• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.343-344
• /
• 2011

최근 3 차원 디스플레이 기술의 발전에 힘입어 3 차원 컨텐츠에 대한 수요도 늘고 있다. 스테레오스코픽(Stereoscopic) 렌즈를 이용하여 3 차원 컨텐츠를 만들거나 여러 장의 2 차원 영상을 이용한 3 차원 복원 연구가 활발히 진행되는 가운데 본 논문에서는 단일 2 차원 영상을 이용해서 깊이 지도를 획득하는 알고리즘을 제안한다. 단일 영상을 보고 3 차원 구조를 파악하는 인간의 시각 체계의 능력에 착안하여 본 논문에서는 단일 영상을 이용하여 깊이 정보를 추출하는 알고리즘을 제안한다. 깊이 단서들 중, 가림 단서를 소개하고 추가로 인간의 시각 체계에서 사용하는 깊이 단서들을 결합하여 기계 학습 알고리즘에 접목시킨다. 실험을 통해 우리는 제안 알고리즘이 물체의 외곽정보를 이용하여 양질의 깊이 지도를 준다는 것을 확인할 수 있다.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2004.10a
• /
• pp.345-351
• /
• 2004

This Paper describes a nonlinear adaptive noise filter using neural network for digital controller system. Back-Propagation Learning Algorithm based MLP (Multi Layer Perceptron)is used an adaptive filters. In this paper. it assume that the noise of primary input in the adaptive noise canceller is not the same characteristic as that of the reference input. Experimental reaults show that the neural network base noise canceller outperforms the linear noise canceller. Especially to make noise cancel close to realtime, Primary input is divided by unit and each divided part is processed for very short time than all the processed data are unified to whole data.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.63-69
• /
• 1999

최근 다국어 정보검색, 기계번역 등과 관련하여 자동 음차 표기 및 복원에 대한 필요성이 증대되고 있다. 특히 영어와 한국어 같이 그 음운구조의 차이가 큰 언어 쌍인 경우에는 간단한 문제가 아니다. 더구나 외래어를 영어로 복원하는 것은 표기의 경우보다 훨씬 어렵다. 본 논문에서는 결정트리 학습을 통한 한/영 자동 음차 복원 방법을 제안하고 기존의 방법 및 로마자 표기법에 기반한 방법에 비교하여 매우 정확하게 복원이 가능함을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.583-585
• /
• 2004

암의 발병을 조기에 예측하고 진단하는 것은 매우 중요하지만 그 과정이 매우 복잡하고 많은 노력이 필요하다. 암이 발생하는 원인은 매우 다양하지만 근본적으로 단백질을 형성하는 유전자에 변화가 오기 때문으로 생각해 볼 수 있다. 유전자 발현 정보로부터 기계적으로 암을 예측하기 위한 과정은 중요한 유전자의 선택, 모델의 학습, 모델을 이용한 예측과정으로 나뉘어 진다. 본 논문에서는 대장암 여부를 유전자 발현 데이터로부터 예측하기 위한 종분화 진화 신경망을 제안한다. 종분화 진화 신경망은 진화 알고리즘을 사용하여 신경망의 구조를 결정하고 종분화 알고리즘을 사용하여 다양한 개체의 생성을 유도한 후 모델의 앙상블을 통해 보다 높은 성능을 내는 방법이다 실험 결과 제안하는 방법이 대장암 예측 cross validation 테스트에서 96.5%의 높은 성능을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.11a
• /
• pp.6-9
• /
• 2010

2007년 Amazon.com이 미국에서 e-Book 전용 단말기 'Kindle'을 출시한 이래, Sony와 대형 서점 Barnes&Noble등 메이저 업체는 물론 다수의 중소업체들이 e-Book 시장에 진출하고 있다. 최근에는 Apple이 iPad를 출시하고 e-Book 시장에 진출한 가운데, Google 역시 6월 이후 e-Book 시장에 진출할 것을 발표함으로써 e-Book 시장의 경쟁이 더욱 치열해지고 있다. e-Book의 급속한 보급증가와 함께 이런 방대한 도서를 관리하는 곳에서 자동 도서 분류의 필요성도 증가하고 있다. 기존의 문서분류 방법들은 대게 수작업, 텍스트(단어)의 집합으로 간주하여 기계 학습방법을 그대로 적용하거나 약간의 변형을 가한 방법들이 대부분 이었다. 본 제안서에서는 데이터 마이닝 분야에서 사용되는 DRFP-Tree 구조를 이용하여 e-Book 내의 문장들의 패턴을 저장하고 이를 사용하여 e-Book을 분류하는 방법을 제안한다.

• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.29-39
• /
• 2021

As damages to individuals, private sectors, and businesses increase due to newly occurring cyber attacks, the underlying network security problem has emerged as a major problem in computer systems. Therefore, NIDS using machine learning and deep learning is being studied to improve the limitations that occur in the existing Network Intrusion Detection System. In this study, a deep learning-based NIDS model study is conducted using the Convolution Neural Network (CNN) algorithm. For the image classification-based CNN algorithm learning, a discrete algorithm for continuity variables was added in the preprocessing stage used previously, and the predicted variables were expressed in a linear relationship and converted into easy-to-interpret data. Finally, the network packet processed through the above process is mapped to a square matrix structure and converted into a pixel image. For the performance evaluation of the proposed model, NSL-KDD, a representative network packet data, was used, and accuracy, precision, recall, and f1-score were used as performance indicators. As a result of the experiment, the proposed model showed the highest performance with an accuracy of 85%, and the harmonic mean (F1-Score) of the R2L class with a small number of training samples was 71%, showing very good performance compared to other models.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.5
• /
• pp.173-187
• /
• 2018

This paper proposes a computer vision and deep learning-based technique for surveillance camera system for vehicle counting as one part of parking lot management system. We applied the You Only Look Once version 2 (YOLOv2) detector and come up with a deep convolutional neural network (CNN) based on YOLOv2 with a different architecture and two models. The effectiveness of the proposed architecture is illustrated using a publicly available Udacity's self-driving-car datasets. After training and testing, our proposed architecture with new models is able to obtain 64.30% mean average precision which is a better performance compare to the original architecture (YOLOv2) that achieved only 47.89% mean average precision on the detection of car, truck, and pedestrian.

• The Journal of Society for e-Business Studies
• /
• v.18 no.2
• /
• pp.95-110
• /
• 2013

This paper presents a novel feature engineering technique that can improve the conventional machine learning-based text classification systems. The proposed method extends the initial set of features by using hyperlink relationships in order to effectively categorize hypertext web documents. Web documents are connected to each other through hyperlinks, and in many cases hyperlinks exist among highly related documents. Such hyperlink relationships can be used to enhance the quality of features which consist of classification models. The basic idea of the proposed method is to generate a sort of ed concept feature which consists of a few raw feature words; for this, the method computes the semantic similarity between a target document and its neighbor documents by utilizing hierarchical relationships in the WordNet ontology. In developing classification models, the ed concept features are equated with other raw features, and they can play a great role in developing more accurate classification models. Through the extensive experiments with the Web-KB test collection, we prove that the proposed methods outperform the conventional ones.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.1
• /
• pp.83-97
• /
• 2021

Wireless sensors that make up the Wireless Sensor Network generally have extremely limited power and resources. The wireless sensor enters the sleep state at a certain interval to conserve power. The Sleep deflation attack is a deadly attack that consumes power by preventing wireless sensors from entering the sleep state, but there is no clear countermeasure. Thus, in this paper, using clustering-based binary search tree structure, the Sleep deprivation attack detection model is proposed. The model proposed in this paper utilizes one of the characteristics of both attack sensor nodes and normal sensor nodes which were classified using machine learning. The characteristics used for detection were determined using Long Short-Term Memory, Decision Tree, Support Vector Machine, and K-Nearest Neighbor. Thresholds for judging attack sensor nodes were then learned by applying the SVM. The determined features were used in the proposed algorithm to calculate the values for attack detection, and the threshold for determining the calculated values was derived by applying SVM.Through experiments, the detection model proposed showed a detection rate of 94% when 35% of the total sensor nodes were attack sensor nodes and improvement of up to 26% in power retention.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1367-1377
• /
• 2020

The purpose of this study is to assess the seismic vulnerability of buildings in Gyeongju city starting with the earthquake that occurred in the city on September 12, 2016, and produce a seismic vulnerability map. 11 influence factors related to geotechnical, physical, and structural indicators were selected to assess the seismic vulnerability, and these were applied as independent variables. For a dependent variable, location data of the buildings that were actually damaged in the 9.12 Gyeongju Earthquake was used. The assessment model was constructed based on random forest (RF) as a mechanic study method and support vector machine (SVM), and the training and test dataset were randomly selected with a ratio of 70:30. For accuracy verification, the receiver operating characteristic (ROC) curve was used to select an optimum model, and the accuracy of each model appeared to be 1.000 for RF and 0.998 for SVM, respectively. In addition, the prediction accuracy was shown as 0.947 and 0.926 for RF and SVM, respectively. The prediction values of the entire buildings in Gyeongju were derived on the basis of the RF model, and these were graded and used to produce the seismic vulnerability map. As a result of reviewing the distribution of building classes as an administrative unit, Hwangnam, Wolseong, Seondo, and Naenam turned out to be highly vulnerable regions, and Yangbuk, Gangdong, Yangnam, and Gampo turned out to be relatively safer regions.