• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.23-30
• /
• 2022

Glaucoma is a chronic neuropathy that affects the optic nerve which can lead to blindness. The detection and prediction of glaucoma become possible using deep neural networks. However, the detection performance relies on the availability of a large number of data. Therefore, we propose different frameworks, including a hybrid of a generative adversarial network and a convolutional neural network to automate and increase the performance of glaucoma detection. The proposed frameworks are evaluated using five public glaucoma datasets. The framework which uses a Deconvolutional Generative Adversarial Network (DCGAN) and a DenseNet pre-trained model achieves 99.6%, 99.08%, 99.4%, 98.69%, and 92.95% of classification accuracy on RIMONE, Drishti-GS, ACRIMA, ORIGA-light, and HRF datasets respectively. Based on the experimental results and evaluation, the proposed framework closely competes with the state-of-the-art methods using the five public glaucoma datasets without requiring any manually preprocessing step.

• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.614-630
• /
• 2023

Dead trees significantly impact forest production and the ecological environment and pose constraints to the sustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmanned aerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly on inefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method was developed in data pre-processing to display tree features in the shadows. For the model structure, the original CSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standard convolutional blocks in the original extraction network were replaced by depthwise separable convolution blocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make the network more robust for low-precision computations. The K-means++ clustering method was also integrated to generate anchor boxes that are more suitable for the dataset. The experimental results show that the improved algorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposed approach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.

• 정보보호학회논문지
• /
• 제22권1호
• /
• pp.43-56
• /
• 2012

무선 센서 네트워크는 재사용이 불가능한 배터리와 제한된 처리능력, 저장 공간을 갖는 다량의 소형 노드로 이루어진다. 이 네트워크에서 노드들은 광범위한 영역에 배치되게 되며 이 노드들은 또한 무선 링크를 통해 노드들 사이에 단거리 통신을 수행한다. 네트워크의 에너지 효율을 위해 동적 클러스터링 기법이 네트워크 수명, 확장, 부하 분산에 효과적인 수단이다. 이 기법은 다수의 노드에 의해 수집되는 데이터가 클러스터 헤드 노드에 의해 집성되어 재전송되는 특징이 있어 해당 노드가 공격자에 노출될 경우 네트워크의 안전을 보장할 수 없게 된다. 그러므로 이러한 클러스터링 기법의 안전한 통신을 위해 노드들 사이에 전송되는 메시지의 암호화와 클러스터 헤드 노드의 보안 유지가 중요하다. 특히, 에너지 효율을 목적으로 설계된 클러스터 기반 프로토콜에서 충분한 데이터 안정성을 보장하기 위해서는 클러스터 구조에 적합한 키 관리 및 인증 기법이 필요하다. 이에 본 논문에서는 계층 클러스터 구조를 갖는 센서 네트워크에 적합한 키 관리 기법을 제안한다. 제안하는 기법은 다항식 키 풀 기반 기법에 기초하며 키 인증 절차를 통해 안정된 네트워크를 유지한다.

• 대한조선학회논문집
• /
• 제55권1호
• /
• pp.66-74
• /
• 2018

In order to evaluate the precision degree of the blocks on the dock, the shipyards recently started to use the point cloud approaches using the 3D scanners. However, they hesitate to use it due to the limited time, cost, and elaborative effects for the post-works. Although it is somewhat traditional instead, they have still used the electro-optical wave devices which have a characteristic of having less dense point set (usually 1 point per meter) around the contact section of two blocks. This paper tried to expand the usage of point sets. Our approach can estimate the rework time to weld between the Pre-Erected(PE) Block and Erected(ER) block as well as the precision of block construction. In detail, two algorithms were applied to increase the efficiency of estimation process. The first one is K-mean clustering algorithm which is used to separate only the related contact point set from others not related with welding sections. The second one is the Concave hull algorithm which also separates the inner point of the contact section used for the delayed outfitting and stiffeners section, and constructs the concave outline of contact section as the primary objects to estimate the rework time of welding. The main purpose of this paper is that the rework cost for welding is able to be obtained easily and precisely with the defective point set. The point set on the blocks' outline are challenging to get the approximated mathematical curves, owing to the lots of orthogonal parts and lack of number of point. To solve this problems we compared the Radial based function-Multi-Layer(RBF-ML) and Akima interpolation method. Collecting the proposed methods, the paper suggested the noble point matching method for minimizing the rework time of block-welding on the dock, differently the previous approach which had paid the attention of only the degree of accuracy.

