• 수면정신생리
• /
• 제8권1호
• /
• pp.67-71
• /
• 2001

Independent Component Analysis (ICA) is a blind source separation method using unsupervised learning and mutual information theory created in the late eighties and developed in the nineties. It has already succeeded in separating eye movement artifacts from human scalp EEG recording. Several characteristic sleep waves such as sleep spindle, K-complex, and positive occipital sharp transient of sleep (POSTS) can be recorded during sleep EEG recording. They are used as stage determining factors of sleep staging and might be reflections of unknown neural sources during sleep. We applied the ICA method to sleep EEG for sleep waves separation. Eighteen channel scalp longitudinal bipolar montage was used for the EEG recording. With the sampling rate of 256Hz, digital EEG data were converted into 18 by n matrix which was used as a original data matrix X. Independent source matrix U (18 by n) was obtained by independent component analysis method ($U=W{\timex}X$, where W is an 18 by 18 matrix obtained by ICA procedures). ICA was applied to the original EEG containing sleep spindle, K-complex, and POSTS. Among the 18 independent components, those containing characteristic shape of sleep waves could be identified. Each independent component was reconstructed into original montage by the product of inverse matrix of W (inv(W)) and U. The reconstructed EEG might be a separation of sleep waves without other components of original EEG matrix X. This result (might) demonstrates that characteristic sleep waves may be separated from original EEG of unknown mixed neural origins by the Independent Component Analysis (ICA) method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2328-2344
• /
• 2022

2D human pose estimation still faces difficulty in low-resolution images. Most existing top-down approaches scale up the target human bonding box images to the large size and insert the scaled image into the network. Due to up-sampling, artifacts occur in the low-resolution target images, and the degraded images adversely affect the accurate estimation of the joint positions. To address this issue, we propose a multi-resolution input feature fusion network for human pose estimation. Specifically, the bounding box image of the target human is rescaled to multiple input images of various sizes, and the features extracted from the multiple images are fused in the network. Moreover, we introduce a guiding channel which induces the multi-resolution input features to alternatively affect the network according to the resolution of the target image. We conduct experiments on MS COCO dataset which is a representative dataset for 2D human pose estimation, where our method achieves superior performance compared to the strong baseline HRNet and the previous state-of-the-art methods.

• 경영정보학연구
• /
• 제18권2호
• /
• pp.173-198
• /
• 2016

본 연구의 목적은 비즈니스 인텔리전스 연구 관점에서 기업부실화 예측 성능을 향상키시는 것이다. 이를 위해 본 연구는 기존 연구들에서 미흡하게 다루어졌던 1) 데이터셋을 구성하는 과정에서 발생하는 바이어스 문제, 2) 거시경제위험 요소의 미반영 문제, 3) 데이터 불균형 문제, 4) 서술적 바이어스 문제를 다루어 경기순환국면을 반영한 기업부실화 예측 프레임워크를 제안하고, 이를 바탕으로 기업부실화 예측 모델을 개발하였다. 본 연구에서는 경기순환국면별로 각각의 데이터셋을 구성하고, 각 데이터셋에서 의사결정나무, 인공신경망 등 단일 분류기부터 앙상블 기법까지 다양한 데이터마이닝 알고리즘을 적용하여 실험하였다. 또한 본 연구는 데이터불균형 문제를 해결하기 위해, 오버샘플링 기법인 SMOTE(synthetic minority over-sampling technique) 기법을 통해 초기 데이터 불균형 상태에서부터 표본비율을 1:1까지 변화시켜 가며, 기업부실화 예측 모델을 개발하는 실험을 하였고, 예측 모델의 변수 선정 시에 선행연구를 바탕으로 재무비율을 추출하고, 여기서 파생된 IT 산출물인 재무상태변동성과 산업수준상태변동성을 예측 모델에 삽입하였다. 마지막으로, 본 연구는 각 순환국면에서 만들어진 기업부실화 예측 모델의 예측 성능 비교와 경기 확장기와 수축기에서의 기업부실화 예측 모델의 유용성에 대해 논의하였다. 본 연구는 비즈니스 인텔리전스 연구 측면에서 기존 연구에서 미흡하게 다루어졌던 4가지 문제점을 검토하고, 이를 해결할 프레임워크를 제안함으로써 기존 연구 대비 기업부실화 예측률을 10% 이상 향상시켰다는 점에서 연구의 의의를 찾을 수 있다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.142-145
• /
• 2005

