• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.655-664
• /
• 2020

Free viewpoint image synthesis technology is one of the important technologies in the MPEG-I (Immersive) standard. RVS (Reference View Synthesizer) developed by MPEG-I and in use in MPEG group is a DIBR (Depth Information-Based Rendering) program that generates an image at a virtual (intermediate) viewpoint from multiple viewpoints' inputs. RVS uses the mesh surface method based on computer graphics, and outperforms the pixel-based ones by 2.5dB or more compared to the previous pixel method. Even though its OpenGL version provides 10 times speed up over the non OpenGL based one, it still shows a non-real-time processing speed, i.e., 0.75 fps on the two 2k resolution input images. In this paper, we analyze the internal of RVS implementation and modify its structure, achieving 34 times speed up, therefore, real-time performance (22-26 fps), through the 3 key improvements: 1) the reuse of OpenGL buffers and texture objects 2) the parallelization of file I/O and OpenGL execution 3) the parallelization of GPU shader program and buffer transfer.

• Aerospace Engineering and Technology
• /
• v.9 no.2
• /
• pp.44-50
• /
• 2010

COMS accommodates multiple payloads; Meteorological Image(MI), Ocean Color Imager(GOCI) and Ka-band communication payloads. In order to improve the quality of MI visible channel, the moon image has been taken into account as backup reference in addition to Albedo monitoring. However, obtaining the moon image by adding special mission schedule is not recommended after IOT, because we may miss chances to obtain meteorological images during the time slots for special imaging. As an alternative solution, an approach extracting moon image from MI FD(Full Disk) image has been proposed when the moon is positioned near to the earth. However, prediction of acquisition time of moon image is somewhat difficult as the moon moves while the MI is scanning type sensor. And the moon can not be seen when it is behind the earth or outside of FD field of view. This paper discusses how effectively the moon can be detected by the MI FD imaging. For that purpose, this paper describes an approach taken to predict the time when the moon image is achievable and then introduces the results obtained from computer simulation.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.79-87
• /
• 2021

In recent years research on automatic font generation with machine learning mainly focus on using transformation-based methods, in comparison, generative model-based methods of font generation have received less attention. Transformation-based methods learn a mapping of the transformations from an existing input to a target. This makes them ambiguous because in some cases a single input reference may correspond to multiple possible outputs. In this work, we focus on font generation using the generative model-based methods which learn the buildup of the characters from noise-to-image. We propose a novel way to train a conditional generative deep neural model so that we can achieve font style control on the generated font images. Our research demonstrates how to generate new font images conditioned on both character class labels and character style labels when using the generative model-based methods. We achieve this by introducing a modified generator network which is given inputs noise, character class, and style, which help us to calculate losses separately for the character class labels and character style labels. We show that adding the character style vector on top of the character class vector separately gives the model rich information about the font and enables us to explicitly specify not only the character class but also the character style that we want the model to generate.

• Korean Journal of Optics and Photonics
• /
• v.15 no.2
• /
• pp.183-189
• /
• 2004

As high quality digital cameras become readily available, digital cameras are being used not only for simple picture recording but also as information storing media in various fields. However, due to the fact that the spectral responses of the camera sensors are different from color matching functions of the CIE standard observer, the color can not be measured using these cameras. This study shows a method for converting camera image to sRGB image, in which color information is preserved. The transfer matrix between camera output signals and CIE stimulus values was determined using a multiple regression method with Macbeth ColorChecker as target colors. The CIE stimulus values for camera output signals can be mapped with a transfer matrix, and these values are converted to sRGB signals. As the result of testing a Kodak DC220 digital camera, the average color difference of Macbeth ColorChecker between true and displayed colors was 2.1 $\Delta$ $E_{ab}$ $^{*}$.$^{*}$.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.160-166
• /
• 2020

As mobile-based hardware technology develops, many kinds of applications are also being developed. In addition, there is an increasing demand to automatically check that the interface of these applications works correctly. In this paper, we describe a method for accurately detecting faulty images from applications by comparing major characteristics from input color images. For this purpose, our method first extracts major characteristics of the input image, then calculates the differences in the extracted major features, and decides if the test image is a normal image or a faulty image dissimilar to the reference image. Experiment results show that the suggested approach robustly determines similar and dissimilar images by comparing major characteristics from input color images. The suggested method is expected to be useful in many real application areas related to computer vision, like video indexing, object detection and tracking, image surveillance, and so on.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.5
• /
• pp.551-555
• /
• 2010

In protein 2-DE image analysis, the accuracy of spot-matching operation which identifies the spot of the same protein in each 2-DE gel image is intensively influenced by the errors caused by the various experimental conditions. This paper proposes an efficient method to find more accurate spot-matching patterns based on multiple reference gel images in spot-matching pattern analysis in protein 2-DE image analysis. Additionally, in order to improve the reduce the execution time which is increased exponentially along with the increasing number of gel images, a "partition then extension" framework is used to find spot-matching pattern of long length and of higher accuracy. In the experiments on real 2-DE images of human liver tissue are used to confirm the accuracy and the efficiency of the proposed algorithm.

