• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1995.06a
• /
• pp.83-86
• /
• 1995

Many image display devices allow only a limited number of colors to be simultaneously displayed. In displaying of natural color image using color palette, it is necessary to construct an optimal color palette and map each pixel of the original image to a color palette with fast. In this paper, we proposed the clustering algorithm using local region block centered one color cluster in the prequantized 3-D histogram. Cluster pairs which have the least distortion error are merged by considering distortion measure. The clustering process is continued until to obtain the desired number of colors. Same as the clustering process, original color image is mapped to palette color via a local region block centering around prequantized original color value. The proposed algorithm incorporated with a spatial activity weighting value which is smoothing region. The method produces high quality display images and considerably reduces computation time.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.11a
• /
• pp.40-41
• /
• 2012

양안식 3D 영상 시청 안정성에 대해 제기되고 있는 지속적인 이슈는 3D 영상 시장 활성화를 위해 반드시 선결 되어야 하는 과제이다. 최근, 이를 위해 양안식 3D 영상이 유발하는 시각적 불편함을 측정하고 이를 예측하는 연구가 활발히 진행되고 있다. 본 논문에서는 양안식 3D 영상에서 시각적 불편함을 야기하는 문제 영역의 양안 시차 변화에 따른 주관적 시각적 불편함을 측정한다. 이를 통하여 양안식 3D 영상에서 시각적 불편함을 주는 문제 영역의 영상 특성이 시청자가 인지하는 시각 피로에 지배적인 영향을 미칠 수 있음을 실험적으로 증명한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1997.06a
• /
• pp.107-112
• /
• 1997

This paper proposes the regularized constrained iterative image restoration algorithms which apply new space-adaptive methods to degraded image signals, and analyzes the convergence condition of the proposed algorithm. First, we introduce space-adaptive regularization operators which change according to edge characteristics of local images in order to effectively prevent the restored edges and boundaries from reblurring. And, pseudo projection operator is used to reduce the ringing artifact which results from extensive amplification of noise components in the restoration process. The analysed algorithm is stable convergent to the fixed point. According to the experimental results for various signal-to-noise ratios(SNR) and blur models, the proposed algorithms other methods and is robust to noise effects and edge reblurring by regularization especially.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1997.06a
• /
• pp.19-24
• /
• 1997

This paper describes an image segmentation technique for the object-oriented coding at very low bit rates. By noting that, in the object-oriented coding technique, each objects are represented by 3 parameters, namely, shape, motion, and color informations, we propose a segmentation technique, in which the 3 parameters are fully exploited. To achieve this goal, starting with the color space conversion and the noise reduction, the input image is divided into many small regions by the K-menas algorithm on the O-K-S color space. Then, each regions are merged, according to the shape and motion information. In simultations, it is shown that the proposed technique segments the input image into relevant objects, according to the shape and motion as well as the colors. In addition, in order to evaluate the performance of the proposed technique, we introduce the notion of the interesting regions, and provide the results of encoding the image with emphasizing the interesting regions.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06b
• /
• pp.141-146
• /
• 1998

Currently virtual studio has used the cromakey method in which an image is captured, and the blue portion of that image is replaced by a graphic image or a real image. The replaced image must be changed according to the camera motion. This paper proposes a novel method to extract camera parameters using the recognition of pentagonal patterns which are painted on the blue screen. The corresponding parameters are position, direction and focal length of the camera in the virtual studio. At first, pentagonal patterns are found using invariant features of the pentagon. Then, the projective transformation of two projected images and the camera parameters are calculated using the matched points. Simulation results indicate that camera parameters are more easily calculated compared to the conventional methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1997.11a
• /
• pp.145-149
• /
• 1997

In this paper, a hierarchical spatio-temporal image segmentation method based on binary split algorithm is proposed. Intensity and displacement vector at each pixel are used for image segmentation. The displacement vectors between two image frames which skip over one or several frames can be approximated by accumulating of the velocity vectors calculated from optical flow between two successive frames when the time interval between the two image frames is short enough or the motion is slow. The pixels whose displacement vector and intensity are ambiguous are precisely decided by the modified watershed algorithm using the proposed priority measure. In the experiment, the region of moving object is precisely segmented.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.59-61
• /
• 2018

Highly compressed images typically not only have low resolution, but are also affected by compression artifacts. Performing image super-resolution (SR) directly on highly compressed image would simultaneously magnify the blocking artifacts. In this paper, a SR method based on deep learning is proposed. The method is an end-to-end trainable deep convolutional neural network which performs SR on compressed images so as to reduce compression artifacts and improve image resolution. The proposed network is divided into compression artifacts removal (CAR) part and SR reconstruction part, and the network is trained by three-step training method to optimize training procedure. Experiments on JPEG compressed images with quality factors of 10, 20, and 30 demonstrate the effectiveness of the proposed method on commonly used test images and image sets.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.59-62
• /
• 2018

In this paper, we propose a contrast enhancement algorithm using guided image filter (GIF). The GIF is used to divide an HDR image into a base layer and a detail layer. The energy scale of base layer determinate the darkness and brightness of the image. However, the detail information in the base layer is difficult to be displayed because of the high brightness and clusters of low brightness. We propose a contrast enhancement method by adjusting the gray level of base layer by subtracting the mean value of itself. It is combined with the detail layer to preserve the detail information. Experiment results show that the proposed algorithm has better performance in detail preservation and contrast enhancement.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.212-215
• /
• 2021

본 논문에서는 6 Fisheye lens 원본 영상에 대하여 Insta360 stitcher, AutoStitch[4], As-Projective-AsPossible(APAP)[5] 스티칭 방법으로 360 도 파노라마 영상을 생성하고 기하학적 왜곡과 컬러 왜곡을 비교 평가한다. 360 도 파노라마 Image Quality Assessment(IQA) 메트릭으로 Natural Image Quality Evaluator(NIQE)[6], Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE)[7], Perception based Image Quality Evaluator(PIQE)[8], Feature Similarity(FSIM)[9] 그리고 high frequency feature 에 대한 Structural Similarity(SSIM)[10]을 측정하여 정량적 평가를 하며 정성적인 비교를 통하여 파노라마 영상의 품질과 평가 메트릭에 대한 벤치마크를 제공한다.

• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.463-473
• /
• 2013

In this paper, we propose a super-resolution (SR) image reconstruction algorithm using multi-view images. We acquire 25 images from multi-view cameras, which consist of a $5{\times}5$ array of cameras, and then reconstruct an SR image of the center image using a low resolution (LR) input image and the other 24 LR reference images. First, we estimate disparity maps from the input image to the 24 reference images, respectively. Then, we interpolate a SR image by employing the LR image and matching points in the reference images. Finally, we refine the SR image using an iterative regularization scheme. Experimental results demonstrate that the proposed algorithm provides higher quality SR images than conventional algorithms.