• 제목/요약/키워드: Identity Information Recognition

Search Result 109, Processing Time 0.024 seconds

Development of Semi-Supervised Deep Domain Adaptation Based Face Recognition Using Only a Single Training Sample (단일 훈련 샘플만을 활용하는 준-지도학습 심층 도메인 적응 기반 얼굴인식 기술 개발)

  • Kim, Kyeong Tae;Choi, Jae Young
    • Journal of Korea Multimedia Society
    • /
    • v.25 no.10
    • /
    • pp.1375-1385
    • /
    • 2022
  • In this paper, we propose a semi-supervised domain adaptation solution to deal with practical face recognition (FR) scenarios where a single face image for each target identity (to be recognized) is only available in the training phase. Main goal of the proposed method is to reduce the discrepancy between the target and the source domain face images, which ultimately improves FR performances. The proposed method is based on the Domain Adatation network (DAN) using an MMD loss function to reduce the discrepancy between domains. In order to train more effectively, we develop a novel loss function learning strategy in which MMD loss and cross-entropy loss functions are adopted by using different weights according to the progress of each epoch during the learning. The proposed weight adoptation focuses on the training of the source domain in the initial learning phase to learn facial feature information such as eyes, nose, and mouth. After the initial learning is completed, the resulting feature information is used to training a deep network using the target domain images. To evaluate the effectiveness of the proposed method, FR performances were evaluated with pretrained model trained only with CASIA-webface (source images) and fine-tuned model trained only with FERET's gallery (target images) under the same FR scenarios. The experimental results showed that the proposed semi-supervised domain adaptation can be improved by 24.78% compared to the pre-trained model and 28.42% compared to the fine-tuned model. In addition, the proposed method outperformed other state-of-the-arts domain adaptation approaches by 9.41%.

Landscape Design of Busan Yeonje Post Office (부산연제우체국청사 조경설계)

  • 김정수;최연철
    • Journal of the Korean Institute of Landscape Architecture
    • /
    • v.30 no.6
    • /
    • pp.109-118
    • /
    • 2003
  • This design is for the landscape of Busan Yeonje Post Office. The design requirements were: the preparation of an open space, the expression of regional identity, total design concept, and application of environmentally-friendly materials. The Busan Yeonje Post Office is not a just post office building but also a place for people including a communications center, rental offices, and a post office, facilities which require publicity, functionality, and symbolism. Therefore, this design reflected the design requirements, and considered the following to include the theme of "Pleasant messages and a pleasant post office with vitality and comfort. First, the open space was to be increased in the adjacent public vacant lot and street park by pulling back the construction line. In this way, the publicity, accessibility, and spatiality of the prepared open space were strengthened. Second, a sense of place for the landscape was expressed through the image of a wave representing Busan, a coastal town. Recognition of the space was strengthened, while regional identity and the concept of transmission of information were expressed. In addition, Citrus junos Sieb., representing 'Happy news' was Planted as a symbolic tree to denote a meaningful resting Place. Third, with respect to the design of reception facilities, considering the general purposes of the construction, post office stamps, envelopes, and postal delivery persons were portrayed on the walls, decorative walls, and environmental formations respectively The flooring was decorated with stamps and waves in keeping with of the total design concept. Fourth, to select environmentally-friendly materials, soil block and eco-grass block were introduced as major paving materials, and folk kinds of trees were included on site. According to the aforesaid plans, the researcher attempted to convert the public agency and office space characterized by feelings of dullness and rigidity into a soft and vivid outdoor space. It is expected that the location will be an open place in the downtown area and be recognized as a 'shared' space for citizens of Yeonje-gu.

Palmprint Identification Algorithm using Hu Invariant Moments (Hu 불변 모멘트를 이용한 장문인식 알고리즘)

  • SHIN Kwang Gyu;RHEE Kang Hyeon
    • Journal of the Institute of Electronics Engineers of Korea CI
    • /
    • v.42 no.2 s.302
    • /
    • pp.31-38
    • /
    • 2005
  • Recently, Biometrics-based personal identification is regarded as an effective method of person's identity with recognition automation and high performance. In this paper, the palmprint recognition method based on Hu invariant moment is proposed. And the low-resolution(750dpi) palmprint image$(5.5Cm\times5.5Cm)$ is used for the small scale database of the effectual palmprint recognition system. The proposed system is consists of two parts: firstly, the palmprint fixed equipment for the acquisition of the correctly palmprint image and secondly, the algorithm of the efficient processing for the palmprint recognition. And the palmprint identification step is limited 3 times. As a results, when the coefficient is 0.001 then FAR and GAR are $0.038\%$ and $98.1\%$ each other. The authors confirmed that FAR is improved $0.002\%$ and GAR is $0.1\%$ each other compared with [3].

