• Journal of the Korea Society of Computer and Information
• /
• v.15 no.4
• /
• pp.47-55
• /
• 2010

In this paper, we proposed a recognition system of smile facial expression for smile treatment training. The proposed system detects face candidate regions by using Haar-like features from camera images. After that, it verifies if the detected face candidate region is a face or non-face by using SVM(Support Vector Machine) classification. For the detected face image, it applies illumination normalization based on histogram matching in order to minimize the effect of illumination change. In the facial expression recognition step, it computes facial feature vector by using PCA(Principal Component Analysis) and recognizes smile expression by using a multilayer perceptron artificial network. The proposed system let the user train smile expression by recognizing the user's smile expression in real-time and displaying the amount of smile expression. Experimental result show that the proposed system improve the correct recognition rate by using face region verification based on SVM and using illumination normalization based on histogram matching.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.33 no.2
• /
• pp.145-155
• /
• 2023

Objectives: A fit test panel is needed to identify the fit performance of a respirator and its face seal. This is a criterion for selecting subjects that can represent the facial characteristics of users. Although anthropometry data has been developed for people in United States and China it is not yet present in Korea. This study aimed to develop a Korean fit test panel and test headform. Methods: For the 7th and 8th waves of the Size Korea anthropometry data, facial measurements of 11,429 people aged 15 to 69 years were used for analysis. PCA and bivariate panel were classified using the ISO16976-2:2022(E) anthropometrics analysis method. Based on this result, a static headform was developemed and a fit test chamber was constructed. Results: Of the 11,429 Korean people used for principal component analysis, 11,300 were included in the ellipse, marking an acceptance rate of 98.87% on PCA panel. The face types were classified into five types. Among them, a large, medium, and small static headform were printed using a 3D printer. In addition, 10,985 people (96.12%) were included in the bivariate panel based on face length and face width. The y-axis (face length) boundary was 97.87 to 134.59 mm, and the x-axis (face width) boundary was 120.75 to 158.23 mm. Conclusions: Compared to the ISO analysis, the Korean principal component was narrower in the width item (PC1) and longer in the length item (PC2). For the future, it is necessary to conduct a fit test using the developed headform and chamber device to confirm the usefulness of this Korean test panel. Therefore, this study is considered valuable as basic research for Korean test panels.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.120-121
• /
• 2006

In this paper, We have designed a face recognition system based on the embedded Linux. This paper has an aim in embedded system to recognize the face more exactly. At first, the contrast of the face image is adjusted with lightening compensation method, the skin and lip color is founded based on YCbCr values from the compensated image. To take advantage of the method based on feature and appearance, these methods are applied to the eyes which has the most highly recognition rate of all the part of the human face. For eyes detecting, which is the most important component of the face recognition, we calculate the horizontal gradient of the face image and the maximum value. This part of the face is resized for fitting the eye image. The image, which is resized for fit to the eye image stored to be compared, is extracted to be the feature vectors using the continuous wavelet transform and these vectors are decided to be whether the same person or not with PNN, to miminize the error rate, the accuracy is analyzed due to the rotation or movement of the face. Also last part of this paper we represent many cases to prove the algorithm contains the feature vector extraction and accuracy of the comparison method.

• Journal of Korea Multimedia Society
• /
• v.8 no.8
• /
• pp.1057-1068
• /
• 2005

This paper proposes a real-time face detection system which detects multiple faces from low resolution video such as web-camera video. First, It finds face region candidates by using AdaBoost based object detection method which selects a small number of critical features from a larger set. Next, it generates reduced feature vector for each face region candidate by using principle component analysis. Finally, it classifies if the candidate is a face or non-face by using SVM(Support Vector Machine) based binary classification. According to experiment results, the proposed method achieves real-time face detection from low resolution video. Also, it reduces the false detection rate than existing methods by using PCA and SVM based face classification step.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1310-1319
• /
• 2016

Face detection is the first step in a wide range of face applications. However, detecting faces in the wild is still a challenging task due to the wide range of variations in pose, scale, and occlusions. Recently, many deep learning methods have been proposed for face detection. However, further improvements are required in the wild. Another important issue to be considered in the face detection is the computational complexity. Current state-of-the-art deep learning methods require a large number of patches to deal with varying scales and the arbitrary image sizes, which result in an increased computational complexity. To reduce the complexity while achieving better detection accuracy, we propose a fully convolutional network-based face detection that can take arbitrarily-sized input and produce feature maps (heat maps) corresponding to the input image size. To deal with the various face scales, a multi-scale network architecture that utilizes the facial components when learning the feature maps is proposed. On top of it, we design multi-task learning technique to improve detection performance. Extensive experiments have been conducted on the FDDB dataset. The experimental results show that the proposed method outperforms state-of-the-art methods with the accuracy of 82.33% at 517 false alarms, while improving computational efficiency significantly.

