• Journal of Information Display
• /
• 제4권4호
• /
• pp.8-12
• /
• 2003

In this paper, we report relatively simple equalizing pulse method for reducing dynamic false contour in PDP. Calculation cost is reduced by limiting pixels to add equalizing pulse and using look-up-table(LUT) for given subfield pattern. Pixels to be modified are determined after comparing selected number of most significant bits(MSB) with those of adjacent pixels. The equalizing pulse amount is determined by consulting LUT, which is for a fixed velocity of 1 pixel/tv field. Even though the suggested scheme does not cover every luminance combination of neighboring pixels, it is expected to work well after appropriate modifications are made according to the velocity.

• Journal of Multimedia Information System
• /
• 제8권4호
• /
• pp.233-242
• /
• 2021

Various anatomical MRI imaging biomarkers for Alzheimer's Disease (AD) identification have been recognized so far. Cortical and subcortical volume, hippocampal, amygdala volume, and genetics patterns have been utilized successfully to diagnose AD patients from healthy. These fundamental sMRI bio-measures have been utilized frequently and independently. The entire possibility of anatomical MRI imaging measures for AD diagnosis might thus still to analyze fully. Thus, in this paper, we merge different structural MRI imaging biomarkers to intensify diagnostic classification and analysis of Alzheimer's. For 54 clinically pronounce Alzheimer's patients, 58 cognitively healthy controls, and 99 Mild Cognitive Impairment (MCI); we calculated 1. Cortical and subcortical features, 2. The hippocampal subfield, amygdala nuclei volume using Freesurfer (6.0.0) and 3. Genetics (APoE ε4) biomarkers were obtained from the ADNI database. These three measures were first applied separately and then combined to predict the AD. After feature combination, we utilize the sequential feature selection [SFS (wrapper)] method to select the top-ranked features vectors and feed them into the Multi-Kernel SVM for classification. This diagnostic classification algorithm yields 94.33% of accuracy, 95.40% of sensitivity, 96.50% of specificity with 94.30% of AUC for AD/HC; for AD/MCI propose method obtained 85.58% of accuracy, 95.73% of sensitivity, and 87.30% of specificity along with 91.48% of AUC. Similarly, for HC/MCI, we obtained 89.77% of accuracy, 96.15% of sensitivity, and 87.35% of specificity with 92.55% of AUC. We also presented the performance comparison of the proposed method with KNN classifiers.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2002년도 Proceedings
• /
• pp.53-60
• /
• 2002

Motion of lung tumors from respiration has been reported in the literature to be as large as of 1-2 cm. This motion requires an additional margin between the Clinical Target Volume (CTV) and the Planning Target Volume (PTV). While such a margin is necessary, it may not be sufficient to ensure proper delivery of Intensity Modulated Radiotherapy (IMRT) to the CTV during the simultaneous movement of the DMLC. Gated treatment has been proposed to improve normal tissues sparing as well as to ensure accurate dose coverage of the tumor volume. The following questions have not been addressed in the literature: a) what is the dose error to a target volume without gated IMRT treatment\ulcorner b) what is an acceptable gating window for such treatment. In this study, we address these questions by proposing a novel technique for calculating the 3D dose error that would result if a lung IMRT plan were delivered without gating. The method is also generalized for gated treatment with an arbitrary triggering window. IMRT plans for three patients with lung tumor were studied. The treatment plans were generated with HELIOS for delivery with 6 MV on a CL2100 Varian linear accelerator with a 26 pair MLC. A CTV to PTV margin of 1 cm was used. An IMRT planning system searches for an optimized fluence map ${\Phi}$ (x,y) for each port, which is then converted into a dynamic MLC file (DMLC). The DMLC file contains information about MLC subfield shapes and the fractional Monitor Units (MUs) to be delivered for each subfield. With a lung tumor, a CTV that executes a quasi periodic motion z(t) does not receive ${\Phi}$ (x,y), but rather an Effective Incident Fluence EIF(x,y). We numerically evaluate the EIF(x,y) from a given DMLC file by a coordinate transformation to the Target's Eye View (TEV). In the TEV coordinate system, the CTV itself is stationary, and the MLC is seen to execute a motion -z(t) that is superimposed on the DMLC motion. The resulting EIF(x,y)is inputted back into the dose calculation engine to estimate the 3D dose to a moving CTV. In this study, we model respiratory motion as a sinusoidal function with an amplitude of 10 mm in the superior-inferior direction, a period of 5 seconds, and an initial phase of zero.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2002년도 Proceedings
• /
• pp.74-82
• /
• 2002

