• 정보보호학회논문지
• /
• 제26권5호
• /
• pp.1259-1267
• /
• 2016

최근 지문 인식을 통한 사용자 인증 기술이 상용화 되면서 위조 지문 이미지 판별이 더욱 중요해졌다. 본 논문에서는 CNN 특징을 이용한 위조 지문 이미지 판별 방법을 제안하였으며, CNN 모델이 실제 지문의 어느 부분에 반응하여 위조지문을 분류하는지 시각화 방법을 통해 분석하였다. 제안하는 방법은 지문영역과 배경영역을 분리하는 전처리 작업 후 CNN 모델을 이용하여 지문의 위조여부를 분류한다. 지문을 단순히 생체지문과 위조지문으로 분류하는 것이 아니라 위조지문을 구성하는 물질별로 분류하여 생체지문과 위조지문들에 대한 특징분석을 제공한다. 실험에 사용한 데이터베이스로는 생체 지문 이미지 6500여 장과 위조 지문 이미지 6000여 장으로 구성되어 있는 LivDet2013을 사용하였으며 위조여부에 대한 ACE 값으로 3.1%, 구성 물질 분류 정확도는 평균 79.58%를 보여 높은 수준의 분류성능을 갖고 있음을 확인하였다.

• Clinical and Experimental Reproductive Medicine
• /
• 제51권1호
• /
• pp.69-74
• /
• 2024

Objective: The purpose of this study was to identify factors associated with twin pregnancy following day 3 double embryo transfer (DET). Methods: This retrospective cohort study incorporated data from 16,972 day 3 DET cycles. The participants were women aged between 18 and 45 years who underwent in vitro fertilization with intracytoplasmic sperm injection (IVF/ICSI) at My Duc Assisted Reproduction Technique Unit (IVFMD), My Duc Hospital, located in Ho Chi Minh City, Vietnam. Results: Of the 16,972 day 3 DET cycles investigated, 8,812 (51.9%) resulted in pregnancy. Of these, 6,108 cycles led to clinical pregnancy, with 1,543 (25.3% of clinical pregnancies) being twin pregnancies. Factors associated with twin pregnancy included age under 35 years (odds ratio [OR], 1.5; 95% confidence interval [CI], 1.32 to 1.71; p<0.001) and cycles involving the transfer of at least one grade I embryo. Relative to the transfer of two grade III embryos, the risk of twin pregnancy was significantly elevated following the transfer of two grade I embryos (OR, 1.40; 95% CI, 1.16 to 1.69; p<0.001) or a combination of one grade I and one grade II embryo (OR, 1.27; 95% CI, 1.05 to 1.55; p=0.001). Conclusion: By analyzing a large number of IVF/ICSI cycles, we identified several predictors of twin pregnancy. These findings can assist medical professionals in tailoring treatment strategies for couples with infertility.

• 충청수학회지
• /
• 제9권1호
• /
• pp.27-60
• /
• 1996

We compute low order terms of automorphisms of a quadric CR manifold defined by $v^a$ =< $A^az$, z > where there is a real vector ${\kappa}{\in}R^m$ such that $det({\kappa}{\cdot}A){\neq}0$.

• 대한수학회논문집
• /
• 제22권3호
• /
• pp.419-428
• /
• 2007

In this paper, we will investigate the Hessian geometry of the homogeneous domain over the hypersurface given by a function F : $\mathbb{R}^n{\rightarrow}\mathbb{R}$ with ${\mid}{\det}\;DdF\mid=1$.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 학술대회지
• /
• pp.562-566
• /
• 2004

In this paper, we propose a method for constructing terrestrial/cable PSIP converting system, so-called a PSIP converter, which is converting terrestrial PSIP into cable PSIP for a data broadcasting service in the interoperable network of terrestrial and cable, and define an interface between the PSIP converter and the OOB SI generator by using PMCP messages compliant to ATSC T3/S1. The exiting PSIP converter just converts terrestrial PSIP into cable PSIP compliant to ATSC and OCAP standard and transmits by a MPEG-2 TS format. That is to say, it is not for the digital data broadcasting but for the digital broadcasting. Say in other word, the proposed PSIP converter is designed to convert and transmit PSIP including DET information which is necessary data event information for data broadcasting service. In addition, the PSIP converter can support various types of PSIP information to the OOB SI generator by using PMCP messages defined by a hierarchical structure as per each channel, audio/video event, data event and so on.

• International Journal of Computer Science & Network Security
• /
• 제21권11호
• /
• pp.59-66
• /
• 2021

Many incidents of Carbon Monoxide (CO) poisoning have occurred because of people being unaware of its presence. There are currently available systems on the market, but they are limited to measuring CO in a certain area and lack vital functions. Additionally, little to no evidence-based information on their quality was available. Thus, a mobile application for detecting CO gas leakage in a vehicle and critical features to assist victims was developed. A usability and functionality test were conducted to determine the product's quality utilizing nine usability and six functionality task scenarios (n=5). Then, a System Usability Scale test was performed to obtain system satisfaction, usability, and learnability (n=50). The usability and functionality test shows that all the tasks given for both tests were 100% successful. The overall score obtained for SUS was 71.4, which indicates good acceptance and usability. Around 20% of respondents claimed that they would need the support of a technical person to be able to use the application and that they needed to learn a lot of things before they could use the application, which indicates the overall high learnability of the application. The result provides empirical evidence that the CO gas leakage detection mobile application is successful and receives good usability, functionality, acceptability, learnability, and satisfaction assessments. DetGas could benefit automobile owners and other stakeholders by mitigating the risk and harm associated with gas leaking that exceeds the safe limit.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• 호남수학학술지
• /
• 제39권4호
• /
• pp.677-688
• /
• 2017

Let P be a finitely generated projective module over a commutative ring R with identity. If P has finite rank, then it will be shown that the map ${\varphi}:Aut_R(P){\rightarrow}U(R)$ defined by ${\varphi}({\alpha})={\det}({\alpha})$ is locally surjective and $Ker({\varphi})=SL_R(P)$.

• Journal of applied mathematics & informatics
• /
• 제33권5_6호
• /
• pp.627-633
• /
• 2015

In this paper, we generalize the concept of the energy of Seidel matrix S(G) which denoted by S^α(G) and obtain some results related to this matrix. Also, we obtain an upper and lower bound for S^α(G) related to all of graphs with |detS(G)| ≥ (n - 1); n ≥ 3.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2010년도 춘계학술대회 논문집
• /
• pp.1213-1214
• /
• 2010