• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.3
• /
• pp.44-51
• /
• 2009

This paper addresses a face verification method using the nonlinear composite filter. This face verification process can be simple and speedy because it does not require any reprocessing such as face detection, alignment or cropping. The optimum nonlinear composite filter is derived by minimizing the output energy due to additive noise and an input scene while maintaining the outputs of training images constant. The filter is equipped with the discrimination capability and the robustness to additive noise by minimizing the outputs of the input scene and the noise, respectively. We build the nonlinear composite filter with two training images and compare the filter with the conventional synthetic discriminant function (SDF) filter. The receiver operating characteristics (ROC) curves are presented as a metric for the performance evaluation. According to the experimental results the optimum nonlinear composite filter is shown to be a robust scheme for face verification in low resolution and noise environments.

• Journal of the Korean Society of Costume
• /
• v.57 no.5 s.114
• /
• pp.99-111
• /
• 2007

The purpose of this study were to identity face satisfaction and make-up behavior according to their lifestyles of middle aged women. The subjects were 350 women of 40 - 59 years old. The results of this study were as follow: 1. Lifestyle factors of middle aged women were self confidence, leisure activities, interests in appearance, economical efficiency, conservatism, and value-orientation. They were classified into four types of dignified & appearance interested group, passive stagnated group, unconfident & economic group, and value-oriented & leisure group according to their lifestyles. 2. They were satisfied with their eyes, lips, and eyebrows, but not satisfied with face tone, face shape, and skin texture. Total face satisfaction level was over average. The face satisfaction level of dignified & appearance interested group and value-oriented & leisure group were higher than other groups, but that of passive stagnated group was the lowest. 3. They pursued internal beauty and natural makeup, and did not follow trendy colors. Dignified & appearance interested group pursued characteristic, elegant, changeable, and various make-up colors, but value-oriented & leisure group pursued more various, changeable, and trendy colors. They did not prefer trendy products because of high price. Dignified & appearance interested group and unconfident & economic group were highly price-oriented, but value-oriented & leisure group purchased trendy products. 4. Dignified & appearance interested group included working women with not much income, but passive stagnated group included low-educated and non-working women. Unconfident & economic group included low-educated non-working women with not much income, but value-oriented & leisure group included highly-educated working women with high income.

• Journal of the Korean Home Economics Association
• /
• v.33 no.2
• /
• pp.168-180
• /
• 1995

The purpose of present study was to examine how each clothes colors on the basis of 5YR 7/3 face color affect clothes colors images as follows : (1) what general consciousness of clothes colors in, (2) how the impression of the harmony of 5YR 7/3 face color and clothes colors is, (3) when we divide clothes colors according to the property of colors- chromatic color and achromatic color, cool color.neutral color.warm color, in tone, in color colume- if there is the difference of visual evaluation, (4) image analysis of 45 clothes colors with the view of each kind of adjectives. The result of this study is as the following: 1. As a result of the analysis of general consciousness on clothes colors, when subjects chose clothes, they most considered colors and they also considered their face colors. They would choose the color of clothes, which were becoming to their having clothes colors or their face colors when they bought clothes. 2. The impressions of coordination of 5YR 7/3 face color and clothes colors consisted of three dimensions - evaluation, activity and harmony. 3. It was known that as a result of the analysis of visual evalutional differences according to dividing the clothes colors by property of colors, there were such notable differences that they might effect the coordination images of face color and clothes colors differently. 4. After arranging 45 clothes colors on the graphs in 17 adjectives, gethering them thogether in each dimension and as the result of the analysis in the evaluation dimension, estimation of yellow, light green column were low and that of achromatic colors were high. That is, it was known that the evalution dimension was concerned with hue of the color properties. In activity dimension, there were different image according to each adjectives. That is, it was known that the evalution dimension was concerned with hue of the color properties. In activity dimension, there were different image according to each adjectives. That is, it was known that the activity demension was concerned with value and chroma of the color properties. In harmony dimension, achromatic columm was high and yellow, green yellow, vivid green columm were low in harmony. That is, it was known that the harmony demension was concerned with hue of the color properties.

• The Journal of Korea Robotics Society
• /
• v.3 no.4
• /
• pp.270-277
• /
• 2008

The development of a face robot basically targets very natural human-robot interaction (HRI), especially emotional interaction. So does a face robot introduced in this paper, named Buddy. Since Buddy was developed for a mobile service robot, it doesn't have a living-being like face such as human's or animal's, but a typically robot-like face with hard skin, which maybe suitable for mass production. Besides, its structure and mechanism should be simple and its production cost also should be low enough. This paper introduces the mechanisms and functions of mobile face robot named Buddy which can take on natural and precise facial expressions and make dynamic gestures driven by one laptop PC. Buddy also can perform lip-sync, eye-contact, face-tracking for lifelike interaction. By adopting a customized emotional reaction decision model, Buddy can create own personality, emotion and motive using various sensor data input. Based on this model, Buddy can interact probably with users and perform real-time learning using personality factors. The interaction performance of Buddy is successfully demonstrated by experiments and simulations.

