• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.8
• /
• pp.1704-1713
• /
• 2011

This paper proposes a method to generate multiview images which use a synthesized image consisting of layered objects. The camera system which consists of a depth camera and a RGB camera is used in capturing objects and extracts 3-dimensional information. Considering the position and distance of the synthesizing image, the objects are synthesized into a layered image. The synthesized image is spaned to multiview images by using multiview generation tools. In this paper, we synthesized two images which consist of objects and human and the multiview images which have 37 view points were generated by using the synthesized images.

• The Journal of the Korea Contents Association
• /
• v.13 no.12
• /
• pp.38-47
• /
• 2013

An image coding based on MRC model, a kind of multi-layer image model, first segments a screen image into foreground, mask, and background layers, and then compresses each layer using a codec that is suitable to the layer. The mask layer defines the position of foreground regions such as textual and graphical contents. The colour signal of the foreground (background) region is saved in the foreground (background) layer. The mask layer which contains the segmentation result of foreground and background regions is of importance since its accuracy directly affects the overall coding performance of the codec. This paper proposes a new layer segmentation algorithm for the MRC based image coding. The proposed method extracts text pixels from the background using morphological top hat filtering. The application of white or black top hat transformation to local blocks is controlled by the information of relative brightness of text compared to the background. In the proposed method, the boundary information of text that is extracted from the edge map of the block is used for the robust decision on the relative brightness of text. Simulation results show that the proposed method is superior to the conventional methods.

• Proceedings of the KSR Conference
• /
• 2010.06a
• /
• pp.533-539
• /
• 2010

The The contactless railway abrasion measurement system have to satisfy two conditions to increase the measurement accuracy as follows. The laser region projected on the rail have to be extracted without the geometrical distortion. The mapping of the acquired laser region data on the rail profile have to be matched with the cross section of rail, exactly. But, the conventional railway abrasion measurement system is required the post image processing with a camera model and a perspective transform for the exact mapping between the cross section of rail and the coordinate data extracted from a line laser region or the raw image obtained from a camera because the image captured from the camera has an oblique viewpoint. So, the measured rail profile data had limits to the measurement accuracy because of a discontinuity point. In this Paper, we propose the 3D rail profile reconstruction method to increase the accuracy of the railway abrasion measurement system applying the modified camera model and perspective transform to the image obtained from the bidirectional rail.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.304-305
• /
• 2010

지금까지는 영화나 3DTV같은 대형 모니터에 사용하기 위하여 입체영상이 제작되어 왔다, 하지만 최근 3D 디스플레이가 스마트폰이나 태블릿PC와 같이 작은 디스플레이 사이즈와 해상도를 가지는 휴대용 기기들에도 적용되어 다양한 형태로 보급되어 보다 많은 관심을 받고 있다. 하지만 모바일 기기와 같은 경우 영화와 고해상도 영상에서와 같은 3D 입체감을 느끼기 어려운 경우가 종종 발생한다. 본 논문에서는 작은 디스플레이 사이즈와 해상도를 가지는 모바일 기기와 같은 장치에서 3D 효과 향상을 위한 깊이맵 재구성의 방법으로 레이어의 개수를 제안하여 3D 효과를 향상 시키는 방법을 제안한다. 이를 위해 고해상도 영상에서 깊이값에 따른 레이어를 6개에서 2개까지 수를 줄여 깊이맵을 생성하였고 영상의 크기를 줄여 영상의 크기와 깊이 레이어에 따라 달라지는 입체 효과를 확인하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.12 no.4
• /
• pp.405-413
• /
• 1987

In this paper, we have discussed three-type median filters(SQUARE, CROSS, X-SHAPE) that preserving edge in an original image while reducing random noise was introduced for image enhancement and edge detection on radar image. Since radar image have a number of parts of curve, we compared results produced by edge detection operater proposed for improving the parts of curve with results of using the existing edge detection methods, such as Roberts, Sobel, Prewitt, Laplacian and Kirsch.

• Journal of Korea Game Society
• /
• v.20 no.3
• /
• pp.15-24
• /
• 2020

We present a reinforcement learning-based framework for generating 2D Lego puzzle from input pixel art images. We devise heuristics for a proper Lego puzzle as stability and efficiency. We also design a DQN structure and train it to maximize the heuristics of 2D Lego puzzle. In legorization stage, we complete the layout of Lego puzzle by adding a Lego brick to the input image using the trained DQN. During this process, we devise a region of interest to reduce the computational loads of the legorization. Using this approach, our framework can present a very high resolutional Lego puzzle.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.6
• /
• pp.411-414
• /
• 2018

In this paper, we propose a radar image extraction scheme for frequency modulated continuous wave radar-based human motion indication. We extracted three-dimensional(3D) range-velocity-angle spectra and generated three micro-profile images by compressing the 3D images in all three directions in every frame. Furthermore, we used body echo suppression to make use of the weak reelection such as in hands and arms. By applying the complete images to classifiers, various human motions can be indicated.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.19 no.6
• /
• pp.663-668
• /
• 2008

Recently, World Radio Conference(WRC)-2007 approved the ultrawide bandwidth of 500 MHz for the use of spaceborne synthetic aperture radar in X-band for the EESS(Earth Exploration Satellite Service) in order to improve the SAR imaging resolution. It is concerned about the interference impact from the spaceborne SAR that may cause to most of ground radars due to the extended ultra wideband. In this paper, in order to predict the interference impact of the ground-based radar from the spaceborne radar, radar interference model is presented using radar characteristic parameters by taking into account the operating environments of the spaceborne and ground based radar in the time, space, and spectrum domains. Using the spaceborne SAR model of TerraSAR-X and ground radar model of meteorological radar recommended by ITU-R, the interference impact was assessed through the computer simulation to see the possible interference impact of the ground based radar operating in the Korean peninsula.

• Journal of Korea Multimedia Society
• /
• v.2 no.4
• /
• pp.409-418
• /
• 1999

The fractal image compression is based on the self-similarity that some area in an image exhibits a very similar shape with other areas. This compression technique has very long encoding time although it has high compression ratio and fast decompression. To cut-off the encoding time, most researches have restricted the search of domain blocks for a range block. These researches have been mainly focused on the coherence between a domain block and a range block, while they have not utilized the coherence among range blocks well. Therefore, we give an analysis on the coherence among range blocks in order to develope an efficient fractal Image compression algorithm. We analysis the range blocks according to not only measures for defining the range block coherence but also threshold of each measure. If these results are joined in a prior work of other fractal compression algorithms, it will give a great effectiveness in encoding time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.4
• /
• pp.935-944
• /
• 2013

In this paper, a new illumination compensation algorithm by segmentation with depth information is proposed to improve the coding efficiency of multi-view images. In the proposed algorithm, a reference image is first segmented into several layers where each layer is composed of objects with a similar depth value. Then we separate objects from each other even in the same layer by labeling each separate region in the layered image. Then, the labeled reference depth image is converted to the position of the distortion image view by using 3D warping algorithm. Finally, we apply an illumination compensation algorithm to each of matched regions in the converted reference view and distorted view. The occlusion regions that occur by 3D warping are also compensated by a global compensation method. Through experimental results, we are able to confirm that the proposed algorithm has better performance to improve coding efficiency.