• 방송공학회논문지
• /
• 제22권1호
• /
• pp.28-41
• /
• 2017

본 논문에서는 사람의 인지특성을 기반으로 대조 민감도에 의해 나타나는 특성을 모델링 한 JND (Just Noticeable Difference) 모델을 비디오 코딩에 적용하여 압축률을 높이는 방법을 제안한다. 제안하는 방법은 JND 모델에 따른 임계치를 기준으로 양자화 단계에서 비가시 신호를 제한하여 주관적 화질을 유지하면서 비트율을 낮추는 방법으로, 변환을 통해 주파수 도메인으로 변환된 잔차 신호들을 양자화 단계에서 입력으로 받아 신호제한 및 양자화를 수행한다. 양자화 단계에서 주파수 도메인의 신호가 JND 관점에서 유사하게 인지되는 기준 임계치를 구해 잔차 신호에서 비가시 신호를 제한하고 양자화를 수행한 후, 최적의 율-인지왜곡 비용을 갖는 양자화 계수를 선택함으로써 비트율을 절감시킨다. 제안하는 알고리즘의 성능 검증은 최신 비디오 압축 표준인 HEVC (High Efficiency Video Coding)의 참조 소프트웨어인 HM16.0에 적용했으며, CTC (Common Test Condition)의 Random Access 모드에서 HM 16.0을 통해 압축된 영상 대비 평균 4.11%, BQTerrace 영상의 양자화 파라미터 22에서 최대 17.22%의 비트율 절감을 보였으며, Low Delay 모드에서 평균 7.16%, 최대 22.55%, All intra 모드에서 평균 13.41%, 최대 21.64%의 비트율 절감을 보였다. 5명의 평가자들의 주관적 화질 측정으로 평균 DMOS (Difference Mean Opinion Score) 값은 최대 약 0.36 최소 0 정도의 분포를 보였다.

• Clinical and Experimental Reproductive Medicine
• /
• 제28권3호
• /
• pp.199-207
• /
• 2001

Objective: The present study was undertaken to examine the effects of magnesium ion in the culture medium on the development of mouse fertilized oocytes either before or after pronuclear formation, and to investigate whether the effect of magnesium ion is related with the redistributional change of mitochondria. Methods : Fertilized oocytes obtained from the oviducts of mice at 15 hr after hCG injection before pronuclear formation (pre-PN) or 21 hr after hCG injection after pronuclear formation (post-PN) were used. The embryos were cultured for 3 days with basic T6 medium-magnesium free and various concentrations of magnesium ion, 0.0, 0.5, 1.0, 2.0, 4.0 or 8.0 mM, respectively. After culture, the developmental stages of embryos and the number of nuclei were evaluated. To observe the effects of magnesium ion on the mitochondrial distribution, fertilized oocytes were collected at 21 hr after hCG injection and cultured for 6 hr with various concentration of magnesium ion. As a control, fertilized oocytes with pronuclei at 27 hr after hCG injection were used. Results: The concentration of magnesium ion to accelerate the in vitro development of mouse fertilized oocytes appeared to be at 2.0 mM for the pre-PN and the post-PN stage embryos. In the mitochondrial redistribution patterns, the embryos cultured in 2.0 mM concentration of magnesium ion showed the highest percentage (22.6%) of distinct perinuclear clustering pattern comparing to other experimental group. Conclusion: The effect of magnesium ion may be related to the cytoplasmic redistribution of mitochondria. This relationship seems to connect the developmental competence of preimplantation mouse embryos in vitro. These results can suggest that higher concentration of magnesium ion (2.0 mM) than those of conventional culture medium ($0.2{\sim}1.2\;mM$) is more suitable for in vitro culture of preimplantation mouse embryos.