Since the successful launch of FORMOSAT -2 satellite by National Space Organization of Taiwan in May 2004, the Remote Sensing Instrument (RSI) on- board the FORMOSAT -2 has continuously acquired images at one panchromatic and four multi-spectral bands (http://www.nspo.org.tw). In general, the RSI performs well and receives high quality images which proved to be very useful for various applications. However, some RSI panchromatic products exhibit obvious trembling artifact that must be removed. Preliminary study reveals that the trembling artifact is caused by the instability of the spacecraft attitude. Though the magnitude of this artifact is actually less than half of a pixel, it affects the applicability of panchromatic products. A procedure removing this artifact is therefore needed for providing image products of consistent quality. Due to the nature of trembling artifact, it is impossible to describe the trembling amount by employing an analytic model. Relied only on image itself, an algorithm determining trembling amount and removing accordingly the trembling artifact is proposed. The algorithm consists of 3 stages. First, a cross-correlation based scheme is used to measure the relative shift between adjacent scan lines. Follows, the trembling amount is estimated from the measured value. For this purpose, the Fourier transform is utilized to characterize random shifts in frequency domain. An adaptive estimation method is then applied to deduce the approximate trembling amount. In the subsequent stage, image re-sampling operation is applied to restore the trembling-free product. Experimental results show that by applying the proposed algorithm, the unpleasant trembling artifact is no longer evident.

• 방송공학회논문지
• /
• 제18권1호
• /
• pp.10-20
• /
• 2013

현재 멀티미디어 장치와 사용자의 요구가 다양해짐에 따라 이를 충족시키기 위하여 이미지 크기는 장치에서 지원하는 해상도나 사용자의 요구에 맞게 조정되어야 한다. 이미지 보간 기법은 크게 공간 도메인과 주파수 도메인에서 수행 될 수 있다. 일반적으로 공간 도메인에서의 보간 방법은 주파수 도메인의 보간 기법에 비해 상대적으로 주관적인 화질 측면에서 좋은 성능을 나타내지만 오버 스무딩 현상으로 인해 객관적인 성능이 낮다. 반대로 주파수 도메인에서의 업샘플링 방법은 객관적인 화질 측면에서 좋은 성능을 나타내지만 블록 열화 현상이 나타나 주관적인 화질 측면 성능이 낮다. 주파수 도메인에서의 보간 기법은 블록 열화 현상 외에도 모든 블록에 대해 동일한 방법으로 고주파 성분을 채워 넣기 때문에 영상의 edge가 많은 부분에서 주관적인 화질 성능을 떨어뜨린다. 본 논문에서는 DCT 계수 분포를 이용해 분류된 edge 방향에 따라 공간 도메인과 주파수 도메인에서의 보간 기법을 적응적으로 사용하는 알고리듬을 제안한다. 제안하는 알고리듬은 두 도메인의 장점을 모두 가지고 있다. 실험 결과를 통해 제안하는 알고리듬이 주관적인 화질 측면의 성능 향상과 함께 객관적인 화질도 향상됨을 알 수 있다.

