interlaced scan image and a image restoration technique for removing motion blur. In order to compensate field mismatch, the edge classification-based linear interpolation technique and the method using the object-based motion compensation are described. We also propose an edge estimation method and an motion-based image segmentation algorithm. For removing motion blur, we adopt an adaptive iterative image restoration method using the motion-based segmentation result to improve the quality of restored image.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.2C
• /
• pp.163-168
• /
• 2009

In this paper, a two-dimensional (2D) modulation code is introduced. The proposed code does not have any isolated pixel that is the most unwanted problem for holographic data storage. The proposed 2D modulation code is simpler than conventional 6/8 code and removes all the isolated 2D ISI patterns. As a result, when the grade of blur is 1.4, the proposed modulation code has better performance overall than conventional 6/8 modulation code. The proposed code has the optimal performance when 4bit quantization is applied.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.13 no.5
• /
• pp.376-382
• /
• 2008

In this paper, scanning backlight unit which can reduce motion blur was designed and fabricated using flat fluorescent lamp. The FFL(flat fluorescent lamp) is in the limelight as a new illuminant of next generation BLU(backlight unit), because of simple assembly and reduction of driving components. In order to control lamp brightness, lamp on-time was controlled. In this experiment, it was confirmed that lamp brightness can be dimmed linealy and in a wide range.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.17 no.3
• /
• pp.26-35
• /
• 1980

The degradation of image due to the finite size of sensing devices has been one of the problems in all digital imaging systems. The basic study on the improvement of the spatial resolution was carried out in both spatial and frequency domains by the resolution recovery techniques which have been used in optics. Here, the techniques were applied to CT (Computerized Tomography) system, and image with finer resolution was obtained by these techniques. The basic theory is described and the results of the simulation are shown.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.70-72
• /
• 2016

본 논문에서는 기존의 인지적 영상 압축 기법에 사용되었던 Just Noticeable Distortion(JND) 모델이 압축과정에서 생기는 왜곡인 양자화 왜곡에 적합하지 않는 다는 것을 보이고, 그 한계점을 해결하기 위하여 Just Noticeable Blur(JNB)의 개념을 적용하여 영상 압축에 적합한 모델을 제시하였다. 주파수 공간에서 영상의 복잡도 특징을 나타내는 Spectral Contras Index(SCI) 값을 사용해서 영상의 DCT 블록별 JNB 를 추정하고 이를 기반으로 영상의 DCT 계수 값을 감소시켜 최신의 DCT 기반 JND 를 적용한 인지적 압축 영상에 비해 더 낮은 PSNR 을 가지면서 왜곡도 인지되지 않는 영상을 얻을 수 있었다. 새롭게 제시한 모델을 적용하면 인지적 영상압축에서 기존의 방법보다 더 낮은 비트율로 유사한 인지적 화질 성능을 발휘할 것으로 예상된다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.399-402
• /
• 2011

카메라를 이용하여 정지 영상을 획득할 때 발생할 수 있는 모션 블러 (motion blur) 현상을 보정하기 위해서는 정확한 블러파라미터 (블러의 방향, 블러의 크기)의 추정이 가장 중요하다. 본 논문에서는 블러된 영상의 로그 스펙트럼 (log spectrum)의 비등방성 (anisotropic)과 엔트로피 (Entropy) 특성에 기반한 블러 파라미터 추정 방법을 제안한다. 제안하는 기법은 1D 로그스펙트럼을 이용한 비등방성의 효과적인 분석과 블러 크기의 정확도를 증가시키기 위한 보상 과정을 포함한다. 실험 결과를 통해 제안하는 기법이 단일 영상에서의 등속 직선 운동의 모션 블러 현상을 더 효과적으로 보정하는 것을 보였다.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 2007.05a
• /
• pp.126-129
• /
• 2007

Hold-type display의 동화상 인식 원리를 이용하여 인간의 시각특성을 고려한 LCD의 동화상 퍼짐(blur) 노이즈 컴퓨터 시뮬레이터를 개발하였다. 그리고 이 시뮬레이터를 이용하여 통화상 퍼짐 노이즈 특성을 실험한 결과 화상의 이동속도가 빠를수록 퍼짐 노이즈가 증가하며 백라이트의 점등비가 낮을수록 노이즈가 저감된다는 것을 알았다. 이는 기존의 연구결과들과 잘 일치하는 것으로 이것으로부터 본 연구에서 개발한 시뮬레이터의 알고리듬이 타당하다고 할 수 있다.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.225-232
• /
• 2017

Understanding actions in videos is an important task. It helps in finding the anomalies present in videos such as fights. Detection of fights becomes more crucial when it comes to sports. This paper focuses on finding fight scenes in Hockey sport videos using blur & radon transform and convolutional neural networks (CNNs). First, the local motion within the video frames has been extracted using blur information. Next, fast fourier and radon transform have been applied on the local motion. The video frames with fight scene have been identified using transfer learning with the help of pre-trained deep learning model VGG-Net. Finally, a comparison of the methodology has been performed using feed forward neural networks. Accuracies of 56.00% and 75.00% have been achieved using feed forward neural network and VGG16-Net, respectively.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.681-684
• /
• 2008

Motion interpolation is adopted and has been spread widely into market since it is effective in reducing motion blur, which is considered as weak characteristic due to slow response time of liquid crystal and hold-type display. 120Hz driving using interpolated frames achieves better moving picture quality with less motion blur and less motion judder. However, errors in the interpolated frames can cause visual artifacts such as static text breakup, halos, and occlusions. This paper focuses on categorizing characteristics of visual artifacts and on reducing side-effects by using information from original frames in special cases.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.827-828
• /
• 2008

Blur identification is the first and the most important step of restoring images. Edge region of the image usually conveys important information of blur parameters. In this paper we propose a region-based edge extraction method for estimating point-spread-function (PSF). As a result, the proposed method can detect the starting and the ending points of a step response, and provides the PSF parameters to the restoration process.