• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.39-51
• /
• 2022

Image-based gender classification and age estimation of human are classic problems in computer vision. Most of researches in this field focus just only one task of either gender classification or age estimation and most of the reported methods for each task focus on accuracy performance and are not computationally light. Thus, running both tasks together simultaneously on low cost mobile or embedded systems with limited cpu processing speed and memory capacity are practically prohibited. In this paper, we propose a novel light-weight gender classification and age estimation method based on ensemble multitasking deep learning with light-weight processing neural network architecture, which processes both gender classification and age estimation simultaneously and in real-time even for embedded systems. Through experiments over various well-known datasets, it is shown that the proposed method performs comparably to the state-of-the-art gender classification and/or age estimation methods with respect to accuracy and runs fast enough (average 14fps) on a Jestson Nano embedded board.

• International journal of advanced smart convergence
• /
• 제9권2호
• /
• pp.203-211
• /
• 2020

This study proposes a method for estimating gender and age that is robust to various external environment changes by applying deep learning-based learning. To improve the accuracy of the proposed algorithm, an improved CNN network structure and learning method are described, and the performance of the algorithm is also evaluated. In this study, in order to improve the learning method based on CNN composed of 6 layers of hidden layers, a network using GoogLeNet's inception module was constructed. As a result of the experiment, the age estimation accuracy of 5,328 images for the performance test of the age estimation method is about 85%, and the gender estimation accuracy is about 98%. It is expected that real-time age recognition will be possible beyond feature extraction of face images if studies on the construction of a larger data set, pre-processing methods, and various network structures and activation functions have been made to classify the age classes that are further subdivided according to age.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2017년도 춘계학술대회
• /
• pp.779-782
• /
• 2017

OpenBR은 안면인식 관련 새로운 방식의 연구, 기존 알고리즘 개선, 상용 시스템과 상호 작용, 인식 성능 측정, 자동화 된 생체 인식 시스템을 배치하기 위한 프레임 워크입니다. 신속한 알고리즘 프로토타이핑을 용이하게 하기 위해 고안되었으며 성숙한 핵심 프레임 워크, 유연한 플러그인 시스템 및 개방형 및 폐쇄형 소스 개발 지원을 특징으로 한다. 기성의 알고리즘은 얼굴 인식, 연령 산정 및 성별 추정과 같은 특정 양식에 대해서도 사용할 수 있다. 본 논문에서는 OpenBR의 프레임 워크의 구성방법에 대해서 기술하고 지원되는 프로그램을 통해서 이용한 안면인식, 성별추정, 나이추정 구현하고 기술하였다.

• 한국가정과교육학회지
• /
• 제19권3호
• /
• pp.91-106
• /
• 2007

The purpose of this study was to provide the information for the development of gender-equality oriented content of Technology Home Economics subject. For this purpose, a total of 404 male and female middle high school students in Chuncheon city were sampled and asked to estimate the degree of practical use and preference for the 47 units of Technology Home Economics subject. Results were summarized as following : 1. The 47 units were classified into 4 groups on the basis of similarity in the degree of practical use and preference: 23 units estimated as 'better than average' by male and female students were classified into group 1; 4 units estimated as 'better than average' by female students but as 'less than average' by male students were classified into group 2; 10 units estimated as 'less than average' by male and female students were classified into group. 3; 10 units estimated as 'far less than average' by male and female students were classified into group 4. Most of the units in Home Economics area were classified Into group 1 or 2, but most of the units in Technology area were classified into group 3 or 4. 2. Gender difference was confirmed between male and female students' estimation of the degree of practical use and preference for the 47 units. In about three-quaters of the units in Home Economics area, female students' estimation of the degree of practical use and preference was higher than male students' estimation. In about half of the units in Technology area, male students' estimation of the degree of practical use and preference was higher than female students' estimation. However, possibility was detected in several units of Technology Home Economics subject that gender difference could be decreased.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.483-486
• /
• 2020

Age estimation and gender classification for human is a classic problem in computer vision. Almost research focus just only one task and the models are too heavy to run on low-cost system. In our research, we aim to apply multitasking learning to perform both task on a lightweight model which can achieve good precision on embedded system in the real time.

• Journal of Nutrition and Health
• /
• 제38권4호
• /
• pp.320-326
• /
• 2005

This study was conducted to estimate dietary choline intake of Koreans according to gender, age and region by using the data on per capita food intake in 1998 and 2001 and the data on choline contents of foods customarily eaten by Koreans. Sixty-four and 97 kinds of food items were used for the estimation of choline intake in 1998 and 2001, respectively. And these corresponded to approximately $90\%$ of the total amount of food intake and energy intake. Based on these food items and choline content, per capita choline intake of Koreans was estimated to be 623.0 mg and 602.4 mg in 1998 and 2001 respectively , 687.2 mg and 659.1 mg in 1998 and 2001 for male, 560.2 mg and 551.0 mg in 1998 and 2001 for female. While the number of foods considered in estimation was higher in 2001 than 1998, choline intake was lower in 2001. The estimated per capita choline intake of male population was higher than that of female population by 108-127 mg, possibly due to the difference in total amount of food intake. The age group with the highest choline intake was different between 2 time points. Although adolescents of 13- 19 years showed the highest intake (712.0 mg) in 1998, mid-aged adults of 30-49 years showed the highest choline intake (662.2 mg) followed by adolescents (645.7 mg) in 2001. In regional comparison, choline intake of the rural population (588.7 mg and 588.3 mg in 1998 and 2001 respectively) was lower than that of population in small to medium-size cities (625.1 mg and 603.9 mg) or in metropolitan cities (637.7 mg and 602.8 mg). This result revealed that per capita choline intake of the Korean population falls within the adequate range with some differences by age, sex and residing area.

