• Journal of the Institute of Electronics and Information Engineers
• /
• 제51권2호
• /
• pp.69-75
• /
• 2014

Existing People Counting System using a single overhead mounted camera has limitation in object recognition and counting in various environments. Those limitations are attributable to overlapping, occlusion and external factors, such as over-sized belongings and dramatic light change. Thus, this paper proposes the new concept of People Counting System by Facial Age Group using two depth cameras, at overhead and frontal viewpoints, in order to improve object recognition accuracy and robust people counting to external factors. The proposed system is counting the pedestrians by five process such as overhead image processing, frontal image processing, identical object recognition, facial age group classification and in-coming/out-going counting. The proposed system developed by C++, OpenCV and Kinect SDK, and it target group of 40 people(10 people by each age group) was setup for People Counting and Facial Age Group classification performance evaluation. The experimental results indicated approximately 98% accuracy in People Counting and 74.23% accuracy in the Facial Age Group classification.

• Journal of Convergence for Information Technology
• /
• 제11권7호
• /
• pp.1-6
• /
• 2021

The purpose of this study is to improve the performance of the artificial neural network system for facial image analysis through the image landmark selection technique. For landmark selection, a CNN-based multi-layer ResNet model for classification of facial image age is required. From the configured ResNet model, a heat map that detects the change of the output node according to the change of the input node is extracted. By combining a plurality of extracted heat maps, facial landmarks related to age classification prediction are created. The importance of each pixel location can be analyzed through facial landmarks. In addition, by removing the pixels with low weights, a significant amount of input data can be reduced.

• Journal of Advanced Navigation Technology
• /
• 제16권1호
• /
• pp.151-157
• /
• 2012

Recently, owing to the technology advances in computer science and image processing, age of face classification have become prevalent topics. It is difficult to estimate age of facial shape with statistical figures because facial shape of the person should change due to not only biological gene but also personal habits. In this paper, we proposed a robust age of face classification method by using Gabor feature and fuzzy support vector machine(SVM). Gabor wavelet function is used for extracting facial feature vector and in order to solve the intrinsic age ambiguity problem, a fuzzy support vector machine(FSVM) is introduced. By utilizing the FSVM age membership functions is defined. Some experiments have conducted to testify the proposed approach and experimental results showed that the proposed method can achieve better age of face classification precision.

• Fashion & Textile Research Journal
• /
• 제11권1호
• /
• pp.103-111
• /
• 2009

This study is to help make-up and coordination for image-making after analysis of facial color and shape of elderly women. The data was analyzed 55-75 years old 212 elderly women's face color and pictures by means of SPSS 12.0 statistics package. On the basis of the colorimetric data on face by Minolta CM2500D, this research considered the analysis of facial color, patternization of facial color and its analysis by age group; for the analysis of facial shape, this research patternized facial shape and analyzed its characteristic using both contour-based facial shape analysis and Kamata facial shape analysis. As for facial color, it was found that the lower age bracket has bright and reddish face, looking fine, while the higher age bracket has a conspicuously yellowish face, looking bad. The community of facial color is classified as 3 types and it was found out that the facial color of the subjects belonging to Type 3, whose L value is the largest, looked the brightest; the face of the subjects belonging to Type 2, whose a value is the largest, was much tinged with red, and the face of the subjects belonging to Type 1, whose b value is the largest were tinged with yellow. According to the analysis of facial shape, there appeared oval & long forms in the classification by contour, while there appeared a lot of downward-directed power and inner-directed power in the classification by Kamata, which is believed to reflect the phenomenon that their chin line becomes roundish and the facial length also tend to be longer due to aging.

• Journal of Broadcast Engineering
• /
• 제28권2호
• /
• pp.238-241
• /
• 2023

Recently deep learning models are widely used for various tasks such as facial recognition and face editing. Their training process often involves a dataset with imbalanced age distribution. It is because some age groups (teenagers and middle age) are more socially active and tends to have more data compared to the less socially active age groups (children and elderly). This imbalanced age distribution may negatively impact the deep learning training process or the model performance when tested against those age groups with less data. To this end, we propose an age-controllable face synthesis technique using a feature decomposition to classify age from facial images which can be utilized to synthesize novel data to balance out the age distribution. We perform extensive qualitative and quantitative evaluation on our proposed technique using the FFHQ dataset and we show that our method has better performance than existing method.