Local Appearance-based Face Recognition Using SVM and PCA (SVM과 PCA를 이용한 국부 외형 기반 얼굴 인식 방법)

  • Park, Seung-Hwan;Kwak, No-Jun
    • Journal of the Institute of Electronics Engineers of Korea SP
    • /
    • v.47 no.3
    • /
    • pp.54-60
    • /
    • 2010
  • The local appearance-based method is one of the face recognition methods that divides face image into small areas and extracts features from each area of face image using statistical analysis. It collects classification results of each area and decides identity of a face image using a voting scheme by integrating classification results of each area of a face image. The conventional local appearance-based method divides face images into small pieces and uses all the pieces in recognition process. In this paper, we propose a local appearance-based method that makes use of only the relatively important facial components. The proposed method detects the facial components such as eyes, nose and mouth that differs much from person to person. In doing so, the proposed method detects exact locations of facial components using support vector machines (SVM). Based on the detected facial components, a number of small images that contain the facial parts are constructed. Then it extracts features from each facial component image using principal components analysis (PCA). We compared the performance of the proposed method to those of the conventional methods. The results show that the proposed method outperforms the conventional local appearance-based method while preserving the advantages of the conventional local appearance-based method.

A Medical Staff Identification System by Using of Beacon, Iris Recognition and Blockchain (비콘과 홍채인식, 블록체인 기반의 의료진 신분확인 시스템 제안)

  • Lim, Se Jin;Kwon, Hyeok Dong;Seo, Hwa Jeong
    • KIPS Transactions on Computer and Communication Systems
    • /
    • v.10 no.1
    • /
    • pp.1-6
    • /
    • 2021
  • Recently, incidents such as proxy surgery (unlicensed medical practice) have been reported in the media that threaten the safety of patients. Alternatives such as the introduction of operating room surveillance camera devices to prevent proxy surgery are emerging, but there are practical difficulties in implementing them due to strong opposition from the medical community. However, the social credibility of doctors is falling as incidents such as proxy surgery occur frequently. In this paper, we propose a medical staff identification system combining Beacon and iris recognition. The system adds reliability by operating on the blockchain network. The system performs primary identification by performing user authentication through iris recognition and proves that the medical staff is in the operating room through beacons. It also ensures patient trust in the surgeon by receiving beacon signals in the background and performing iris authentication at random intervals to prevent medical staff from leaving the operating room after only performing initial certification.

The Security Problem Analysis for Reversibility of Transformed Biometric Information Data on Eigenvector-based face Authentication (특성 벡터를 이용한 얼굴 인증 시스템에서 변환된 생체 정보 데이터의 가역성에 대한 보안 문제 분석)

  • Kim, Koon-Soon;Kang, Jeon-Il;Nyang, Dae-Hun;Lee, Kyung-Hee
    • Journal of the Korea Institute of Information Security & Cryptology
    • /
    • v.18 no.3
    • /
    • pp.51-59
    • /
    • 2008
  • The biometrics has been researched as a means for authenticating user's identity. Among the biometrics schemes for face recognition, the eigenvector-based schemes, which use eigenvector made from training data for transforming test data to abstracted data, are widely adopted. From those schemes, however, it is hard to expect cancelable feature, which is a general concept for security in the biometrics. In this paper, we point out the security problem that is the recovery of valuable face information from the abstracted face data and consider a possible attack scenario by showing our experiment results.

Development of Electronic Identification System of Individual Dairy Cow for Stockvreeding Automatization I. Transmitting and Receiving Circuit Design and Manufacture (젖소의 사양관리 자동화를 위한 전자개체인식장치 개발 I.송, 수신부 회로설계 및 제작)

  • 한병성;정길도;최명호;김용준;김명순;강복원
    • Journal of Veterinary Clinics
    • /
    • v.13 no.2
    • /
    • pp.171-176
    • /
    • 1996
  • In this study, dldctronic identification system of individual dairy cow was developed for autocatization of stoxkvreeding management. To automize the breeding management, it is necessary to obtain and analyze the individual information distinguished from others perferentially. Electronic identification system can distinguish individual livestock from others with electromagnetic wave signal recognition system. Electoronic identification system consists of transmitter transmitting the oscillated signal and receiver set. The transmitted signal from transmitter clung to individual livestock is received from the receiving antenna and the signal in different according to the established value of the register. By distinct signal recieved from the reciever, wi can distinguish the identity of a livestock from others clearly. This system can manage $2^{12}$ individuals with a reciever theoretically. However in order to reduce the errors by analogous signal, this system uses only triple number and can manage 1365 individuals with a reciever practically. This system can be connevtted to Max 232 and microcomputer for the breeding management efficiently.

