• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.3
• /
• pp.135-138
• /
• 2014

This paper devices a facial expression recognition method robust to overfitting using 2D-DCT and EHMM algorithm. In particular, this paper achieves enhanced recognition performance by setting up a large window size for 2D-DCT feature extraction and extracting the observation vectors of EHMM. The experimental results on the CK facial expression database and the JAFFE facial expression database showed that the facial expression recognition accuracy was improved according as window size is large. Also, the proposed method revealed the recognition accuracy of 87.79% and showed enhanced recognition performance ranging from 46.01% to 50.05% in comparison to previous approaches based on histogram feature, when CK database is employed for training and JAFFE database is used to test the recognition accuracy.

• ETRI Journal
• /
• v.33 no.2
• /
• pp.219-229
• /
• 2011

In this paper, we propose a transformation scheme used to analyze online gaming traffic properties and develop a traffic model. We analyze the packet size and the inter departure time distributions of a popular first-person shooter game (Left 4 Dead) and a massively multiplayer online role-playing game (World of Warcraft) in order to compare them to the existing scheme. Recent online gaming traffic is erratically distributed, so it is very difficult to analyze. Therefore, our research focuses on a transformation scheme to obtain new smooth patterns from a messy dataset. It extracts relatively heavy-weighted density data and then transforms them into a corresponding dataset domain to obtain a simplified graph. We compare the analytical model histogram, the chi-square statistic, and the quantile-quantile plot of the proposed scheme to an existing scheme. The results show that the proposed scheme demonstrates a good fit in all parts. The chi-square statistic of our scheme for the Left 4 Dead packet size distribution is less than one ninth of the existing one when dealing with erratic traffic.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.12B
• /
• pp.2401-2406
• /
• 1999

Various techniques extracting feature vectors have been studied for the cut detection in compressed video data. In case of using the histogram of occurrence of pixel's values as a feature vector, the precise detection of cuts would not be expected because of not considering the spatial correlation of pixels. And more sophisticated algorithms such as CCV(Color Coherent Vector) and Correlrogram tend to be used. Though these methods can be able to detect cuts rather precisely, they require much more processing time because of a enormous amount of computations. In this paper we propose a method of the cut detection using spatial correlation of DC values of luminance components in MPEG video sequence. This requires less processing time and also It can increase the rates of detecting the correct cuts by using advanced comparative method.

• Journal of Electrical Engineering and Technology
• /
• v.11 no.4
• /
• pp.951-960
• /
• 2016

Tower insulators in electric power transmission network play a crucial role in preserving the reliability of the system. Electrical utilities frequently face the problem of flashover of insulators due to pollution deposition on their surface. Several research works based on leakage current (LC) measurement has been already carried out in developing diagnostic techniques for these insulators. Since the LC signal is highly intermittent in nature, estimation of pollution severity based on LC signal measurement over a short period of time will not produce accurate results. Reports on the measurement and analysis of LC signals over a long period of time is scanty. This paper attempts to use Random Forest (RF) classifier, which produces accurate results on large data bases, to analyze the pollution severity of high voltage tower insulators. Leakage current characteristics over a long period of time were measured in the laboratory on porcelain insulator. Pollution experiments were conducted at 11 kV AC voltage. Time domain analysis and wavelet transform technique were used to extract both basic features and histogram features of the LC signal. RF model was trained and tested with a variety of LC signals measured over a lengthy period of time and it is noticed that the proposed RF model based pollution severity classifier is efficient and will be helpful to electrical utilities for real time implementation.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.11a
• /
• pp.380-383
• /
• 2003

The purpose of this study is to develope prototype EPID system and improve image quality of radiation therapy field imaging system using CCD camera. For this research we used Linac(Clinac 4/100), Copper metal plate, $Gd_2O_2S_2$ phosphor and CCD camera(Photronic). In this study we find best thickness of buil-up metal plate and acquired projection image of humanoid head phantom. Also we enhanced raw image data using superposition and histogram stretching method. Through the thickness optimized of metal plate and image processing, we confirmed of an improved image quality of an EPID system using CCD camera. As result, highest quality image was acquired at 1mm thickness of Copper metal plate and improved image quality by image processing methods.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.11a
• /
• pp.212-215
• /
• 2003