In Intensity Modulated Radiotherapy (IMRT), radiation is delivered in a multiple of Multileaf Collimator (MLC) subfields. A subfield with a small leaf-to-leaf opening is highly sensitive to a leaf-positional error. We introduce a method of identifying and rejecting IMRT plans that are highly sensitive to a systematic MLC gap error (sensitivity to possible random leaf-positional errors is not addressed here). There are two sources of a systematic MLC gap error: Centerline Mechanical Offset (CMO) and, in the case of a rounded end MLC, Radiation Field Offset (RFO). In IMRT planning system, using an incorrect value of RFO introduces a systematic error ΔRFO that results in all leaf-to-leaf gaps that are either too large or too small by (2ㆍΔRFO), whereas assuming that CMO is zero introduces systematic error ΔCMO that results in all gaps that are too large by ΔCMO = CMO. We introduce a concept of the Average Leaf Pair Opening (ALPO) that can be calculated from a dynamic MLC delivery file. We derive an analytic formula for a fractional average fluence error resulting from a systematic gap error of Δ$\chi$ and show that it is inversely proportional to ALPO; explicitly it is equal to, (equation omitted) in which $\varepsilon$ is generally of the order of 1 mm and Δx=2ㆍΔRFO+CMO. This analytic relationship is verified with independent numerical calculations.

• 조명전기설비학회논문지
• /
• 제30권3호
• /
• pp.1-8
• /
• 2016

In this paper, we proposed a new driving scheme to reduce the motion blur and save the power for OLEDs(organic light emitting diodes). We adopted a DVS (dynamic voltage scaling) method to reduce power consumption and the division of TV field to improve motion blur. In the proposed scheme, BEW (Blur Edge Width) was decreased to the ratio of 1/4 compared to the conventional scheme under the optimal conditions. In this scheme, the gray levels to which the DVS method can be applied were divided into much smaller groups depending on the number of subfields. Therefore, our scheme does not guarantee less power consumption for every image compared to the conventional scheme. However, the new scheme can move the gray levels adopting the DVS to higher gray levels. Thus, we can save power even when having images at high gray levels.

