• 정보보호학회논문지
• /
• 제19권3호
• /
• pp.11-25
• /
• 2009

지문샘플의 품질은 지문인식 시스템에서 매칭 성능에 영향을 주는 중요한 요소이다. 지문인식 시스템에서 낮은 품질의 지문영상을 제거하면 인식 에러율이 현저하게 줄어드는 것을 알 수 있다. 본 논문에서는 minutiae 기반의 지문인식 알고리즘을 이용하여 단일 샘플 품질 측정방법에 대한 유효성을 평가하였다. 우선, minutiae 기반의 지문인식 알고리즘의 매칭성능에 영향을 주는 여러 가지 요소에 대하여 검증하고, 다음, 현재 흔히 사용하는 측정방법에 대해 연구하고, 그 중 효과적인 품질측정방법들을 선택하여 NIST, QualityCheck와 Verifinger 5.0을 이용하여 비교 분석 하였다. FVC 데이터베이스로 실험한 결과 단일 측정방법도 매칭성능을 효과적으로 향상함을 알 수 있었다.

• 정보보호학회논문지
• /
• 제18권5호
• /
• pp.59-69
• /
• 2008

본 연구의 목적은 지문인식 평가용 데이터베이스의 난이도(Level of Difficulty)를 지문영상 품질 평가를 통하여 정량화 하는 것이다. 본 논문에서는 지문 데이터베이스의 난이도 정량화 방법으로 품질 분포도 분석 방법과 품질 차이 분석 방법을 제안한다. 품질 분포도 분석 방법은 지문 영상의 품진 등급별 빈도수를 기반으로 데이터베이스 전체의 난이도를 표현한 방법이고, 품질 차이 분석 방법은 데이터베이스를 구성하는 동일 지문의 영상들 간의 각 품질 차이별 빈도수를 동시발생 매트릭스(Co-occurrence Matrices)를 이용하여 난이도로 나타낸 방법이다. 두 방법론에 의한 실험 결과 지문영상 품질 기반의 데이터베이스의 난이도는 인식 성능과 상관관계를 가지며, 품질 분포도 기반의 난이도보다 품질 차이 기반의 난이도가 더 높은 상관관계를 보임을 확인할 수 있었다. 특히 MPQ(Matching Pairwise Qualities) 동시발생 매트릭스의 OQ(Opposite Qualities) Block 기반의 난이도 정량화 방법이 인식 성능과 가장 높은 상관관계를 나타내는 것을 알 수 있었다. 본 연구를 통해 지문영상 품질 기반의 지문 데이터베이스의 난이도를 정량화할 수 있었고, 난이도와 인식성능이 높은 상관관계를 가짐을 알 수 있다.

• 정보보호학회논문지
• /
• 제17권2호
• /
• pp.29-40
• /
• 2007

입력이 되는 지문 영상의 품질은 지문 인식 시스템의 전체 인식 성능에 다양하게 영향을 미친다. 지문 영상의 품질은 추출, 정합 등의 알고리즘 내에서도 중요한 정보로 활용되는데, 특히 센서 입력부에서의 품질 측정 및 평가는 불필요한 처리 과정을 줄여 시스템의 전반적인 안정성을 유도하는데 사용되는 만큼 매우 중요한 과정이 된다. 본 논문에서는 기존의 많은 연구 결과들과는 달리 센서, 인식 알고리즘, 손가락의 특성에 독립적인 지문 영상 품질을 정의하였다. 또한 이의 측정을 위하여 지문 영상의 지역적 특성들만을 이용한 새로운 품질 측정 방법도 제안하였다. 제안하는 품질 측정 방법은 블록별 품질 특성 분류 결과를 이용하여 전체 지문 영상의 품질을 수치화하는 것으로, 본 논문에서는 FVC에서 사용된 몇 가지 서브 데이터베이스들을 이용하여 제안하는 방법의 유용함을 증명하였다. 실험 결과는 NIST의 NFIQ, AWARE사의 QualityCheck 품질 측정 알고리즘들과 비교되었는데, 이들 알고리즘들에 비해 제안하는 측정 방법이 인간의 시각적 분류 기준과 유사한 결과를 보였다.

• Journal of Information Processing Systems
• /
• 제12권1호
• /
• pp.83-99
• /
• 2016

The latest research on the image-based fingerprint matching approaches indicates that they are less complex than the minutiae-based approaches when it comes to dealing with low quality images. Most of the approaches in the literature are not robust to fingerprint rotation and translation. In this paper, we develop a robust fingerprint matching system by extracting the circular region of interest (ROI) of a radius of 50 pixels centered at the core point. Maximizing their orientation correlation aligns two fingerprints that are to be matched. The modified Euclidean distance computed between the extracted orientation features of the sample and query images is used for matching. Extensive experiments were conducted over four benchmark fingerprint datasets of FVC2002 and two other proprietary databases of RFVC 2002 and the AITDB. The experimental results show the superiority of our proposed method over the well-known image-based approaches in the literature.