• Journal of Preventive Medicine and Public Health
• /
• 제33권3호
• /
• pp.323-333
• /
• 2000

Objectives : Chronic HBsAg carriers are the principal source of infection for other susceptible people, and are themselves at high risk of developing serious liver diseases. In Korea, it has been estimated that 65-75% of the HBsAg positives remained as persistent carriers. Additionally, familial clustering of MBV infection has frequently been observed among carriers. Some would become progressive, chronic hepatitis patients, and others would not. The aim of this study was to evaluate the association between various factors, such as the duration of infection, type of virus, mutation of precore/core region in HBV, major histocompatibility class-I, and developing chronic liver diseases among familial HBV carriers. Methods : Chronic carrier status was identified by repeated serological tests for HBsAg at intervals of six months or more. A familial chronic carrier was defined when the disease was observed in a family member over two generations. Two families were recruited, among which a total of 20 chronic HBsAg carriers(11 carriers in No.1, and 9 in No.2 family) were identified. Data on the general characteristics and liver disease status were collected. Identification of the HBV-DNA was successful only for 13 subjects among the 20 carriers. Analysis of viral DNA in terms of subtype, pre-core and core region mutations was carried out. The type of major histocompatibility class-1 for the 13 subjects was also analysed. Results & Conclusions : Seven of 10 chronic HBV carriers of the 1st generation and one of 10 of the 2nd generation were clinical patients with chronic hepatitis, the others, three of the 1 st and nine of the 2nd generation, were asymptomatic carriers. This data indicates that the duration of HBV carriage is one of the major factors for disease severity. The subtype of HBsAg analysed using MBV-DNA identified in 13 carriers were adr, and the pattern of precore nonsense mutation in HBV-DNA was identical among family members, which meads that the same virus strains were transmitted between the family members. The association between the precore or core mutations in HBV-DNA and the disease severity was not observed. While it was suggested that a specific type of MHC class-I may be related to disease progression.

• 한국분말야금학회:학술대회논문집
• /
• 한국분말야금학회 2001년도 추계학술강연 및 발표대회
• /
• pp.7-7
• /
• 2001

The increasing interest in light weight materials coupled to the need for cost -effective processing have combined to create a significant opportunity for aluminum P/M. particularly in the automotive industry in order to reduce fuel emissions and improve fuel economy at affordable prices. Additional potential markets for Al PIM parts include hand tools. Where moving parts against gravity represents a challenge; and office machinery, where reciprocating forces are important. Aluminum PIM adds light weight, high compressibility. low sintering temperatures. easy machinability and good corrosion resistance to all advantages of conventional iron bm;ed P/rv1. Current commercial alloys are pre-mixed of either the AI-Si-Mg or AL-Cu-Mg-Si type and contain 1.5% ethylene bis-stearamide as an internal lubricant. The powder is compacted in closed dies at pressure of 200-500Mpa and sintered in nitrogen at temperatures between $580~630^{\circ}C$ in continuous muffle furnace. For some applications no further processing is required. although most applications require one or more secondary operations such as sizing and finishing. These sccondary operations improve the dimension. properties or appearance of the finished part. Aluminum is often considered difficult to sinter because of the presence of a stable surface oxide film. Removal of the oxide in iron and copper based is usually achieved through the use of reducing atmospheres. such as hydrogen or dissociated ammonia. In aluminum. this occurs in the solid st,lte through the partial reduction of the aluminum by magncsium to form spinel. This exposcs the underlying metal and facilitates sintering. It has recently been shown that < 0.2% Mg is all that is required. It is noteworthy that most aluminum pre-mixes contain at least 0.5% Mg. The sintering of aluminum alloys can be further enhanced by selective microalloying. Just 100ppm pf tin chnnges the liquid phase sintering kinetics of the 2xxx alloys to produce a tensile strength of 375Mpa. an increilse of nearly 20% over the unmodified alloy. The ductility is unnffected. A similar but different effect occurs by the addition of 100 ppm of Pb to 7xxx alloys. The lend changes the wetting characteristics of the sintering liquid which serves to increase the tensile strength to 440 Mpa. a 40% increase over unmodified aIloys. Current research is predominantly aimed at the development of metal matrix composites. which have a high specific modulus. good wear resistance and a tailorable coefficient of thermal expnnsion. By controlling particle clustering and by engineering the ceramic/matrix interface in order to enhance sintering. very attractive properties can be achicved in the ns-sintered state. I\t an ils-sintered density ilpproaching 99%. these new experimental alloys hnve a modulus of 130 Gpa and an ultimate tensile strength of 212 Mpa in the T4 temper. In contest. unreinforcecl aluminum has a modulus of just 70 Gpa.