• The korean journal of orthodontics
• /
• 제22권4호
• /
• pp.881-920
• /
• 1992

This study was designed to assess profile preferences among Korean youths in the year 1992. Facial esthetics was evaluated by means of silhouette profiles, eliminating the influence of a number of aspects that may affect judgment when normal lateral photographs are used. The main points of preference to be clarified here are as follows. First, on facial convexity, Second, on nasion depth, Third, on mentolabial sulcus depth, Fourth, on the position of upper and lower lips, Fifth, on facial type according to Angle's classification of malocclusion, Sixth, on Song's tangents. The 54 subjects printed in questionnaire as black and white silhouettes were selected from 300 tracings from cephalometric radiographs of people whose age ranging from 11 to 20 years. Photographs of six female subjects were retouched by computer graphic software and printed in color and black/white photographs which were used for adaptation of eyes of participants in selecting profiles in silhouette. They constitute 2 questions. The 54 subjects were grouped as 22 questions, each of them composed of 6 subjects, according to the aspects to be clarified. Twenty four questions in total were asked to assess profile preferences. For the assessment, the profile line, the facial esthetic triangle, Song's tangents, and Angle's classification of malocclusion were introduced. The profile line is composed of 11 component points which are Trichion, Glabella, Nasion, Pronasale, Subnasale, Labrale superius, Stomion, Labrale inferius, Supramentale, Pogonion, and Gnathion. The facial esthetic triangle is composed of 3 tangents: A-tangent which is the tangent of dorsum of nose, B-tangent which is the line passing through Sn and Ls, and C-tangent which is drawn on the turning point of the curve which lies between mentolabial sulcus (Sm) and pogonion (Pg). Angle's classification has 3 types of malocclusion which are Class I, Class II, and Class III. Class II malocclusion is subdivided into Division 1 and Division 2. The participants of the survey were composed of 861 college students (448 male students, 413 female students) whose majors grouped as Fine Arts. Liberal Arts, and Natural Sciences, and whose mean age 21.8 years. The statistics program SPSS/PC + of SPSS Inc. was used to analyze answers of participants. Crosstabulation, Chi-square test, and Kendall test were done. The conclusions are as follows: First, Korean youths have a tendency to prefer the slightly convex face to the flat or concave face. Second, they prefer a moderately deep nasion. Third, they prefer a moderately deep mentolabial sulcus. Fourth, they prefer the position of lips which are near to Ricketts' E-line. The position of the upper lip which is slightly posterior to E-line is preferred. The upper lip which lies too far anterior or posterior to the lower lip is not perferred. Fifth, they prefer most, according to Angle's Classification of Malocclusion, Class I facial profile which has a slight inclination to Class II division 2. The order of preference is Class I, Class II division 2, Class III, and Class II division 1. Sixth, they prefer the type 2 and 3 of Song's tangents. The facial profile within which A-and B-tangent meet is preferred. The facial profile which has Cotangent that .meets with A-tangent slightly posterior to the crossing point of A-and B-tangent or that parallels with B-tangent is preferred.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• 제41권11호
• /
• pp.1035-1045
• /
• 2017

Classification of age and gender has been carried out through different approaches such as facial-based and audio-based classifications. One of the limitations of facial-based methods is the reduced recognition rate over large distances, while another is the prerequisite of the faces to be located in front of the camera. Similarly, in audio-based methods, the recognition rate is reduced in a noisy environment. In contrast, gait-based methods are only required that a target person is in the camera. In previous works, the view point of a camera is only available as a side view and gait data sets consist of a standard gait, which is different from an ordinary gait in a real environment. We propose a feature extraction method using skeleton models from an RGB-D sensor by considering characteristics of age and gender using ordinary gait. Experimental results show that the proposed method could efficiently classify age and gender within a target group of individuals in real-life environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권7호
• /
• pp.2480-2495
• /
• 2014