• 대한의생명과학회지
• /
• 제16권3호
• /
• pp.197-200
• /
• 2010

All citizens over 17 year old living in Korea have to be fingerprinted to obtain a certificate of resident registration. For this reason, human identification through fingerprints has been used actively in crime scene investigation. The fingerprint is so unique that it is one of the most certain ways to identify oneself and it can differentiate between genetically identical twins. Fingerprints gained in crime scene indicate a direction of criminal investigation in conjecturing a suspect. Fingerprints help a reunion of family got scattered for a long time and make it possible to get a personal identification for missing person who met with natural calamity. We developed a new fingerprinting method using safranine O, so as to develop fingerprints on the adhesive tapes and non-porous papers in various physical environments. Results were compared to the preexisting fingerprinting method, the minutiae numbers of fingerprints were greatly increased in our newly developed safranine O fingerprinting method. This newly developed safranine O method showed a quantity and quality comparable to the preexisting fingerprinting method routinely used in these days. In our hands, the safranine O fingerprinting method is another easy and obvious choice when the forensic case sample is available for fingerprints on the adhesive tapes and non-porous papers.

• 한국약용작물학회지
• /
• 제14권2호
• /
• pp.117-123
• /
• 2006

The purpose is to study the identification method of Huangjinju powder for injection and the medicinal materials by the fingerprint off-ray Diffraction Fourier (XRDF). We used the same method on both the studying of Huangjinju and the medicinal materials. Then we selected a few components alignment to compare. We analyzed the data by setting up the deviation $d({\AA})$ as ${\pm}0.05$ to calculate the rate of special mark on the sample (Px) and on the patent (P). The special XRDF of Huangjinju$[d({\AA})/(I/I_0)]$ have 5 peaks that have not expressed in medicinal materials. Therefore Px is 22.73%. Flos Trollii Chinensis has 3 special marks and Px is 17.65%. Flos Chrysanthemi Indici has 1 special mark and Px is 3.57%. Its coincided interplanar spacing with the patent is $2.907{\AA}$. Flos Lonicerae Japonicae has 6 special marks and Px is 23.08%. Its special mark in the patent are 4.95/14 and 4.50/15, respectively. The P is 9.09%. Its coincided interplanar spacing with the patent is $2.910{\AA}\;and\;3.05{\AA}$, respectively. The number of special XRDF mark peaks of baicalin is 9 and Px is 18.37%. Its coincided interplanar spacing with the patent is $2.910{\AA}$. It has visible mark and specificity adopting XRDF fingerprint to identify Huangjinju and medicinal materials. Establishing the quality standard is a synthetic index that depends both on special marks in the medicinal materials of the patent and on the coincidence peak data.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2000년도 International Symposium on Bioinformatics
• /
• pp.21-22
• /
• 2000

Expressed sequence tags (EFTs) are the partial segments of cDNA produced from 5 or 3 single-pass sequencing of cDNA clones, error-prone and generated in highly redundant sets. Advancement and expansion of Genomics made biologists to generate huge amount of ESTs from variety of organisms-human, microorganisms as well as plants, and the cumulated number of ESTs is over 5.3 million, As the EST data being accumulate more rapidly, it becomes bigger that the needs of the EST analysis tools for extraction of biological meaning from EST data. Among the several needs of EST analyses, the extraction of protein sequence or functional motifs from ESTs are important for the identification of their function in vivo. To accomplish that purpose the precise and accurate identification of the region where the coding sequences (CDSs) is a crucial problem to solve primarily, and it will be helpful to extract and detect of genuine CD5s and protein motifs from EST collections. Although several public tools are available for EST analysis, there is not any one to accomplish the object. Furthermore, they are not targeted to the plant ESTs but human or microorganism. Thus, to correspond the urgent needs of collaborators deals with plant ESTs and to establish the analysis system to be used as general-purpose public software we constructed the pipelined-EST analysis system by integration of public software components. The software we used are as follows - Phred/Cross-match for the quality control and vector screening, NCBI Blast for the similarity searching, ICATools for the EST clustering, Phrap for EST contig assembly, and BLOCKS/Prosite for protein motif searching. The sample data set used for the construction and verification of this system was 1,386 ESTs from human intrathymic T-cells that verified using UniGene and Nr database of NCBI. The approach for the extraction of CDSs from sample data set was carried out by comparison between sample data and protein sequences/motif database, determining matched protein sequences/motifs that agree with our defined parameters, and extracting the regions that shows similarities. In recent future, in addition to these components, it is supposed to be also integrated into our system and served that the software for the peptide mass spectrometry fingerprint analysis, one of the proteomics fields. This pipelined-EST analysis system will extend our knowledge on the plant ESTs and proteins by identification of unknown-genes.