• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.7
• /
• pp.821-826
• /
• 1999

This paper proposes a face recognition method using feature spaces and image vectors in the image plane. We obtain the 2-D feature space using the self-organizing map which has two inputs from the axis of the given image. The image vector consists of its weights and the average gray levels in the feature space. Also, we can reconstruct an normalized face by using the image vector having no connection with the size of the given face image. In the proposed method, each face is recognized with the best match of the feature spaces and the maximum match of the normally retrieval face images, respectively. For enhancing recognition rates, our method combines the two recognition methods by the feature spaces and the retrieval images. Simulations are conducted on the ORL(Olivetti Research laboratory) images of 40 persons, in which each person has 10 facial images, and the result shows 100% recognition and 14.5% rejection rates for the 20$\times$20 feature sizes and the 24$\times$28 retrieval image size.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2020-2023
• /
• 2003

In this paper, our proposed system uses the regional Gabor wavelet and Neural Network to implement face recognition similar to human face recognition system, because the Gator wavelet expresses visual recognition system of human mathematically and the regional Neural Network is robust to white noise and partial illumination. This system consists of two stages of building database and recognizing face. One is composed by using the supervised learning of Neural Network. At this time, the Neural Network is applied to the upper and the lower part of face images respectively. The Backpropagation algorithm is used to learn Neural Network. Another consists of calibration of slope of face image, measurement of illumination variant using deviation with average face image and similarity comparison using Euclidean distance measure.

• Journal of radiological science and technology
• /
• v.44 no.6
• /
• pp.679-688
• /
• 2021

This study is a questionnaire about the lesson environment that radiation major students prefer in a non-face-to-face live online lesson environment for a total of 133 students, 65 second graders and 68 third graders who are enrolled in the department of radiology at a university located in the Seoul metropolitan area. And checked the satisfactory level by grade. The questionnaire consists of three categories: 1st real-time non-face-to-face lectures, 2nd professor lectures, and 3rd corona lectures. A total of 14 questions, with multiple choice and descriptive response methods. As an evaluation method, in the case of a multiple-choice question, the average was calculated using a 5-point Likert scale. As a result of conducting the independent sample T-test of the SPSS program, the response by grade was P > 0.05, and no significant result was shown by the contents of the questionnaire survey of the second grade. As for the lecture method of the department of radiology after the end of Covid-19 virus, it is better to promote face-to-face lessons in radiation training subjects and non-face-to-face real-time education in subjects centered on radiation theory.

• Journal of the Korean Society of Costume
• /
• v.57 no.5 s.114
• /
• pp.151-165
• /
• 2007

Appearance matters in society today. Women want to feel good and look their best. They do make-up, wear garment and accessory for their good looking. Doing make-up, we have to know how we are look and to consider face shape. But it is difficult to recognize face shape. Because there is no standard face shape of adult women of quantitative analysis. The purpose of this study was to offer standard face shape of adult women in Korea. Furthermore, the study was to determine and differentiate face shape of each age group to set the basic data for the Korean beauty industry. In this study, photographs of 600 Korean women, age between $20{\sim}50's$, were indirectly measured in Venus face2D program. The measurements were analyzed by statistical methods. As a result of basic statistical data analysis, the average lengths of face were 196mm, lengths of forehead-hairline between eyebrows were 62mm, lengths of eyebrow between noses were 68mm, length of nose between chin were 66mm, and width of face were 150mm. By comparing to each age group's face using ANOVA, the statistically noticeable differences were found in measurements.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.11C
• /
• pp.1040-1048
• /
• 2009

This paper proposes an auto flesh tone balance algorithm for the picture that is taken for people. General white balance algorithms bring neutral region into focus. But, other objects can be basis if its spectral reflectance is known. In this paper the basis for white balance is human face. For experiment, first, transfer characteristic of image sensor is analyzed and camera output RGB on average face chromaticity under standard illumination is calculated. Second, Output rate for the image is adjusted to make RGB rate for the face photo area taken under unknown illumination RGB rate that is already calculated. Input tri-stimulus XYZ can be calculated from camera output RGB by camera transfer matrix. And input tri-stimulus XYZ is transformed to standard color space (sRGB) using sRGB transfer matrix. For display, RGB data is encoded as eight-bit data after gamma correction. Algorithm is applied to average face color that is light skin color of Macbeth color chart and average color of various face colors that are actually measured.