• 대한의용생체공학회:의공학회지
• /
• 제21권4호
• /
• pp.419-431
• /
• 2000

본 논문에서는 디지털 스캔 컨버터(DSC, digital scan converter)에 의한 왜곡이 근본적으로 발생하지 않는 새로운 초음파 영상 방법으로서 표시장치의 화소에 직접 수신 집속하는 기법(DPBF, display-pixel-based focusing)과 이를 구현하기 위한 초음파 영상 장치의 구조를 제안한다. 제안한 기법은 극 좌표에 위치하는 샘플링 점들 대신에 직각좌표상의 화소들에 직접 수신 집속 함으로써 DSC의 역할을 제거하였다. DPBF 기법을 적용한 영상에는 DSC에 의한 허상 및 구획화 현상 등이 나타나지 않으며, 일반적인 방법에 의한 영상에 비해 보다 세밀한 영상을 제공한다는 사실을 검증할 수 있었다. 본 논문은 DSC에 의한 보간 효과가 두 개의 비교 영상 A, B 간에 미치는 정도를 정량적으로 평가하기 위해서 영상거침 비(ICRA/B, Image Coarseness Ratio)를 새로운 성능 측정 방법으로 정의하였다. 이 ICRA/B를 이용하여 비교 영상간 거친 정도를 정량적으로 측정하였다.

• 대한의용생체공학회:의공학회지
• /
• 제35권5호
• /
• pp.151-159
• /
• 2014

Computed tomography (CT) is one of the most widely used medical imaging modality. However, substantial x-ray dose exposed to the human subject during the CT scan is a great concern. Region-of-interest (ROI) CT is considered to be a possible solution for its potential to reduce the x-ray dose to the human subject. In most of ROI-CT scans, the ROI is set to a circular shape whose diameter is often considerably smaller than the full field-of-view (FOV). However, an arbitrarily shaped ROI is very desirable to reduce the x-ray dose more than the circularly shaped ROI can do. We propose a new method to make a non-circular convex-shaped ROI along with the image reconstruction method. To make a ROI with an arbitrary convex shape, dynamic collimations are necessary to minimize the x-ray dose at each angle of view. In addition to the dynamic collimation, we get the ROI projection data with slightly lower sampling rate in the view direction to further reduce the x-ray dose. We reconstruct images from the ROI projection data in the compressed sensing (CS) framework assisted by the exterior projection data acquired from the pilot scan to set the ROI. To validate the proposed method, we used the experimental micro-CT projection data after truncating them to simulate the dynamic collimation. The reconstructed ROI images showed little errors as compared to the images reconstructed from the full-FOV scan data as well as little artifacts inside the ROI. We expect the proposed method can significantly reduce the x-ray dose in CT scans if the dynamic collimation is realized in real CT machines.

• 대한원격탐사학회지
• /
• 제26권3호
• /
• pp.347-359
• /
• 2010

Ortho-photos provide valuable spatial and spectral information for various Geographic Information System (GIS) and mapping applications. The absence of relief displacement and the uniform scale in ortho-photos enable interested users to measure distances, compute areas, derive geographic locations, and quantify changes. Differential rectification has traditionally been used for ortho-photo generation. However, differential rectification produces serious problems (in the form of ghost images) when dealing with large scale imagery over urban areas. To avoid these artifacts, true ortho-photo generation techniques have been devised to remove ghost images through visibility analysis and occlusion detection. So far, the Z-buffer method has been one of the most popular methods for true ortho-photo generation. However, it is quite sensitive to the relationship between the cell size of the Digital Surface Model (DSM) and the Ground Sampling Distance (GSD) of the imaging sensor. Another critical issue of true ortho-photo generation using high resolution satellite imagery is the scan line search. In other words, the perspective center corresponding to each ground point should be identified since we are dealing with a line camera. This paper introduces alternative methodology for true ortho-photo generation that circumvents the drawbacks of the Z-buffer technique and the existing scan line search methods. The experiments using real data are carried out while comparing the performance of the proposed and the existing methods through qualitative and quantitative evaluations and computational efficiency. The experimental analysis proved that the proposed method provided the best success ratio of the occlusion detection and had reasonable processing time compared to all other true ortho-photo generation methods tested in this paper.