• Geomechanics and Engineering
• /
• v.30 no.4
• /
• pp.363-372
• /
• 2022

One of the most frequent issues in tunnel excavation is the collapse of rock blocks and the dropping of rock fragments from the tunnel face. The tunnel face can be reinforced using a number of techniques. One of the most popular and affordable solutions is the use of face longitudinal dowels, which has benefits including high strength, flexibility, and ease of cutting. In order to examine the reinforced face, this work shows the longitudinal deformation profile and ground response curve for a tunnel face. This approach is based on assumptions made during the analysis phase of problem solving. By knowing the tunnel face response and dowel behavior, the interaction of two elements can be solved. The rock element equation derived from the rock bolt method is combined with the dowel differential equation to solve the reinforced ground response curve (GRC). With a straightforward and accurate analytical equation, the new differential equation produces the reinforced displacement of the tunnel face at each stage of excavation. With simple equations and a less involved computational process, this approach offers quick and accurate solutions. The FLAC3D simulation has been compared with the suggested analytical approach. A logical error is apparent from the discrepancies between the two solutions. Each component of the equation's effect has also been described.

• The Journal of Industrial Distribution & Business
• /
• v.12 no.11
• /
• pp.17-31
• /
• 2021

Purpose: Despite the fact that face (i.e. Chemyon) is deeply-rooted in Korean culture and significantly affects the behavior of Korean people, the effect of face on leisure consumers' consumption behavior has only reported mixed findings, that is, significant and/or insignificant face effects have been reported. It is necessary to integrate prior research findings, and comprehensively examine the effect of face on leisure consumers' consumption behavior. The purpose of this study was to investigate the effect (i.e. effect size, and moderating variables) of face on leisure consumers' consumption behavior through meta-analysis. Research design, data and methodology: Among 1,019 face-related academic studies, retrieved from the academic research information services (RISS), 34 studies and 300 cases examining the effect of face on leisure consumers' consumption behavior were finally included for meta-analysis. Face measured as face sensitivity and/or a face sensitivity sub-component (shame-consciousness, formality-consciousness, and other-consciousness) were integrated in the meta-analysis. Leisure consumers' consumption behavior was classified as antecedents of purchase (overall conspicuous consumption tendency, overall symbolic consumption tendency, personality, high price, high quality, brand seeking, fashion seeking, enjoyment, other person (interpersonal) consideration, position, reference group, and attitude), purchase (purchase intention, unplanned purchase, purchase, and expenditure), and post-purchase (satisfaction, repurchase, and post-purchase). The data used in the meta-analysis was comprised of correlation coefficients, and the meta-analysis was performed using the R-program. Results: The overall mean effect size of face on leisure consumers' consumption behavior was .248. It was found that the effect size was the largest in the order of shame-consciousness face, formality-consciousness face, and other-consciousness face. Among the types of leisure consumers' consumption behavior categorized as dependent variables, the effect size was found to be largest in the order of position, attitude, reference group, post-purchase behavior, brand seeking, personality, trend seeking, etc. In addition, it was found that the leisure types moderated the effect size of face on leisure consumers' consumption behavior. The effect size was found to be largest in the order of skin diving, baseball, various leisure participation, dance, gambling, golf, etc. Conclusions: Face moderately or significantly influence leisure consumers' consumption behavior.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.891-894
• /
• 1998

This paper introduces an automatic face region extraction method. This method consists of two part: face recognition and extraction of facial organs which are eye, eyebrow, nose and mouth. In first stage, we use genetic algorithms(GAs) to get face region in complex background. In second stage, we use Geometrical Face Model to textract eye, eyebrow, nose and mouth. In both stage, stochastic component is used to deal with the problems caused by had lighting condition. According to this value, blurring number is determined. Average Computation time is less than 1 sec, and using this method we can extract facial feature efficiently from several images which has different lightning condition.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.73-78
• /
• 1994

Wear and fracture mode of ceramic tool for hardened SKD11 steel was investigated by face milling in this study. The cutting force and Acoustic Emission(AE) signal were utilized to detect the wear and fracture of ceramic tool. The following conclusions were obtained : (1) The wear and fracture modes of ceramic tool are characterized by three types: \circled1wear which has normal wear and notch wear, \circled2 wear caused by scooping on the rake face, \circled3 large fracture caused by thermal crack in the rake face. (2) The wear behaviour of ceramic tool can be detected by the increase of mean cutting force and the variation of the AE RMS voltage. (3) The catastrophic fracture of ceramic tool can be detected by the cutting force(Fz-component). (4) As the hardness of work material increased, Acoustic Emission RMS value and mean cutting force(Fz) increased linearly, but the tool life decreased.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.11a
• /
• pp.240-243
• /
• 2002

In this paper, we propose a face region detection based on skin-color distribution and facial feature extraction algorithm in color still images. To extract face region, we transform color using general skin-color distribution. Facial features are extracted by edge transformation. This detection process reduces calculation time by a scale-down scanning from segmented region. we can detect face region in various facial Expression, skin-color deference and tilted face images.