• 한국컴퓨터그래픽스학회논문지
• /
• 제16권4호
• /
• pp.25-33
• /
• 2010

디지털 카메라가 널리 보급되면서 사람들은 언제, 어디서나 사진을 찍고 값싼 저장장치에 많은 수의 사진을 저장할 수 있게 되었다. 하지만 많은 수의 사진 중 원하는 사진을 효율적으로 탐색하는 것은 어려운 문제로 남아 있다. 본 논문은 위치 정보를 갖는 대규모 사진집합을 신속하고 직관적으로 탐색하는 새로운 방법을 제안한다. 전체 사진집합을 구조화하기 위해 지리적 거리가 가장 근접한 사진들을 묶어 군집화하고, 이러한 과정을 반복하여 최종적으로 모든 사진이 하나의 군집으로 병합되는 계층적 군집화를 수행한다. 또한 모든 군집의 컨벡스 헐과 넓이를 미리 계산하여 사진 탐색 시에는 미리 계산된 데이터와 현재 탐색 중인 지리 영역에 포함되는 군집들의 넓이를 비교해 적절한 넓이의 군집들을 선택적으로 시각화한다. 이 때 군집은 포함되는 모든 사진의 위치를 보여주는 대신 컨벡스 헐로 시각화하여 군집의 정확한 공간적 범위를 쉽게 파악할 수 있다. 사용자는 관심 군집을 클릭하여 해당 군집으로 신속하게 이동할수 있으며, 시스템은 관심 군집을 지도 영역에 정확히 채워 보일 수 있도록 자동적으로 지도 이동과 축척 조절을 하고 적절한 넓이의 하위 군집들로 분할하여 시각화한다. 특정주제 검색, 사진분류 등의 일반적인 사진탐색 예제를 통하여 제안된 방법의 유용성을 확인하였다.

• Asian-Australasian Journal of Animal Sciences
• /
• 제12권4호
• /
• pp.515-519
• /
• 1999

With the purpose of determining the influence of sexual partners on the oestrous behaviour and to evaluate the accuracy of predicting the time from implant withdrawal to sexual receptivity following a treatment with Synchromate B (SMB), 15 adult Brahman cows were used in each of three phases. During phase I and n, random pairs of animals were induced to display oestrus one pair after the other at daily intervals, while in phase III, cows were induced alternately, every other day, one cow on the 1st day, two on the 3rd, one on the 5th, two on the 7th until all cows were treated. Sixty six percent of the cows in phases I and II, and 80% in phase III came into oestrous after treatment. The interval between implant withdrawal and, expected and observed oestrous was statistically different in all phases. Clustering of oestrous was evident. Cows displayed sexual receptivity within a. range of -24 to +96; -24 to +72 and -216 to +192 hours after implant withdrawal for the three phases, respectively, with a tendency for cows treated first (within treatments), to delay their oestrus signs and vice versa. In phase III, four cows showed oestrous behaviour with the implant in place. These in spite of not observing pre-ovulatory follicles. Correlation values of 0.99, 0.93 and 0.90 (P<0.05) were found respectively among treatments, between the number of cows coming into oestrus and the number of mounts observed. These findings suggest that there are social and behavioural factors in a herd that may override exogenous synchronisation treatments.