• 한국지구과학회지
• /
• 제34권3호
• /
• pp.257-273
• /
• 2013

본 연구는 과학관 도슨트의 활동에 대한 전문화 방안을 모색하는데 목적을 두고 8년간 도슨트 활동을 지속한 박 도슨트를 연구 참여자로 선정하여 생애사적 접근방법으로 연구하였다. 박 도슨트 선정과정은 다음과 같다. 1단계 목적표집, 편의표집을 통해 56명의 도슨트를 선정, 2단계 심화연구 수락자를 기준으로 목적표집을 통해 11명의 도스트를 선정했다. 11명을 대상으로 설문지 분석과 심층면담을 거쳐 본 연구에 적극적인 1인을 3단계 목적표집을 통해 선정하였다. 주로 면담을 실시하고 참여관찰과 문서 등의 수집 분석을 통해 박 도슨트가 어떻게 도슨트 활동을 하게 되었는지, 그리고 현재의 도슨트 활동은 어떠한지, 도슨트 전문성을 위해 어떠한 노력을 하는지 등을 생애사적으로 기술하였다. 연구의 결과는 다음과 같다. 첫째, 박 도슨트는 교육적 신념이 투철하며 이는 생애사적 경험을 통해 형성된 신념이 무엇보다도 도슨트 활동에 큰 영향을 미치고 있음을 알 수 있다. 둘째, 박 도슨트는 활동을 통해서 전시해설에 관한 실천적 지식이 형성 및 체계화 되고 있었다. 셋째, 박 도슨트는 전문가로 활동하기 위해서 전시해설을 위한 전문성을 기르고 이를 좀 더 효과적으로 획득할 수 있는 방법으로 주변 환경을 충분히 활용하고 있었다. 연구 결과를 바탕으로 체계적인 도슨트 제도의 구축 및 운영과 도슨트 활동에 따른 지원 및 구축은 과학관에서의 도슨트 교육의 활성화를 위해서는 중요한 부분임을 인식할 수 있다.

• 한국대기환경학회지
• /
• 제25권6호
• /
• pp.523-538
• /
• 2009

Recently, there has been a keen demand for real-time automatic monitoring of VOCs not only in Korea but other developed countries. We carried out this study to evaluate and to optimize the performance of a continuous automatic monitoring system for hazardous VOCs (HVOCs) in the ambient atmosphere, using an on-line GC system. The online system normally consisted of a Nafion dryer prior to a cold trap of an automatic thermal desorption apparatus and a GC system equipped with two detectors, i.e. PID and ECD. Preliminary tests conducted to check out any contamination of the system revealed an evidence of significant artifact formation of benzene, and it was found that the Nafion dryer (even brand new one) is the source of the benzene artifact. Thus, all the subsequent experiments in this study was carried out inevitably by removing the Nafion dryer. The on-line GC method was investigated with a variety of QC/QA performance criteria such as repeatability, linearity, lower detection limits, and accuracy. In order to find out the best operating condition for the on-line GC system, three different types (in terms of adsorption strength) of cold trap combinations were tested, i.e. (i) Tenax-TA and Carbopack-B combination (weak and hydrophobic); (ii) Tenax-TA, Carbopack-X and Carboxen-1000 combination (strong and hydrophilic); and (iii) Tenax-TA and Carbopack-X combination (medium and hydrophobic/hydrophilic). The USEPA TO-17 manual method was selected as a reference method to evaluate the performance of the on-line method. A series of experiments revealed that the system performance was superior to others when a cold trap packed with hydrophilic adsorbents (Tenax-TA/Carbopack-X/Carboxen-1000 combination) was used and operated at $25^{\circ}C$. However, the system with a cold trap packed with a combination of Tenax-TA and Carbopack-X is more recommended for field applications since the carboxen-1000 adsorbent is too sensitive to water vapor, and hence the performance of the system might be very unstable to humid samples or during rainy days. Furthermore, the precision and accuracy criteria of the Tenax-TA/ Carbopack-X combination were generally compatible with the triple adsorbents cold trap. The continuous automatic monitoring method is, thus, considered very useful to real-time monitoring to understand the variations of VOCs concentrations in ambient air, as it adopts much simpler procedures in sampling, analysis, and data integration steps than manual monitoring methods. However, it should be noted that there is a high possibility of benzene artifacts formation through the Nafion dryer, which is often installed to remove water vapor in air samples before being adsorbed onto the cold trap. Therefore, if a Nafion dryer is used in any studies of monitoring VOCs, the benzene contamination should be carefully examined before carrying out obtaining the data.