• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.492-497
• /
• 2009

Computer-Generated Hologram (CGH) is made for three dimensional image of a virtual object. Error diffusion method is used for the phase quantization of CGH, and it is known to be effective to the image quality improvement of the reconstructed image. However, the image quality of the reconstructed image from the CGH using error diffusion method depends on the selection of error diffusion coefficient. In this paper, we derived the relational expression to obtain the error diffusion coefficient from the position of the input object and size of the input object for CGH. As a result, the method of this thesis was able to obtain an excellent reconstructed image compared with the case to derive the error diffusion coefficient from only the position of the input image.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.112-115
• /
• 2022

The process of deep learning usually needs to deal with massive data which has greatly limited the development of deep learning technologies today. Convolutional Neural Network (CNN) structure is often used to solve image classification problems. However, a large number of images may be required in order to train an image in CNN, which is a heavy burden for existing computer systems to handle. If the image data can be compressed under the premise that the computer hardware system remains unchanged, it is possible to train more datasets in deep learning. However, image compression usually adopts the form of lossy compression, which will lose part of the image information. If the lost information is key information, it may affect learning performance. In this paper, we will analyze the effect of image compression on deep learning performance on CNN-based cat and dog classification. Through the experiment results, we conclude that the compression of images does not have a significant impact on the accuracy of deep learning.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.825-828
• /
• 2022

In this paper, we propose an efficient image size selection method for video-based point cloud compression. The current MPEG video-based point cloud compression reference encoding process configures a threshold on the size of images while converting point cloud data into images. Because the converted image is compressed and restored by the legacy video codec, the size of the image is one of the main components in influencing the compression efficiency. If the image size can be made smaller than the image size determined by the threshold, compression efficiency can be improved. Here, we studied how to improve the compression efficiency by selecting the best-fit image size generated during video-based point cloud compression. Experimental results show that the proposed method can reduce the encoding time by 6 percent without loss of coding performance compared to the test model 15.0 version of video-based point cloud encoder.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.58-60
• /
• 2019

Double JPEG compression detection is one of the most important ways of exposing the integrity of the JPEG image in image forensics. Several methods have been proposed for discriminating against the double JPEG image. In this paper, we propose a new method for restoring the JPEG compressed image and making the detector confused by introducing a Generative Adversarial Network (GAN). First, a generator network is designed for restoring the JPEG compressed image and analyzed the quality. Then, the restored image is tested with the double compression detector for evaluating the robustness of the proposed GAN model. The detection accuracy reduces from 98% to 58%.

• Journal of Broadcast Engineering
• /
• v.1 no.2
• /
• pp.96-107
• /
• 1996

In case of remote image retrieval via shared network or low speed link in order to make a decision for target image problems such as transmission delay are encountered. In this paper browsing and optimization techniques are proposed for fast retrieval of Image by the multiresolution representation and progressive transmission. The proposed network model was analyzed and evaluated for system's performance improvement. Interactive user-system using several multiresolution representation has shown better performance in transmission delay minimization over the single resolution image retrieval system.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.99-102
• /
• 2009

In this paper a hierarchical stereo matching algorithm based on feature extraction is proposed. The boundary (edge) as feature point in an image is first obtained by segmenting an image into red, green, blue and white regions. With the obtained boundary information, disparities are extracted by matching window on the image boundary, and the initial disparity map is generated when assigned the same disparity to neighbor pixels. The final disparity map is created with the initial disparity. The regions with the same initial disparity are classified into the regions with the same color and we search the disparity again in each region with the same color by changing block size and search range. The experiment results are evaluated on the Middlebury data set and it show that the proposed algorithm performed better than a phase based algorithm in the sense that only about 14% of the disparities for the entire image are inaccurate in the final disparity map. Furthermore, it was verified that the boundary of each region with the same disparity was clearly distinguished.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.360-363
• /
• 2009

Variation of viewing position of the observer is one of factors of image distortion in the stereoscopic display. A rotation movement of the observer makes the stereoscopic image distortion and that is caused by different horizontal position of each eye of the observer. It is different from horizontal and depth directional movement of the observer. In this paper, we showed the numerical simulation result about the distortion analysis and the correction of the stereoscopic image in rotation movement of the observer.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.816-820
• /
• 2009

In landscape simulatin, it is necessary to express very realistic image generated by computer graphics. One solution is to use texture mapping; however, it needs a lot of work and time to obtain images for texture mapping since there are huge variety of images for buildings, roads, stations and so on, and the landscape image is diverse due to the weather and time. Especially, weathered images such as stain on walls, crack on roads and so forth, are needed to make the landscape image very realistic. These weathered images do not have to be strict so that it saves a lot of work and time for obtaining the images for texture mapping if we can generate a variety of weathered images automatically. Therefore, this paper describes how to generate a variety of weathered images automatically by changing the weathered shape of the original image.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.714-718
• /
• 2009

3D reconstruction of a human face from an image sequence remains an important problem in computer vision. We propose a method, based on a factorization algorithm, that reconstructs a 3D face model from short image sequences exhibiting rotational motion. Factorization algorithms can recover structure and motion simultaneously from one image sequence, but they usually require that all feature points be well tracked. Under rotational motion, however, feature tracking often fails due to occlusion and frame out of features. Additionally, the paucity of images may make feature tracking more difficult or decrease reconstruction accuracy. The proposed 3D reconstruction approach can handle short image sequences exhibiting rotational motion wherein feature points are likely to be missing. We implement the proposal as a reconstruction method; it employs image sequence division and a feature tracking method that uses Active Appearance Models to avoid the failure of feature tracking. Experiments conducted on an image sequence of a human face demonstrate the effectiveness of the proposed method.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.180-181
• /
• 2015

In this paper, we present a consistent and efficient edit propagation method that is applied for light field data. Unlike conventional sparse edit propagation, the coherency between light field sub-aperture images is fully considered by utilizing light field consistency in the optimization framework. Instead of directly solving the optimization function on all light field sub-aperture images, the proposed optimization framework performs sparse edit propagation in the extended focus image domain. The extended focus image is the representative image that contains implicit depth information and the well-focused region of all sub-aperture images. The edit results in the extended focus image are then propagated back to each light field sub-aperture image. Experimental results on test images captured by a Lytro off-the-shelf light field camera confirm that the proposed method provides robust and consistent results of edited light field sub-aperture images.