• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
• /
• pp.1279-1282
• /
• 2003

Estimation of human motion states is important enabling technologies for realizing a pervasive computing environment. In this paper, an improved method fur estimating human motion state from accelerometer data is introduced. Our method fur estimating human motion state utilizes various statistics of accelerometer data, such as mean, standard variation, skewness, kurtosis, eccentricity, as features for classification, and therefore is expected to be more robust than other existing methods that rely on only a few simple statistics. A series of experiments fur testing the effectiveness of the proposed method has been performed, and its result is presented.

• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2000년도 창립총회 및 춘계학술발표회
• /
• pp.24-27
• /
• 2000

The development of land classification and soil productivity rating systems (SPR) are examined for their application to valuation of agricultural land in South Dakota, USA. The application of SPR data to land valuation work conducted by real estate appraisers, tax assessors, and economists are discussed along with an assessment of its benefits and limitations.

• 전기전자학회논문지
• /
• 제24권1호
• /
• pp.97-105
• /
• 2020

열 영상은 온도에 따라 방출하는 에너지의 차이를 나타낸 영상이다. 주야간 사용이 가능하기 때문에 군사적인 용도로 많이 활용되고 있으나, 열 영상은 물체의 경계가 불명확하고 흐릿하게 표현되는 경우가 많으며 화염 등의 열기로 인해 경계부분이 변질되는 단점이 있다. 따라서, 열 영상을 이용하여 표적의 종류를 분류할 때 정확하게 분할된 경계선을 이용할 경우 효과적으로 분류 할 수 있지만, 물체의 경계가 잘못 추출되는 경우 분류의 정확도가 크게 감소한다. 본 논문에서는 이러한 단점을 극복하기 위해서 표적 영상의 분할 신뢰도에 따라 형태특징과 지역특징의 분류결과를 융합하는 계층적 분류기법을 제안하였으며, 연속 영상 기반으로 분류 결과를 갱신하는 기법을 새롭게 제안하여 차량 표적 분류 정확도를 개선하였다. 제안하는 방법은 실제 군용 표적 4종(전차, 장갑차, 상용차, 군용트럭)이 있는 다양한 자세의 열 영상 20,000장 이상을 이용하여 성능을 검증하였으며, 우수한 성능의 기존 방법 대비 정확도 개선에 효과가 있음을 확인하였다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2016년도 제54차 하계학술대회논문집 24권2호
• /
• pp.51-54
• /
• 2016

In computer vision especially in image processing, it has become popular to apply deep convolutional networks for supervised learning. Convolutional networks have shown a state of the art results in classification, object recognition, detection as well as semantic segmentation. However, supervised learning has two major disadvantages. One is it requires huge amount of labeled data to get high accuracy, the second one is to train so much data takes quite a bit long time. On the other hand, unsupervised learning can handle these problems more cheaper way. In this paper we show efficient way to learn features for classification in an unsupervised way. The network trained layer-wise, used backpropagation and our network learns features from unlabeled data. Our approach shows better results on Caltech-256 and STL-10 dataset.

• 한국공작기계학회논문집
• /
• 제13권3호
• /
• pp.24-29
• /
• 2004

The grinding process plays a key role which decides the quality of a product finally. But the grinding process is very irregular, so it is very difficult to analyse the process accurately. Therefore it is very important in the aspect of precision and automation to reduce the idle time and to decide the proper dressing time by watching. In this study, we choose the method which can be observed directly by using of computer vision and then apply pattern classification technique to the method of measuring the wheel surface. Pattern classification technique is proper to analyse complicated surface image. We observe the change of the wheel surface by using of the gray level run lengths which are representative in this technique.

• 한국생물정보학회:학술대회논문집
• /
• 한국생물정보시스템생물학회 2005년도 BIOINFO 2005
• /
• pp.85-90
• /
• 2005

In this paper, firstly we report experimental results on applying information extraction (IE) methodology to the task of summarizing clinical trial design information in focus on ‘Compared Treatment’, ‘Endpoint’ and ‘Patient Population’ from clinical trial MEDLINE abstracts. From these results, we have come to see this problem as one that can be decomposed into a sentence classification subtask and an IE subtask. By classifying sentences from clinical trial abstracts and only performing IE on sentences that are most likely to contain relevant information, we hypothesize that the accuracy of information extracted from the abstracts can be increased. As preparation for testing this theory in the next stage, we conducted an experiment applying state-of-the-art sentence classification techniques to the clinical trial abstracts and evaluated its potential in the original task of the summarization of clinical trial design information.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2000년도 춘계학술대회논문집 - 한국공작기계학회
• /
• pp.447-452
• /
• 2000

The grinding plays a key role which decide the quality of a product finally. But the grinding process is very irregular, so it is very difficult to analyse the process accurately. Therefore it is very important in the aspect of precision and automation to reduce the idle time and to decide the proper dressing time by visualizing. In this study, we choose the direct method of observation by making use of computer vision, and apply pattern classification technique to the method of measuring the wheel surface. Pattern classification technique is proper to analyse complex surface image. We observe the change of the wheel surface by making use of the gray level run lengths which are representative prince in this technique.

• Communications for Statistical Applications and Methods
• /
• 제24권6호
• /
• pp.543-559
• /
• 2017

Tree-based regression and classification ensembles form a standard part of the data-science toolkit. Many commonly used methods take an algorithmic view, proposing greedy methods for constructing decision trees; examples include the classification and regression trees algorithm, boosted decision trees, and random forests. Recent history has seen a surge of interest in Bayesian techniques for constructing decision tree ensembles, with these methods frequently outperforming their algorithmic counterparts. The goal of this article is to survey the landscape surrounding Bayesian decision tree methods, and to discuss recent modeling and computational developments. We provide connections between Bayesian tree-based methods and existing machine learning techniques, and outline several recent theoretical developments establishing frequentist consistency and rates of convergence for the posterior distribution. The methodology we present is applicable for a wide variety of statistical tasks including regression, classification, modeling of count data, and many others. We illustrate the methodology on both simulated and real datasets.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.101-102
• /
• 2007

Heart diseases are critical and should be detected as soon as possible. A stethoscope is a simple device to find cardiac disorder but requires keen experiences in heart sounds. We evaluate a cardiac disorder classifier by using heart sounds recorded by a digital wireless stethoscope developed in this work. The classifier uses hidden Markov models with circular state transition to model the heart sounds. We train the classifier using two kinds of data: One recorded by using our stethoscope and the other sampled from a clean heart sound database. In classification experiments using 165 sound clips, the classifier shows the classification accuracy of 82% in classifying 6 cardiac disorder categories.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2013년도 추계학술대회 논문집
• /
• pp.346-353
• /
• 2013

Industrial processes need to be monitored in real-time based on the input-output data observed during their operation. Abnormalities in an induction motor should be detected early in order to avoid costly breakdowns. To early identify induction motor faults, this paper effectively estimates spectral envelopes of each induction motor fault by utilizing a linear prediction coding (LPC) analysis technique and an expectation maximization (EM) algorithm. Moreover, this paper classifies induction motor faults into their corresponding categories by calculating Mahalanobis distance using the estimated spectral envelopes and finding the minimum distance. Experimental results shows that the proposed approach yields higher classification accuracies than the state-of-the-art approach for both noiseless and noisy environments for identifying the induction motor faults.