This paper describes a hierarchical method for image-based age estimation that combines age group classification and age value estimation. The proposed method uses a coarse-to-fine strategy with different appearance features to describe facial shape and texture. Considering the damage to continuity between neighboring groups caused by fixed divisions during age group classification, a dynamic grouping technique is employed to allow non-fixed groups. Based on the given group, an ordinal hyperplane ranking (OHRank) model is employed to transform age estimation into a series of binary enquiry problems that can take advantage of the intrinsic correlation and ordinal information of age. A set of experiments on FG-NET are presented and the results demonstrate the validity of our solution.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 한국컴퓨터정보학회 2024년도 제69차 동계학술대회논문집 32권1호
• /
• pp.343-345
• /
• 2024

얼굴 연령 분류 기법은 신원 확인 시스템 고도화, 유동 인구 통계 자동화 시스템 구축, 연령 제한 콘텐츠 관리 시스템 고도화 등 다양한 분야에 적용할 수 있는 확장 가능성을 가진다. 넓은 확장 가능성을 가지는 만큼 적용된 시스템의 안정성을 위해서는 얼굴 연령 분류 기법의 높은 정확도는 필수적이다. 따라서, 본 논문에서는 Vision Transformer(ViT) 기반 분류 알고리즘의 얼굴 연령 분류 성능을 비교 분석한다. ViT 기반분류 알고리즘으로는 최근 널리 사용되고 있는 ViT, Swin Transformer(ST), Neighborhood Attention Transformer(NAT) 세 가지로 선정하였으며, ViT의 얼굴 연령 분류 정확도 65.19%의 성능을 확인하였다.

• Proceedings of the KAIS Fall Conference
• /
• 한국산학기술학회 2010년도 추계학술발표논문집 2부
• /
• pp.878-881
• /
• 2010

For better diagnosis and prescription in Korean traditional medicine, Lee Je-Ma (1837-1900) created Sasang Constitution classification which was divided into four groups of Taeyangin, Soyangin, Taeumin and Soumin based on both body shape and natural disposition. The purpose of this study was to investigate the characteristics of facial skin parameters (hydration, lipid and pH) on forehead and cheek according to Sasang Constitution classifications of Taeumin, Soyangin and Soumin in Korean. Eighty-nine Korean female subjects were recruited for this study and the average age of them was 19.9${\pm}$0.84 years. The four groups by the Sasang Constitution were classified by questionnaire for the Sasang Constitution classification proposed by Kyung-Hee Oriental Medicine Hospital. Consequently, thirty-eight (42.7%) among the subjects were grouped into Soumin, twenty-nine (32.6%) into Taeumin, twenty (22.5%) into Soyangin and two (2%) into Taeyangin. Taeyangin group was excluded from statistical analysis due to small subjects. Hydration, lipid and pH parameters on forehead and cheek were measured by using non-invasive instruments of Corneometer (CM 825, Schwarzhaup, Germany), Sebumeter (SM 815, Schwarzhaup, Germany) and Skin-pH-meter (pH 905, Schwarzhaup, Germany), respectively. The measurements by the same investigator were performed under standardized condition with a room temperature of $21^{\circ}C$ and a humidity level of 40% to 50%. As a result, hydration (F=25.481, p=.000), lipid (F=5.753, p=.005) and pH (F=5.010, p=.009) of the forehead skin showed significant differences in the order of Taeumin, Soyangin and Soumin. Hydration (F=23.216, p=.000), lipid (F=6.898 p=.002) and pH (F=5.070, p=.008) of the cheek skin showed significant differences in the order of Taeumin, Soyangin and Soumin. In conclusion, facial skin surface seemed to be dependent on Sasang Constitution classification in Korean. These findings indicated that Sasang Constitution classification might be an useful esthetic treatment for caring facial skin in the future.