• Korean Journal of Remote Sensing
• /
• v.30 no.5
• /
• pp.547-557
• /
• 2014

This paper presents a method for designing circular template used in similarity measurement for image registration. Circular template has translation and rotation invariant property, which results in correct matching of control points for image registration under the condition of translation and rotation between reference and sensed images. Circular template consisting of the pixels located on the multiple circumferences of the circles whose radii vary from zero to a certain distance, is converted to two-dimensional Discrete Polar Coordinate Matrix (DPCM), whose elements are the pixels of the circular template. For sensed image, the same type of circular template and DPCM are created by rotating the circular template repeatedly by a certain degree in the range between 0 and 360 degrees and then similarity is calculated using mutual information of the two DPCMs. The best match is determined when the mutual information for each rotation angle at each pixel in search area is maximum. The proposed algorithm was tested using KOMPSAT-2 images acquired at two different times and the results indicate high accurate matching performance under image rotation.

• Journal of Broadcast Engineering
• /
• v.12 no.2
• /
• pp.128-136
• /
• 2007

Since the multi-view video is a set of video sequences captured by multiple array cameras for the same three-dimensional scene, it can provide multiple viewpoint images using geometrical manipulation and intermediate view generation. Although multi-view video allows us to experience more realistic feeling with a wide range of images, the amount of data to be processed increases in proportion to the number of cameras. Therefore, we need to develop efficient coding methods. One of the possible approaches to multi-view video coding is to generate an intermediate image using view interpolation method and to use the interpolated image as an additional reference frame. The previous view interpolation method for multi-view video coding employs fixed size block matching over the pre-determined disparity search range. However, if the disparity search range is not proper, disparity error may occur. In this paper, we propose an efficient view interpolation method using initial disparity estimation, variable block-based estimation, and pixel-level estimation using adjusted search ranges. In addition, we propose a multi-view video coding method based on H.264/AVC to exploit the intermediate image. Intermediate images have been improved about $1{\sim}4dB$ using the proposed method compared to the previous view interpolation method, and the coding efficiency have been improved about 0.5 dB compared to the reference model.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.341-344
• /
• 2007

A major obstacle to classify and validate Land Cover maps is the high cost of generating reference data or multiple thematic maps for subsequent comparative analysis. In case of inaccessible area such as North Korea, the high resolution satellite imagery may be used as in situ data so as to overcome the lack of reliable reference data. The objective of this paper is to investigate the possibility of utilizing QuickBird (0.6m) of North Korea obtained from Google Earth data provided thru internet. Monthly NDVI images of nine months from the summer of 2004 were classified into L=54 cluster using ISODATA algorithm, and these L clusters were assigned to 7 classes; coniferous forest, deciduous forest, mixed forest, paddy field, dry field, water and built-up area. The overall accuracy and Kappa index were 85.98% and 0.82, respectively, which represents about 10% point increase of classification accuracy than our previous study based on GCP point data around North Korea. Thus we can conclude that Google Earth may be used to substitute the traditional in situ data collection on the site where the accessibility is severely limited.

• Journal of Internet of Things and Convergence
• /
• v.6 no.2
• /
• pp.37-43
• /
• 2020

The paper proposed the method of calculating the latest approach angle that can reliably recognize multiple laser images even with the change in separation distance between screen and laser launch device. This method recognizes the angle of the laser pattern angle by using the distance of the laser pattern angle, and the angle extraction of the laser detects the laser image from the acquired image using the labeling algorithm, and performs the huff conversion to extract the angle of the straight line. The distance of the reference angle and angle of the laser image extracted using Euclidean distance among similarity scales is calculated, and the furnace is recognized using the calculated distance result value. Experiments with changing the separation distance to "200 cm to 400 cm" showed 100% recognition of individual strands at all separation distances. The experiment confirmed the reliability of the proposed method.