• 제목/요약/키워드: Face identification

검색결과 304건 처리시간 0.035초

얼굴 인식을 통한 신뢰성 있는 디지털 도어록 제어 시스템 (Reliable Digital Door Lock Control System using Face Recognition)

  • 임지민;김찬;차원석;한태문;허규원;송상근;이상준
    • 전기전자학회논문지
    • /
    • 제17권4호
    • /
    • pp.499-504
    • /
    • 2013
  • IT 기술 발전과 함께 다양한 분야에서 자동화가 진행되고 있다. 홈오토메이션 또한 현재 다양한 형태로 적용되고 있다. 현재 이용 중인 디지털 도어록 시스템은 단순히 비밀번호 입력 및 마그네틱 전자키 방식을 통한 잠금장치의 개폐만을 지원하고 있다. 본 논문에서는 홈오토메이션과 디지털 도어록 시스템 안전성 확보를 위한 지능화된 얼굴인식기술을 활용한 도어록 시스템을 제안한다. 제안 시스템은 원격지에서 방문자의 신원을 확인한 후 이를 바탕으로 잠금장치의 개폐를 조정할 수 있으며, 이를 통해 안전한 출입 통제 관리가 가능하도록 지원해준다.

눈동자를 이용한 사용자 인증기법 (A Scheme for User Authentication using Pupil)

  • 이재욱;강보선;이근호
    • 디지털융복합연구
    • /
    • 제14권9호
    • /
    • pp.325-329
    • /
    • 2016
  • 얼굴인증은 다양한 생체인증 중에서 거부감이 적고 변조가 어려워서 각광받고 있다. 얼굴인증의 알고리즘은 어떻게 알고리즘을 만드느냐에 따라서 정확성과 속도에 많은 차이를 가져온다. 눈동자를 추적 및 검출하여 얼굴의 검출 데이터와 함께 데이터를 추출함으로써 오탐률을 개선하고 정확하게 얼굴로 인증을 할 수 있도록 알고리즘을 연구하였다. Cascade를 통해서 얼굴을 검출하고 관심영역으로 지정 후 얼굴 영역을 균등하게 4등분하여 검출되는 객체의 좌표 값을 저장한다. 또한 검출된 눈에서 눈동자를 검출하기 위하여 이진화를 진행하고 Hough 변환을 통해 눈동자를 검출한다. 추출된 눈동자의 중심좌표를 저장하고 계산하여 데이터 매칭을 통해 얼굴 인증을 한다. 눈동자를 추적과 함께 얼굴의 데이터를 계산하여 정확하고 최적화된 얼굴인증 알고리즘을 연구한다.

동공의 반사특징을 이용한 얼굴위조판별 시스템 (Fake Face Detection System Using Pupil Reflection)

  • 양재준;조성원;정선태
    • 한국지능시스템학회논문지
    • /
    • 제20권5호
    • /
    • pp.645-651
    • /
    • 2010
  • 최근 지능형 범죄가 늘면서 첨단 보안 기술에 대한 요구가 점차 늘어나고 있다. 현재까지 보고된 위조영상검출방법은 실용화를 위하여 정확도 개선이 요구된다. 본 논문에서는 사람의 얼굴에 대하여 동공의 반사광을 이용한 얼굴위조판별 시스템을 제안한다. 제안된 시스템은 먼저 다중 스케일 가버특징 벡터를 기반으로 눈의 위치를 찾은 후 2단계의 템플릿 매칭을 통해서 설정된 적용범위를 벗어나는 눈에 대하여 위조판별을 고려하지 않음으로써 정확도를 높이는 방법을 사용한다. 신뢰도가 확보된 눈의 위치를 기반으로 적외선 조명에 반사되는 동공의 특징을 이용하여 눈위치 근처에서의 화소값을 계산하여 위조 여부를 판단한다. 실험을 통하여 본 논문에서 제안한 방법이 더욱 신뢰성 높은 위조판별시스템임을 확인하였다.