This paper presents constitution of a moderate price thermography system for on-line measurement. All objects with some temperature above absolute zero radiate in the infrared. The intensities and spectra of the infrared radiated from some object depend on the conditions and temperature distributions on the surface of the objects, and the temperature distributions differ from each others with different undersurface structures. Consequently, infrared radiation is useful for diagnosis of the conditions on the surface and undersurface of electric power apparatus. But the present, because engineers directly measure the temperatures of apparatuses in off-line, the measured data do not always have the information of the past. The proposed system is able to analyze not only the information of the past but the trend of deterioration, and the system is able to compensate for the distortion of surface temperature as to weather conditions. The thermal image histogram is equalized to upgrade observability, and a pan-tilter is adopted to control of direction for any target point.

• Journal of Chest Surgery
• /
• v.53 no.4
• /
• pp.184-190
• /
• 2020

Radiation therapy (RT) has improved patient outcomes, but treatment-related complication rates remain high. In the conventional 2-dimensional and 3-dimensional conformal RT (3D-CRT) era, there was little room for toxicity reduction because of the need to balance the estimated toxicity to organs at risk (OARs), derived from dose-volume histogram data for organs including the lung, heart, spinal cord, and liver, with the planning target volume (PTV) dose. Intensity-modulated RT (IMRT) is an advanced form of conformal RT that utilizes computer-controlled linear accelerators to deliver precise radiation doses to the PTV. The dosimetric advantages of IMRT enable better sparing of normal tissues and OARs than is possible with 3D-CRT. A major breakthrough in the treatment of esophageal cancer (EC), whether early or locally advanced, is the use of proton beam therapy (PBT). Protons deposit their highest dose of radiation at the tumor, while leaving none behind; the resulting effective dose reduction to healthy tissues and OARs considerably reduces acute and delayed RT-related toxicity. In recent studies, PBT has been found to alleviate severe lymphopenia resulting from combined chemo-radiation, opening up the possibility of reducing immune suppression, which might be associated with a poor prognosis in cases of locally advanced EC.

• Journal of the Korea Society of Computer and Information
• /
• v.7 no.3
• /
• pp.1-8
• /
• 2002

Due to the fast development in computer application technologies, a video is now being more widely used than ever in many areas. The current information analyzing systems are basically built to process text-based data. Thus, it has little bits Problems when it needs to correctly represent the ambiguity of a video, when it has to process a large amount of comments. or when it lacks the objectivity that the jobs require. We would like to purpose the method that is capable of analyze a large amount of video efficiently. To extract the color, we translate the color from RGB to HSI and use the information that matches with the representative colors. To extract the shape information, we use improved moment invariants(IMI) so that we can solve many problems of histogram intersection.

• Journal of the Korean Society of Visualization
• /
• v.20 no.2
• /
• pp.45-54
• /
• 2022

We analyzed the low wall-shear stress area in the intracranial aneurysm that occurred at an anterior communicating artery with a special emphasis on vortical structures close to the wall. We reconstructed the aneurysm model from patient CTA data. We assumed blood as an incompressible Newtonian fluid and treated the blood vessel as a solid wall. The pulsatile boundary condition was applied at the inlet of the anterior cerebral artery. From the instantaneous flow field, we computed the histogram of the wall-shear stress over the aneurysm wall and found the low wall-shear stress event (< 0.4 Pa). This extreme event was due to the low wall-shear stress area that occurred at the daughter sac. We found that the merging of two vortices induced the low wall-shear stress area; one arises from the morphological characteristics of the daughter sac, and the other is formed by a jet flow into the aneurysm sac. The latter approaches the daughter sac, which ultimately leads to the strong ejection event near the daughter sac.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.292-298
• /
• 2022

There is a limit to improving the image quality through global processing of images taken under backlighting because too bright and dark parts are mixed in one scene. This paper introduces a method to improve the quality of a photo by making two virtual images that improve the dark and bright areas of a backlit photo, and fusing them with the original image into a Laplacian pyramid. The proposed method reduces the computational burden by using histogram stretching and gamma transformation that can be simplified with LUT when creating the two virtual images. In addition, in order to obtain a color-enhanced image, contrast conversion was performed only on the luminance using the HSV coordinate system. The proposed technique showed its effectiveness by calculating several NIQA indicators using standard image data sets.