• 융합정보논문지
• /
• 제11권6호
• /
• pp.49-54
• /
• 2021

최근 스마트 헬스케어 기술의 발전에 따라 일상적인 질환에 대한 관심이 증가하고 있다. 이에 따라 헬스케어 데이터를 통해 예측 모델로 질병을 분석하거나 예측하는 연구들이 증가하고 있다. 그러나 헬스케어 데이터에는 양성 데이터와 음성 데이터의 불균형이 존재한다. 이는 특정 질환을 가진 환자에 비하여 상대적으로 환자가 아닌 사람이 많아 데이터 수집에 어려움이 있어 발생하는 현상이다. 데이터 불균형은 질병 예측 및 탐지 시 진행하는 모델의 성능에 영향을 끼치기 때문에 이를 제거할 필요가 있다. 따라서 본 연구에서는 오버샘플링과 결측값 대치를 통해서 데이터 불균형을 해소한다. AutoML을 기반으로 여러 모델의 성능을 파악하고 모델 중 상위 3개의 모델을 앙상블한다.

• 한국통신학회논문지
• /
• 제36권2C호
• /
• pp.78-85
• /
• 2011

본 논문에서는 카메라 자체에서 움직임을 검출하고 분류된 객체를 추척할 수 있는 지능형 PTZ 카메라 시스템을 제안하였다. 추적하고자 하는 객체가 검출되면 분류하고, 객체의 움직임에 따라 PTZ 카메라가 실시간으로 추적한다. 검출을 위해 GMM을 사용하였고 검출성능을 높이기 위해 그림자 제거 기법을 적용하였다. 검출된 객체의 분류를 위해 Legendre 모멘트를 적용하였다. 본 논문에서는 카메라의 초점 조절을 사용하지않고 영상의 중심과 객체와의 방향, 거리, 속도 정보만을 이용하여 PTZ 카메라의 움직임을 제어하는 방법을 제안하였다. TI DM6446 Davinci를 이용하여 실시간으로 객체의 검출, 분류와 추적이 가능한 카메라 시스템을 구성하였다. 실험 결과 사람과 차량을 구분하고, 움직임의 속도가 빠른 차량에 대해서도 본 추적시스템은 안정적으로 동작함을 확인하였다.

• 한국전자파학회논문지
• /
• 제19권3호
• /
• pp.294-303
• /
• 2008

ISAR(Inverse Synthetic Aperture Radar) 영상은 표적에 대한 RCS(Radar Cross Section)를 2차원 공간에 표현하며, 표적구분에 이용될 수 있다. 2차원 IFFT(Inverse fast Fourier Transform)를 이용하여 쉽고 빠르게 ISAR 영상을 만들 수 있다. 하지만 IFFT를 이용하여 만든 ISAR 영상은 측정된 주파수 대역 폭과 각도 영역이 작아질 경우 해상도가 떨어지게 된다. 이를 해결하기 위해 AR(Auto Regressive), MUSIC(Multiple SIgnal Classification), Modified MUSIC과 같은 고해상도 스펙트럼 예측 기법을 이용하여 주파수 대역 폭과 각도 영역이 작아도 높은 해상도의 ISAR 영상을 만들 수 있다. 본 논문에서는 IFFT, AR, MUSIC, Modified MUSIC 기법을 적용하여 만든 ISAR 영상을 이용하여 표적 구분에 이용하고, 표적 구분에 적절한 ISAR 영상을 얻기 위한 고해상도 기법을 연구한다. 그리고 표적 구분 결과를 보여준다.

• 대한산업공학회지
• /
• 제28권2호
• /
• pp.147-161
• /
• 2002

In novelty detection, one attempts to discriminate abnormal patterns from normal ones. Novelty detection is quite difficult since, unlike usual two class classification problems, only normal patterns are available for training. Auto-Associative Multi-Layer Perceptron (AAMLP) has been shown to provide a good performance based upon the property that novel patterns usually have larger auto-associative errors. In this paper, we give a mathematical analysis of 2-layer AAMLP's output characteristics and empirical results of 2-layer and 4-layer AAMLPs. Various activation functions such as linear, saturated linear and sigmoid are compared. The 2-layer AAMLPs cannot identify non-linear boundaries while the 4-layer ones can. When the data distribution is multi-modal, then an ensemble of AAMLPs, each of which is trained with pre-clustered data is required. This paper contributes to understanding of AAMLP networks and leads to practical recommendations regarding its use.

• Industrial Engineering and Management Systems
• /
• 제13권4호
• /
• pp.454-462
• /
• 2014

Retention of possible churning customer is one of the most important issues in customer relationship management, so companies try to predict churn customers using their large-scale high-dimensional data. This study focuses on dealing with large data sets by reducing the dimensionality. By using six different dimension reduction methods-Principal Component Analysis (PCA), factor analysis (FA), locally linear embedding (LLE), local tangent space alignment (LTSA), locally preserving projections (LPP), and deep auto-encoder-our experiments apply each dimension reduction method to the training data, build a classification model using the mapped data and then measure the performance using hit rate to compare the dimension reduction methods. In the result, PCA shows good performance despite its simplicity, and the deep auto-encoder gives the best overall performance. These results can be explained by the characteristics of the churn prediction data that is highly correlated and overlapped over the classes. We also proposed a simple out-of-sample extension method for the nonlinear dimension reduction methods, LLE and LTSA, utilizing the characteristic of the data.