Multimodal Biometrics Recognition from Facial Video with Missing Modalities Using Deep Learning

  • Maity, Sayan;Abdel-Mottaleb, Mohamed;Asfour, Shihab S.
    • Journal of Information Processing Systems
    • /
    • 제16권1호
    • /
    • pp.6-29
    • /
    • 2020
  • Biometrics identification using multiple modalities has attracted the attention of many researchers as it produces more robust and trustworthy results than single modality biometrics. In this paper, we present a novel multimodal recognition system that trains a deep learning network to automatically learn features after extracting multiple biometric modalities from a single data source, i.e., facial video clips. Utilizing different modalities, i.e., left ear, left profile face, frontal face, right profile face, and right ear, present in the facial video clips, we train supervised denoising auto-encoders to automatically extract robust and non-redundant features. The automatically learned features are then used to train modality specific sparse classifiers to perform the multimodal recognition. Moreover, the proposed technique has proven robust when some of the above modalities were missing during the testing. The proposed system has three main components that are responsible for detection, which consists of modality specific detectors to automatically detect images of different modalities present in facial video clips; feature selection, which uses supervised denoising sparse auto-encoders network to capture discriminative representations that are robust to the illumination and pose variations; and classification, which consists of a set of modality specific sparse representation classifiers for unimodal recognition, followed by score level fusion of the recognition results of the available modalities. Experiments conducted on the constrained facial video dataset (WVU) and the unconstrained facial video dataset (HONDA/UCSD), resulted in a 99.17% and 97.14% Rank-1 recognition rates, respectively. The multimodal recognition accuracy demonstrates the superiority and robustness of the proposed approach irrespective of the illumination, non-planar movement, and pose variations present in the video clips even in the situation of missing modalities.

프라이버시 침해 방지를 위한 얼굴 정보 변환 메커니즘 (Face Information Conversion Mechanism to Prevent Privacy Infringement)

  • 김진수;김상춘;박남제
    • 한국정보기술학회논문지
    • /
    • 제17권6호
    • /
    • pp.115-122
    • /
    • 2019
  • CCTV(Closed-circuit Television)는 사고 예방 및 시설 안전을 위해 매년 설치대수가 증가함에 따라 1인당 CCTV에 노출되는 횟수가 증대되고 있으며, 노출되는 대상의 프라이버시 보호를 위해 지능형 영상감시 시스템 기술이 각광받고 있다. 지능형 영상감시 시스템은 촬영된 영상 데이터에 대한 단순한 식별에서 피사체의 행동 유형과 현장 상황 판단 등을 수행하거나, 촬영된 피사체의 정보가 노출될 수 있는 정보를 외부로 유출되지 않도록 프라이버시 보호를 위한 처리 과정을 진행한다. 제안된 기술은 영상감시 시스템에 적용되어 영상감시 시스템으로부터 촬영된 원본 영상 정보를 유사 영상 정보로 변환함으로서 외부에 원본 영상 정보가 유출되지 않도록 하는 기술이다. 본문에서는 미리 설정된 유사도에 근접하는 가상의 얼굴 이미지를 삽입하는 영상 변환메커니즘을 제안한다.

전국 한방의료기관 한의사 대상 한의 변증활용 현황 조사 (A National Survey on Utilization of Pattern Identification among Korean Medicine Doctors)

  • 여민경;박기현;장은수;이영섭
    • 대한예방한의학회지
    • /
    • 제19권3호
    • /
    • pp.45-55
    • /
    • 2015
  • Objective : The aim of this study was to survey the present utilization of pattern identification(PI) in the clinical field among Korean Medicine Doctors. Method : This survey was conducted from Oct. 1 to Oct. 31, 2014 by face-to-face interview using structured questionnaire. The subjects were 400 korean medicine doctors who worked in hospitals or clinics. The questionnaire consisted of two sections - a demographics section(i.e., sex, age, clinical experience, place of work, residence and so on) and the main research section(i.e., the PI utilization rate, the utilization of PI systems, the utilization of PI in the clinical domain and in the treatment domain, the correlation between utilization PI and so on). Results : This study revealed that subjects used the PI to 66.7% of their patients on average. The utilization rate of PI in patients was significantly higher for female physicians than for male physicians, for physicians worked in clinics than for physicians who worked in hospitals and for physicians with more than 10 years clinical experience than for physicians with less than 10 years clinical experience. In the utilization of PI systems, Visceral PI was most as 92.0%. There were significant difference on the PI utilization rate according to utilization of Meridian and Collateral PI, Triple Energizer PI, Defense, Qi, Nutrient and Blood PI, Six-Meridian PI and Six Excesses PI. As a result of analysing the PI utilization rate by clinical domain and in the treatment domain, the Treatment and the Drug Treatment showed the highest values. Among utilized PI systems, the correlation coefficients between Defense, Qi, Nutrient and Blood PI and Triple Energizer PI showed the highest value, but the correlation coefficients between Sasang Constitutional Medicine with the each PI showed substantially lower value. Conclusion : The results of this study demonstrate that usage of PI was higher than usage of U code(in KCD) usage in clinical field. we suggested that additional studies on using PI and developing more appropriate standardized tool should be conducted to widen scope of PI's utilization.