• Archives of Plastic Surgery
• /
• v.46 no.5
• /
• pp.405-413
• /
• 2019

Background Face transplantation has naturally evolved from reconstructive procedures. However, few institutions perform face transplantations, because it is time-consuming and it is necessary to justify non-vital organ transplantation. We investigated the process of organ donation from brain-dead patients and the possibility of incorporating face transplantation into the donation process. Methods A retrospective review was performed of 1,074 brain-dead patients from January 2015 to December 2016 in Korea. We analyzed the time intervals from admission to brain death decisions (first, second, and final), the causes of brain death, and the state of the transplanted organs. Results The patient base (n=1,074) was composed of 747 males and 327 females. The average period between admission to the first brain death decision was 8.5 days (${\pm}15.3$). The average time intervals between the first brain death decision and medical confirmation using electroencephalography and between the first brain death decision and the final determination of brain death were 16 hours 58 minutes (${\pm}14hours$ 50 minutes) and 22 hours 57 minutes (${\pm}16hours$ 16 minutes), respectively. The most common cause of brain death was cerebral hemorrhage/stroke (42.3%), followed by hypoxia (30.1%), and head trauma (25.2%). Conclusions When face transplantation is performed, the transplantation team has 22 hours 57 minutes on average to prepare after the first brain death decision. The cause of brain death was head trauma in approximately one-fourth of cases. Although head trauma does not always imply facial trauma, surgeons should be aware that the facial tissue may be compromised in such cases.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2667-2682
• /
• 2015

Due to the wide application of face recognition (FR) in information security, surveillance, access control and others, it has received significantly increased attention from both the academic and industrial communities during the past several decades. However, partial face occlusion is one of the most challenging problems in face recognition issue. In this paper, a novel method based on linear regression-based classification (LRC) algorithm is proposed to address this problem. After all images are downsampled and divided into several blocks, we exploit the evaluator of each block to determine the clear blocks of the test face image by using linear regression technique. Then, the remained uncontaminated blocks are utilized to partial occluded face recognition issue. Furthermore, an improved Distance-based Evidence Fusion approach is proposed to decide in favor of the class with average value of corresponding minimum distance. Since this occlusion removing process uses a simple linear regression approach, the completely computational cost approximately equals to LRC and much lower than sparse representation-based classification (SRC) and extended-SRC (eSRC). Based on the experimental results on both AR face database and extended Yale B face database, it demonstrates the effectiveness of the proposed method on issue of partial occluded face recognition and the performance is satisfactory. Through the comparison with the conventional methods (eigenface+NN, fisherfaces+NN) and the state-of-the-art methods (LRC, SRC and eSRC), the proposed method shows better performance and robustness.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.245-250
• /
• 2019

In this paper, we propose an ellipsoid modeling method for coding of a face depth picture. The ellipsoid modeling is firstly based on a point of a nose tip which is defined as the lowest value of the depth in the picture. The proposed ellipsoid representation is simplified through a difference of depth values between in the nose tip and in left or right boundary point of the face. Parameters of the ellipsoid are calculated through coordinates and depth values to minimize differences from the actual depth pixels. A picture is predicted by the modeled ellipsoid for coding of the face depth picture. In simulation results, an average MSEs between the face depth picture and the predicted picture is measured as 20.3.

• Journal of the Korean Society of Radiology
• /
• v.15 no.4
• /
• pp.519-524
• /
• 2021

As the COVID-19 situation continued to spread to the local community along with the spread due to influx, each university had to conduct all online classes and partially online classes. The purpose of this study was to investigate the satisfaction of learners with the content and lecture contents by paying attention to online non-face-to-face classes according to the change of the class environment in the Corona 19 situation. Satisfaction survey on online non-face-to-face class major subjects was analyzed using questionnaires from June 1 to June 11, 2021, targeting 2nd and 3rd year students in the Department of Radiology at S University in Daegu. As a result of the study, satisfaction with content and class content was found to be an average of 3.78 ± 0.75 points, and learning satisfaction was found to be an average of 3.00 ± 1.14 points. In addition, when taking online non-face-to-face classes, the correlation between students' class attitude and content and class content satisfaction was the highest (r=0.555, p<0.01), and it was found that there was also a correlation between content and class content satisfaction and learning satisfaction. (r=0.331, p<0.01). I think that satisfaction with non-face-to-face online classes can be improved if the quality of content is improved during non-face-to-face online major classes as well as more active interactions between students and professors.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.4
• /
• pp.461-468
• /
• 2003

In a sequence of images obtained by surveillance cameras, facial regions appear very small and their colors change abruptly by lighting condition. This paper proposes a new face detection scheme, robust on complex background, small size, and lighting conditions. The proposed method is consisted of three processes. In the first step, the candidates for the face regions are selected using face color distribution and motion information. In the second stage, the non-face regions are removed using face color ratio, boundary ratio, and average of column-wise intensity variation in the candidates. The face regions containing eyes and mouth are segmented and classified, and then they are scored using their topological relations in the last step. To speed up and improve a performance the above process, a block based image segmentation technique is used. The experiments have shown that the proposed algorithm detects faced regions with more than 91% of accuracy and less than 4.3% of false alarm rate.