• 노동경제논집
• /
• 제43권1호
• /
• pp.93-114
• /
• 2020

롤모델 효과에 따라 여성 교수의 존재는 여학생에게 유의한 영향을 미칠 수 있다. 특히 남학생이 많은 '남초학과'로 전공을 선택하거나, 대학원으로의 진학을 결정하는 데 여성 교수의 존재가 연관이 있을 수 있다. 다만 학부 입학, 박사과정 학생의 성취도 등과 관련하여 교수의 성별이 여학생에게 미치는 영향을 분석한 연구는 있으나 교수의 성비에 따른 여학생 취업에 관한 실증적인 연구가 부족하여 실제로 여성 교원의 비율이 여학생의 취업에 얼마나 영향을 주는지 판단하기 어렵다. 이에 본 연구는 여성 교수의 존재가 여성 대졸자의 취업에 미치는 영향을 분석하였다. 대학알리미 자료를 이용하여 분석한 결과, 합동회귀모형 추정에서는 전임교원 중 여성의 비율이 1%포인트 증가하면 여학생의 취업률이 0.102%포인트 유의하게 상승하는 것으로 나타났으나 학교-학과의 고정효과를 고려한 분석에서는 유의한 결과가 나타나지 않았다. 여성 교원의 비율이 여학생의 취업률에 미치는 영향을 전공계열별로 구분하여 살펴보아도 이와 유사한 결과가 도출되었다. 학생의 취업률은 본 연구에서 사용한 변수 외에 관측되지 않는 개인의 성향, 능력, 가정환경 등 여러 기타 요인에 영향을 받을 수 있다. 즉, 본 연구의 추정에서는 우리나라 대학의 여학생 취업에 있어서 여성 교수의 롤모델 효과에 대한 실증적 증거를 발견하지 못하였다.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제59권3호
• /
• pp.217-235
• /
• 2020

본 연구는 TIMSS 2011와 2015 수학 성취도 평가를 바탕으로 초등학교 4학년 남녀 간의 성차가 드러나는 문항들을 내용과 인지영역에 따라 분석하였다. 연구결과, 성차가 유의미하게 차이나는 20% 문항들에서 내용 영역별 성차를 보면 남학생은 수와 측정영역에서 여학생은 도형영역에서 상대적으로 우위에 있는 것으로 나타났다. 또한 인지 영역별 성차를 보면, 남학생이 지식, 적용, 추론 영역에서 여학생보다 상대적인 우위에 있는 것으로 드러났다.

• 대한인간공학회지
• /
• 제30권4호
• /
• pp.535-540
• /
• 2011

Objective: The aim of this paper is to survey the gender differences Work-related musculoskeletal disorders among agriculture workers. Background: The number of occupational injuries has been decreasing trend in Korea, but it has been increasing in agriculture. The ratio of women worker shows steady increasing trend among the agricultural worker in Korea. Method: This study investigated through the questionnaire to 358 agricultural workers who are working for agriculture located in Gyeongnam, Gyeongbuk and Jeju. Results: The task types occurring frequently and the weight of a heavy object handled according to the sex were showed statistically significant difference. The estimation of WMSDs patients by sex appeared statistically very significant difference. Conclusion: The prevalence rate of WMSDs in female was about 21 percent higher than the prevalence rate of WMSDs in male. Application: The result of this study will help improvement of work environment and prevention of WMSDs in agricultural work.

• Journal of Preventive Medicine and Public Health
• /
• 제42권4호
• /
• pp.215-222
• /
• 2009

Objectives : Vigorous physical activity is a well-known method to promote people s health. This research aims to investigate whether perceived neighborhood characteristics affect vigorous physical activity among adult Seoul residents (aged 19 to 64). Methods : Utilizing the 2005 Seoul Citizens Health Indicators Survey data, this study estimates the probability of vigorous physical activity. Particular attention is given to the effects of three perceived neighborhood characteristics satisfaction with relationship to the neighborhood, satisfaction with park and recreational facilities, and satisfaction with public security). Logistic regression models are analyzed separately by gender for the parameter estimation. Results : Vigorous physical activity is positively associated with three perceived neighborhood characteristics for women, while neither significant nor substantive association is found for men. Conclusions : As vigorous physical activity among Seoul citizens is differentially affected by perceived neighborhood characteristics and by gender, a different approach will be needed to increase vigorous physical activity of men and women in Seoul.