전국 한방의료기관 한의사 대상 만성질환의 변증활용 현황 조사 (A National-wide Survey on Utilization of Pattern Identification for Chronic Diseases among Korean Medicine Doctors)

  • 여민경;박기현;이영섭
    • 대한예방한의학회지
    • /
    • 제21권2호
    • /
    • pp.23-34
    • /
    • 2017
  • Objectives : The aim of this study was to survey the present utilization of pattern identification(PI) by chronic disease and using PI system in the clinical field among Korean medicine doctors. Methods : This survey was conducted from Oct. 1 to Oct. 31, 2014 by face-to-face interview using structured questionnaire. The subjects were 400 Korean medicine doctors who worked in hospitals or clinics. The questionnaire consisted of whether use clinical experience by chronic disease and the utilization of PI, the utilization of PI systems, the PI utilization rate, and the correlation between utilization PI and so on. General elements consisted of sex, age, clinical experience, place of work, district and so on. Results and Conclusions : This study revealed that clinical experience by chronic disease used 24.0-90.8%. The most chronic diseases were chronic fatigue and chronic pain, and least disease was cancer. Experience of the utilization of PI among Korean medicine doctors who have clinical experience was 87.8-97.1%. Also, the most utilized disease was chronic gastritis & gastric ulcer, and least disease was dyslipidemia. In this case, the most common cases were using the one PI between the utilization of PI systems, and the utilization ratio of Visceral PI was the highest. Phi coefficient between the PI systems in chronic diseases divided into positive correlation and negative correlation. The correlation of dementia between Triple energizer PI and Six excesses PI was the highest at positive correlation, and the frequency of occurrence between Defense, Qi, Nutrient and Blood PI was the highest at positive correlation. The correlation of osteoarthropathy between Meridian and Collateral PI and Sasang Constitutional Medicine was the highest at negative correlation. Also the frequency of occurrence between Qi, Blood, Fluid and Humor PI and Sasang Constitutional Medicine was the highest at negative correlation. We hope that additional studies on systematic PI research of chronic disease which needed to be in korean oriental medicine and meet the needs of clinical consumer continue based on this study.

갱년기장애 및 폐경기 후 증후군 변증진단 도구의 변증분류 조정과 진단의 간의 진단일치도 연구 (A Study on Aadjustment of the Patterns, and the Correlation between the Diagnostic Tool for Climacteric and Postmenopausal Syndrome Pattern Identification (CaPSP) and Korean Medicine Doctors' Diagnosis)

  • 이인선;김종원;전수형;지규용;강창완
    • 대한한방부인과학회지
    • /
    • 제34권1호
    • /
    • pp.1-14
    • /
    • 2021
  • Objectives: We studied for the adjustment of the patterns of 'The Diagnostic Tool for Climacteric and Postmenopausal Syndrome Pattern Identification (hereinafter CaPSPI)' (studyI) and the correlation between CaPSPI and Korean medicine doctors' diagnosis which was carried out without knowing the results of CaPSPI (studyII). Methods: The studyI followed the previous study method in 2018 (2018-3). The studyII was conducted from June 1, 2019 to July 10, 2020 with ◯◯ University Korean Medicine Hospital IRB's approval (2019-4). Doctors' diagnosis was conducted face-to-face with the subjects. Doctors' diagnosis was carried out based on the Kupperman's questionnaire, 'Diagnosis System of Oriental Medicine (hereinafter DSOM)' and four examinations (四診) records. The diagnosis was marked with 0 for 'no', 1 for 'somewhat', 2 for 'yes' and 3 for 'very yes'. The correlation between CaPSPI and the mean of doctors diagnostic scores were investigated statistically. Results: The studyI showed that heart-heat (心火) pattern was added. The Factor loading coefficient for heart-heat was 0.551 to 0.789, and the Cronbach's coefficient was 0.896. The studyII showed that the diagnosis (Kappa statistic) of two doctors showed statistically significant concordance (all eight patterns), with correlation of them were 0.3 or higher. And the correlation between the CaPSPI score and the mean of doctors' diagnostic score showed a statistically significant correlation, with liver qi depression (肝鬱) being the highest at 0.552 and dual deficiency of the heart-spleen (心脾兩虛) being the lowest at 0.301. Conclusions: Since the diagnosis results of CaPSPI showed a significant correlation with the diagnosis of Korean traditional medicine experts, it was believed that the CaPSPI results can be trusted and used for clinical purposes.