  • PDF

The Biometric based Mobile ID and Its Application to Electronic Voting

  • Yun, Sung-Hyun;Lim, Heui-Seok
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.7 no.1
    • /
    • pp.166-183
    • /
    • 2013
  • It requires a lot of costs and manpower to manage an election. The electronic voting scheme can make the election system economic and trustful. The widespread use of smart phones causes mobile voting to be a major issue. The smart phone can be used as a mobile voting platform since it can carry out many services in addition to basic telephone service. To make mobile voting practical and trustful, we analyzed two subjects of study. Firstly, the way to make a biometric based mobile ID, which has legal binding forces. In mobile voting, user identification is accomplished on line since the voter should be able to vote wherever they go. The digital ID conducts a similar role to the need for a resident card. The user's identity is bound to the resident card legally. To bind the user's identity to the smart phone, we use USIM. Biometric recognition is also needed to authenticate the user, since the user cannot prove him or her on line face-to-face. The proposed mobile ID can be reissued by means of introducing a random secret value. Secondly, the mobile voting scheme is proposed where candidates can accept election results without doubt. The goal of an election is to select a leader among two or more candidates. Existing electronic voting schemes mainly focus on the study of ballot verification accomplished by voters. These approaches are not safe against collusion attacks where candidates and the election administration center are able to collude to fabricate election results. Therefore, a new type of voting and counting method is needed where candidates can directly take part in voting and counting stages. The biometric based multi-signature scheme is used to make the undeniable multi-signed ballot. The ballot cannot be verified without the help of all candidates. If candidates accept election results without a doubt, the fairness of the election is satisfied.

User Authentication Method using EEG Signal in FIDO System (FIDO 시스템에서 EEG 신호를 이용한 사용자 인증 방법)

  • Kim, Yong-Ki;Chae, Cheol-Joo;Cho, Han-Jin
    • Journal of the Korea Convergence Society
    • /
    • v.9 no.1
    • /
    • pp.465-471
    • /
    • 2018
  • Recently, biometric technology has begun to be used as a fusion of IT technology and financial system. Using this biometric technology, FIDO(Fast Identity Online) technology, Samsung and Apple started Samsung Pay and Apple Pay service. FIDO authentication technology replaces existing authentication methods such as passwords. Among the biometric technologies, fingerprint recognition technology is attracting attention because it can minimize the device and user rejection at a relatively low price. However, fingerprint information has a limited number of users and it can not be reused if fingerprint information is leaked by an external attacker. Therefore, in this paper, we propose a method to authenticate a user using EEG signal which is one of biometrics technologies. W propose a method to use EEG signal measurement value in FIDO system by using convenience channel by using short channel EEG device. And propose a method to utilize EEG signal when the user recognizes a specific entity by measuring the EEG signal before and after recognizing a specific entity.

A Study on the Measure for Improving the Website of ScienceON

  • Younghee Noh
    • International Journal of Knowledge Content Development & Technology
    • /
    • v.14 no.3
    • /
    • pp.93-113
    • /
    • 2024
  • This study was conducted for the purpose of deriving the existing ScienceON elements and reorganizing them to increase user convenience and utilization. Towards this end, implications were derived by analyzing the current status of the ScienceON's website, case studies and analysis of major similar institution websites, and user satisfaction surveys. Based on the derived contents, key terms were analyzed and redefined, the menu system was reorganized, and the menu names were redefined, and the menu system was reorganized to match the direction and identity of the redefined ScienceON. The study results based on such are as follows. First, the ScienceON's brand selection criteria and direction were based on service sustainability, service inclusiveness, recognition of portal characteristics, user friendliness and intuitiveness. Considering the meaning of branding, branding strategy, and selection criteria for ScienceON branding, ScienceON, Ontong ScienceON, ScienceONPlatform, science information portal, ScienceON, Mecca of science and technology information, and 5 brand names such as ScienceON were proposed. Second, to improve accessibility, usability, and satisfaction, it would be necessary to redefine current terms. Towards this end, it would be necessary to change the terminology of knowledge infrastructure search and to unify and organize the terminology. Third, in the current menu, the services provided as analysis services and curation services are organized in the beta service menu, and data such as papers, patents, reports, and trends are also serviced only through integrated search, and hence, each data type and service type classification must be improved.