• Journal of KIISE
• /
• v.43 no.8
• /
• pp.869-877
• /
• 2016

In this paper, we propose an automated segmentation method of femoral cartilage in knee MR images using multi-atlas-based locally-weighted voting. The proposed method involves two steps. First, to utilize the shape information to show that the femoral cartilage is attached to a femur, the femur is segmented via volume and object-based locally-weighted voting and narrow-band region growing. Second, the object-based affine transformation of the femur is applied to the registration of femoral cartilage, and the femoral cartilage is segmented via multi-atlas shape-based locally-weighted voting. To evaluate the performance of the proposed method, we compared the segmentation results of majority voting method, intensity-based locally-weighted voting method, and the proposed method with manual segmentation results defined by expert. In our experimental results, the newly proposed method avoids a leakage into the neighboring regions having similar intensity of femoral cartilage, and shows improved segmentation accuracy.

• Korean Journal of Legislative Studies
• /
• v.26 no.2
• /
• pp.35-71
• /
• 2020

The purpose of this study is to verify the effect of electoral competition on voters deciding on strategic split-ticket voting under the mixed-member electoral system. As result, the more competitive the constituencies are, the more voters choose to vote for the major parties. The results of logistic regression analysis including interaction terms showed that the more competitive the constituencies are, the more voters choose for candidates from the major parties. Also, the finding shows that major party supporters are less affected by electoral competition than minor party supporters in choosing a candidate in the single-seat districts. In the case of minor party supporters, the more competitive the constituencies were, the more likely they were to choose the major party candidate instead of the minor party candidate. Based on these results, it can be inferred that voters are affected by the presence or behavior of other voters in local constituencies under the first-past-the-post rule. Because of the psychology of not wanting their votes to be useless, voters cast their ballots more strategically as the competition in constituencies intensifies, and as the competition in constituencies slackens, such tendencies weaken, and this trait is particularly evident among minor party supporters.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.4
• /
• pp.29-38
• /
• 2018

In this paper, we propose an automatic segmentation method of meniscus in knee MR images by automatic meniscus localization, multi-atlas-based locally-weighted voting, and patch-based edge feature classification. First, after segmenting the bone and knee articular cartilage, the volume of interest of the meniscus is automatically localized. Second, the meniscus is segmented by multi-atlas-based locally-weighted voting taking into account the weights of shape and intensity distribution in the volume of interest of the meniscus. Finally, to remove leakage to the collateral ligaments with similar intensity, meniscus is refined using patch-based edge feature classification considering shape and distance weights. Dice similarity coefficient between proposed method and manual segmentation were 80.13% of medial meniscus and 80.81 % for lateral meniscus, and showed better results of 7.25% for medial meniscus and 1.31% for lateral meniscus compared to the multi-atlas-based locally-weighted voting.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.1
• /
• pp.183-191
• /
• 2022

The Korea Stewardship Code 'Principles on the Fiduciary Responsibilities of Institutional Investors' was introduced in 2016 and the National Pension Service adopted it in 2018. the National Pension Service casted 'dessent' vote on the agenda which is able to reduce the ownership interest of shareholder in general meeting. This paper examines whether 'dissent' voting affected on the ownership interest of shareholder or not. The 'dissent' vote on the agenda are related to revision artical of corperation, appointment or compensation of director and auditor, approval of financial statements ect. The proxies of earnings management is discretionary accruals calculated by modified Jones model. The control variablies are size of assets, liabilities per assets, returns on assets. The results of this study are as followings. First, the 'dissent' voting on the agenda are related to revision artical of corperation, M&A, approval of financial statements ect. are not significant because their sample size is too small, Second, the 'dissent' voting on appointment of director and auditor affected on reduction of discretionary accruals. So the National Pension Service activism shall affect on increasing the ownership interest of shareholder. Third, the 'dissent' voting on compensation of director and auditor is not affected on reduction of discretionary accruals. This results show that 'unconditional dissent voting' on the agenda in general meeting is not to reduce the ownership interest of shareholder.

• Digital Contents
• /
• no.11 s.66
• /
• pp.72-73
• /
• 1998

제4차 JTC1/SC32 회의가 10월 21일 오후 3시 30분 한국데이터베이스진흥센터 회의실에서 개최됐다. 이번 회의는 개념스키마 모델링 설비(CSMF)의 최종위원회안(FCD)의 검토 및 투표, CSMF 프로젝트 분할 서면 투표, 메타데이터 오픈 포럼 안내, 컴퓨터 2천년 문제, 위원의 해임 및 추천 등에 대한 안건을 처리했다.