• ETRI Journal
• /
• 제44권4호
• /
• pp.613-623
• /
• 2022

Autism spectrum disorder (ASD) is a developmental disorder associated with cognitive and neurobehavioral disorders. It affects the person's behavior and performance. Autism affects verbal and non-verbal communication in social interactions. Early screening and diagnosis of ASD are essential and helpful for early educational planning and treatment, the provision of family support, and for providing appropriate medical support for the child on time. Thus, developing automated methods for diagnosing ASD is becoming an essential need. Herein, we investigate using various machine learning methods to build predictive models for diagnosing ASD in children using facial images. To achieve this, we used an autistic children dataset containing 2936 facial images of children with autism and typical children. In application, we used classical machine learning methods, such as support vector machine and random forest. In addition to using deep-learning methods, we used a state-of-the-art method, that is, automated machine learning (AutoML). We compared the results obtained from the existing techniques. Consequently, we obtained that AutoML achieved the highest performance of approximately 96% accuracy via the Hyperpot and tree-based pipeline optimization tool optimization. Furthermore, AutoML methods enabled us to easily find the best parameter settings without any human efforts for feature engineering.

• 전자공학회논문지S
• /
• 제34S권11호
• /
• pp.114-125
• /
• 1997

This paper presents a method that automatically recognizes dimension sets from the mechanical drawings, and that classifies 6 types dimension sets according to functional purpose. In the proposed method, the object and closed-loop symbols are separated from the character-free drawings. Then object lines and interpretation lines are vectorized. And, after recognizing dimension sets(consistings of arrowhead, shape line, tail lines, extension lines, text-string, and feature control frame), we classify recognized dimension sets as horizontal, vertical, angular, diametral, radial, and leader dimension sets. Finally the proposed method converts classified dimension sets into AutoCAD data by using AutoLisp language. By using the methods of geometric modeling, the proposed method readily recognized and classifies dimension sets from complex drawings. Experimetnal results are presented, which are obtained by applying the proposed method to drawings drawn in compliance with the KS drafting standard.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.360-362
• /
• 2016

본 논문에서는 실 승마 코칭을 수행하기 위해 Stacked 오토엔코더를 이용한 승마 보법을 분류하고자 한다. Staked Auto-encoder(SAE)에서 은닉층 수를 조절하여 승마데이터에 적합하게 쌓고, 성능을 비교하고 은닉층의 수를 수정한다. 데이터베이스 구축 환경은 16개의 관성센서로 이루어진 무선 네트워크로 구성된 슈트를 착용하고 국가대표급 승마 전문가로부터 데이터베이스를 취득한다. DB를 이용하여 보법별(평보, 속보, 경속보, 구보)로 각각 특징들(볼기 y축 포지션, 허리각도)을 이용하여 보법분류를 한다. 구축된 승마 모션데이터로 실험한 결과, 은닉층의 수가 1층일 때 성능은 95%를 보여주었고 은닉층의 수가 2층일 때 94%의 성능을 나타내었다.

• 자동차안전학회지
• /
• 제15권4호
• /
• pp.32-38
• /
• 2023

With the transition from internal combustion engine vehicles to eco-friendly cars, it has become essential to systematically construct an education system for electric vehicles based on high-voltage electric devices. In this study, we discussed the establishment of an educational system for electric vehicles based on high-voltage electric devices and proposed methods for qualifications after completing the education. To ensure systematic education, we presented a classification of learners according to their levels and job competencies. Additionally, we emphasized the importance of providing adequate practical training equipment for courses that require higher qualifications. Finally, to distinguish between the levels of completion of training and practical skills, we highlighted the necessity of implementing a system to certificates to individuals who have successfully completed the systematic training program.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권1호
• /
• pp.10-17
• /
• 2017

We propose classification algorithms for human and dog movement. The proposed algorithms use micro-Doppler signals obtained from humans and dogs moving in four different directions. A two-stage classifier based on a support vector machine (SVM) is proposed, which uses a radial-based function (RBF) kernel and $16^{th}$-order linear predictive code (LPC) coefficients as feature vectors. With the proposed algorithms, we obtain the best classification results when a first-level SVM classifies the type of movement, and then, a second-level SVM classifies the moving object. We obtain the correct classification probability 95.54% of the time, on average. Next, to deal with the difficult classification problem of human and dog running, we propose a two-layer convolutional neural network (CNN). The proposed CNN is composed of six ($6{\times}6$) convolution filters at the first and second layers, with ($5{\times}5$) max pooling for the first layer and ($2{\times}2$) max pooling for the second layer. The proposed CNN-based classifier adopts an auto regressive spectrogram as the feature image obtained from the $16^{th}$-order LPC vectors for a specific time duration. The proposed CNN exhibits 100% classification accuracy and outperforms the SVM-based classifier. These results show that the proposed classifiers can be used for human and dog classification systems and also for classification problems using data obtained from an ultra-wideband (UWB) sensor.