• Journal of Korea Multimedia Society
• /
• v.19 no.5
• /
• pp.848-855
• /
• 2016

Face recognition has received a great deal of attention for the wide range of applications in real-world scenario. In this scenario, mismatches (so called heterogeneity) in terms of resolution and illumination between gallery and test face images are inevitable due to the different capturing conditions. In order to deal with the mismatch problem, we propose a local feature learning method using deep canonical correlation analysis (DCCA) for heterogeneous face recognition. By the DCCA, we can effectively reduce the mismatch between the gallery and the test face images. Furthermore, the proposed local feature learned by the DCCA is able to enhance the discriminative power by using facial local structure information. Through the experiments on two different scenarios (i.e., matching near-infrared to visible face images and matching low-resolution to high-resolution face images), we could validate the effectiveness of the proposed method in terms of recognition accuracy using publicly available databases.

• Journal of Korean Society for Atmospheric Environment
• /
• v.10 no.3
• /
• pp.183-190
• /
• 1994

Performance of domestic glass fabrics was tested in a Pulse- jet cleaned fabric filter under simulated coal combustion. Pulse Pressure were 2.5, 4.0kgf/$\textrm{cm}^2$ and pulse air nozzle diameter were 4.0, 6.0mm Pressure drop and penetration turned out to be low at small pulse air nozzle diameter and low pulse air pressure. Fractional penetration through the dust cake and fabric at face velocity of 1.7m/min was higher than that at face velocity of 1.0m/min. As a consequense, the performance of domestic glass fabrics was better with face velocity of less than 1.0m/min, pulse air pressure of 2.5 kgf/$\textrm{cm}^2$ and pusle air nozzle diameter of 4.0mm.

• Journal of Korea Multimedia Society
• /
• v.15 no.5
• /
• pp.606-614
• /
• 2012

Face detection using AdaBoost algorithm is one of the fastest and the most robust face detection algorithm so many improvements or extensions of this method have been proposed. However, almost all previous approaches deal with only frontal face and suffer from limited discriminant capability for inclined face because these methods apply the same features for both frontal and inclined face. Also conventional approaches for detecting inclined face which apply frontal face detecting method to inclined input image or make different detectors for each angle require heavy computational complexity and show low detection rate. In order to overcome this problem, a method for detecting inclined face using JointBoost is proposed in this paper. The computational and sample complexity is reduced by finding common features that can be shared across the classes. Simulation results show that the detection rate of the proposed method is at least 2% higher than that of the conventional AdaBoost method under the learning condition with the same iteration number. Also the proposed method not only detects the existence of a face but also gives information about the inclined direction of the detected face.

• Proceedings of the Korean Society of Tribologists and Lubrication Engineers Conference
• /
• 1999.06a
• /
• pp.197-202
• /
• 1999

A mechanical face seal is a tribe-element intended to control the leakage of working fluid at the interface of a rotating shaft and its housing. The leakage of working fluid decreases as the seal surfaces get closer each other. But a very small seal clearance results in a drastic reduction of seal life because of high wear and heat generation. Therefore, in the design of mechanical face seals the compromise between low leakage and acceptable life is important and presents a difficult design problem. And the gap geometry of seal clearance affects seal performance very much and becomes an important design variable. In this study the Reynolds equation for the sealing dam of mechanical face seals is numerically analyzed using the Galerkin Finite Element Method, which can be readily applied to various seal geometries. The film pressures of the sealing dam are analyzed, including the effects of the seal face coning and tilt. Then, opening forces, restoring moments, leakages, and dynamic coefficients are calculated.

• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.173-180
• /
• 2018

This paper proposes a method to improve the performance of face recognition via super-resolution method using sparse representation and deep learning from low-resolution facial images. Recently, there have been many researches on ultra-high-resolution images using deep learning techniques, but studies are still under way in real-time face recognition. In this paper, we combine the sparse representation and deep learning to generate super-resolution images to improve the performance of face recognition. We have also improved the processing speed by designing in parallel structure when applying sparse representation. Finally, experimental results show that the proposed method is superior to conventional methods on various images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.151-154
• /
• 2003

In this paper, we propose a method to recognize a face image of the 3D using the LVQ network. LVQ network of the proposed method, We used the front view of a face image to get to a coded light to a training data, can group a face image including the side of various angle. For an usefulness authentication of this algorithm, Various experiment which classifies a face image of the angle was the low.