• Information Systems Review
• /
• v.25 no.4
• /
• pp.47-65
• /
• 2023

In this paper, we investigate how the online voting system can be a trust-based system from a technical perspective. Under four principles of voting, we finely evaluate the existing belief that offline voting is safer and more reliable than online voting based on procedural processes, technical principles. Many studies have suggested the ideas for implementing online voting system, but they have not attempted to strictly examine the technologies of online voting system from the perspective of voting requirements, and usually verification has been insufficient in terms of practical acceptance. Therefore, this study aims to analyze how the technologies are utilized to meet the demanding requirements of voting based on the technologies proven in the field. In addition to general data encryption, online voting requires more technologies for preventing data manipulation and verifying voting results. Moreover, high degree of confidentiality is required because voting data should not be exposed not only to outsiders but also to managers or the system itself. To this end, the security techniques such as Blind Signature, Bit Delegation and Key Division are used. In the case of blockchain-based voting, Mixnet and Zero-Knowledge Proof are required to ensure anonymity. In this study, the current status of the online voting system is analyzed based on the field system that actually serves. This study will enhance our understanding on online voting security technologies and contribute to build a more trust-based voting mechanism.

• Journal of the Korean Geographical Society
• /
• v.49 no.3
• /
• pp.371-389
• /
• 2014

This study aims to analyze the redistricting problems in non-autonomous Gu. Although non-autonomous Gu is a just local administrative district, it has been regarded as an important and basic spatial unit in electoral redistricting. By the reform of Public Official Election Act in 2012, however, non-autonomous Gu is distinguished from local governments like Si, Gun and autonomous Gu, in boundary delimitation for the 19th National Assembly election, and some are divided into a part of another constituency. About these background, this study points out the following problems. First, in national scale, the reform of Act made the malapportionment in constituencies of non-autonomous Gus, comparing with those of local governments. Second, there was the discriminative application of Act in each non-autonomous Gu and it will make the malapportionment worse in next election, considering the reorganization of local administrative system. Finally, this study propose that it is necessary to select one from a variety of redistricting principles, especially between the prevention of gerrymandering, the representativeness of local government and the apportionment, prior to another amendment of redistricting system and the debate about political reform.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.27-32
• /
• 2019

Research that implements the classification process through Deep Learning algorithm, one of the outstanding human abilities, includes a unimodal model, a multi-modal model, and a multi-modal method using music videos. In this study, the results were better by suggesting a system to analyze each song's spectrum into short samples and vote for the results. Among Deep Learning algorithms, CNN showed superior performance in the category of music genre compared to RNN, and improved performance when CNN and RNN were applied together. The system of voting for each CNN result by Deep Learning a short sample of music showed better results than the previous model and the model with Softmax layer added to the model performed best. The need for the explosive growth of digital media and the automatic classification of music genres in numerous streaming services is increasing. Future research will need to reduce the proportion of undifferentiated songs and develop algorithms for the last category classification of undivided songs.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.71-80
• /
• 2024

In this study, a proposed ensemble learning technique aims to enhance the semantic segmentation performance of images captured by Unmanned Aerial Vehicles (UAVs). With the increasing use of UAVs in fields such as urban planning, there has been active development of techniques utilizing deep learning segmentation methods for land cover segmentation. The study suggests a method that utilizes prominent segmentation models, namely U-Net, DeepLabV3, and Fully Convolutional Network (FCN), to improve segmentation prediction performance. The proposed approach integrates training loss, validation accuracy, and class score of the three segmentation models to enhance overall prediction performance. The method was applied and evaluated on a land cover segmentation problem involving seven classes: buildings,roads, parking lots, fields, trees, empty spaces, and areas with unspecified labels, using images captured by UAVs. The performance of the ensemble model was evaluated by mean Intersection over Union (mIoU), and the results of comparing the proposed ensemble model with the three existing segmentation methods showed that mIoU performance was improved. Consequently, the study confirms that the proposed technique can enhance the performance of semantic segmentation models.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.267-270
• /
• 2007

기계학습에서 분류기들의 집합으로 구성된 앙상블 분류기는 단일 분류기에 비해 정확도가 높다는 것이 입증되었다. 본 논문에서는 새로운 앙상블 학습으로서 데이터의 지역 기반 분류기들의 앙상블 학습을 제시하여 기존의 앙상블 학습과의 비교를 통해 성능을 검증하고자 한다. 지역 기반 분류기의 앙상블 학습은 데이터의 분포가 지역에 따라 다르다는 점에 착안하여 학습 데이터를 분할하고 해당하는 지역에 기반을 둔 분류기들을 만들어 나간다. 이렇게 만들어진 분류기들로부터 지역에 따라 가중치를 둔 투표를 하여 앙상블 방법을 이끌어낸다. 본 논문에서 제시한 앙상블 분류기의 성능평가를 위해 UCI Machine Learning Repository에 있는 11개의 데이터 셋을 이용하여 단일 분류기와 기존의 앙상블 분류기인 배깅과 부스팅등의 정확도를 비교하였다. 그 결과 기본 분류기로 나이브 베이즈와 SVM을 사용했을 때 새로운 앙상블 방법이 다른 방법보다 좋은 성능을 보이는 것을 알 수 있었다.