갱년기장애 사용처방과 변증분류에 관한 전문가 합의 연구 (A Study on the Experts' Agreement on the Patterns Classification of the Prescription for the Use of Menopausal Disorder)

  • 전수형;박영희;김종원;지규용;이지연;이인선
    • 대한한방부인과학회지
    • /
    • 제35권2호
    • /
    • pp.70-86
    • /
    • 2022
  • Objectives: This study attempted to provide prescriptions that can be used according to the classification of menopausal disorders in order to increase the clinical utilization of CaPSPI (Diagnostic System for Climacteric and Postmenopausal Syndrome Pattern Identification). Methods: Delphi method survey was conducted for expert consensus on the Pattern Identification that could be used for each prescription by selecting frequent prescriptions based on the literature researched on the prescription of Korean medicine used for women with menopausal disorders. 11 experts participated in the questionnaire for posterity and 7 experts for sasang constitution. The survey was divided into the main prescription that can be mainly used for the patterns and the minor prescription that can be used although the suitability is somewhat low. Expert consensus was obtained through the second survey, and the pattern classification of prescriptions that were not confirmed here was confirmed through a face-to-face meeting. Results: There were 56 prescriptions mainly used for treatment for menopause, 45 for posterity and 11 for Sasang constitution. There were 69 identifications of 45 posterity prescriptions and 17 classifications of 11 Sasang constitutional prescriptions including major and minor patterns. By pattern identifications, the number of major/minor prescriptions was as follows. 肝鬱 14/5, 腎陰虛 8/2, 腎陽虛 4/1, 肝腎陰虛 2/7, 腎陰陽兩虛 3/1, 心腎不交 6/3, 心脾兩虛 15/2, 心火 7/6. Conclusions: In the future, by accumulating and analyzing the data on the diagnosis and treatment using CaPSPI, the clinical status of Korean medicine for menopausal disorders can be reviewed, and it can be used as the basis for the clinical practice guidelines for menopausal disorders. In addition, it is thought that analyzing the results of the diagnosis of pattern identifications and the prescription of Sasang constitution can be helpful in research on the posterity and Sasang constitution.

노인의 우울증상 식별력에 영향을 미치는 요인 (Factors affecting the ability of older adults to identify symptoms of depression)

  • 이선혜;고정은
    • 한국노년학
    • /
    • 제29권2호
    • /
    • pp.529-546
    • /
    • 2009
  • 노인우울은 유병율이 높을 뿐만 아니라 상당히 심각한 신체적, 심리사회적, 경제적 부담을 초래함에도 불구하고, 노화의 과정으로 또는 신체질환의 일부로 여겨지는 가운데 개입이 지연되고 있다. 건강문제를 해결하기 위한 시도가 일반적으로 당사자나 주변인들의 문제 인식에서 비롯되는 점에 비추어 볼 때, 우울증상에 대한 정확한 식별은 자기 자신은 물론 주위 노인들의 증상에 대한 조기발견과 개입을 촉진하는데 도움이 될 수 있다. 이에 본 연구에서는 대도시에 거주하는 노인 104명에 대한 면접 설문조사를 통해 우울증상에 대한 노인들의 식별력과 증상의 개념화 방식을 탐색하고 이에 영향을 미치는 요인들에 대한 분석을 실시했다. 연구결과, 우울증상을 식별할 수 있었던 노인은 전체 응답자의 14%에 불과했으며, 그 밖에는 외로움, 정서적 문제, 경제적 문제 등으로 인식했다. 75세 이상의 나이, 주된 정신건강정보원이 대중매체인 경우 식별력이 감소하는 것으로, 정신건강 지식이나 정보의 주된 원천이 정신건강 관련교육인 경우 식별력 증진에 도움이 되는 것으로 나타났다. 이와 같은 결과에 기초할 때, 노인의 우울증상 식별력을 증진시키기 위한 전략으로 정보지 비치나 일회성 행사위주의 정보전달보다 면대면의 구체적 정보제공과 상담 방식이 효과적일 것으로 보이며, 대중매체의 정신건강 및 질환 콘텐츠에 대한 면밀한 검토에 초점을 둔 후속 연구도 필요할 것이다.