• 한국도서관정보학회지
• /
• 제52권1호
• /
• pp.109-128
• /
• 2021

본고에서는 LRM 노멘 개체를 수용하기 위한 전거제어시 고려사항을 모색하고자 문헌연구, RDA 규칙분석, 국내 목록 전문가 의견조사를 실시하였다. 그 결과 전거제어를 위해 노멘의 속성 요소, 목록기술, MARC 전거포맷 측면에서 고려사항을 제안하였다. 첫째, 노멘의 속성으로 LRM 노멘의 유형, 스킴, 이용대상자, 이용맥락, 참고정보원, 언어, 문자, 문자변환과 RDA에서 추가한 식별상태, 주기, 구분불가한 이름지시자의 기술을 가능한 상세히 기술해야 한다. 둘째, 노멘의 속성요소 및 관계요소를 기술하기 위해 목록규칙인 RDA에 제시된 대로 비구조화, 구조화, 식별자, IRI로 기술하며, 이중 구조화된 기술을 위해서는 어휘인코딩스킴(VES), 스트링인코딩스킴(SES)이 작성되어야 한다. 또한, 전거형접근점에 대한 구조 및 우선 이름/표제에 대한 목록 규칙도 마련되어야 한다. 셋째, LRM 노멘을 반영할 수 있도록 MARC 21 전거포맷을 마련하기 위해 Maxwell의 확장(안)을 바탕으로 추가 확장 방안을 제안하였다. (1) 노멘 마다 속성이 입력될 수 있도록 4XX, 5XX에 속성이 기술되어야 하며, 1XX, 5XX와 4XX에 기술될 노멘의 속성을 구분하였다. (2) 노멘 속성으로 유형, 언어, 문자, 문자변환, 이용맥락, 유효기간을 기술하기 위해서는 MARC 21내에 표시기호 및 식별기호가 추가되어야 한다. 이에 368, 381, 377의 식별기호 확장과 이용맥락과 유효기간을 기술하기 위한 표시기호를 추가할 것을 제안하였다. 본고에서 제안한 LRM 노멘을 위한 전거제어시 고려사항은 국내에서 LRM을 반영한 전거제어 방안을 마련하는 기반이 될 것이다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문집
• /
• pp.693-698
• /
• 2002

For the evaluation of the plasma display panel (PDP)'s noise, vibration and sound characteristics of fanless PDP are measured and investigated. PDP is a type of two-electrode vacuum tube which operates on the same principle as a household fluorescent light. An inert gas such as argon or neon is injected between two glass plates on which transparent electrodes have been formed, and the glass is illuminated by generating discharge. For this discharge, both high voltage and currents are needed and cause an acoustic noise. We investigated the noise characteristics connected with both a electromagnetic elements from SMPS to panel through X, Y and logic board, and a mechanical elements form panel to case through transfer path which related with vibration and heat. To reduce the noise of PDP, a discharge pulse memory design related with both higher brightness and lower power consumption is important and mechanical characteristics connected with dissipation process of both heat and vibration generated by panel discharge must be investigated.

• 한국도서관정보학회지
• /
• 제22권
• /
• pp.61-93
• /
• 1995

This paper discussed which of the three basic entries of serials that changed their titles-earliest, latest, or successive entry cataloging-is the most efficient, informative, and economical in a manual and an online environment. The results of the study are as follow: 1. Earliest entry cataloging or as an alternative, successive entry cataloging advocated by Cutter in 1976 was replaced in ALA Cataloging Rules by latest entry cataloging and with AACR1 and AACR2, latest entry cataloging was replaced by successive entry cataloging again. 2. Entry method of serial cataloging should be determined in consideration of goals of catalog, efficiency of catalog retrieval, theoretical and practical standpoint of entry methods, average cataloging time and cost, the state of adaption. 3. In a card catalog environment, if the numbering and class number of the serial that title proper changed into initials itself continue and the title history must not be too complicated, the serial is desirable to entry under the latest title. Except for above case, successive entry cataloging enables serials to be more quickly and economically made available to users than with either earliest or latest entry cataloging. 4. In an automated catalog environment, successive entry cataloging has more potential of satisfying and impairs the efficiency of both accessing and processing a library's serials collection. However, when initial cataloging by MARC format is done, careful screen design and complex programming will be needed. And subfield containing the local record number should be added to the programming means to link the various records together, i.e., 780 (preceding title) and 785 (succeeding title) fields for all related titles.

• 전기학회논문지
• /
• 제56권12호
• /
• pp.2189-2195
• /
• 2007

In this paper, we propose a newly developed selective reset waveform of a ac PDP using the wide erase pulse technique with the control of address bias voltage. Although it is generally understood that the wide pulse erasing methode shows the narrow driving margin in an opposite discharge type ac PDP, we could obtain a moderate driving margin in a 3-electrode surface discharge type ac PDP. The obtained driving margin shows a strong dependency on the sustain voltage and the address bias voltage. The lower the sustain and the address bias voltage, the wider the driving margin. The pulse width of the proposed waveform is only $10{\mu}s$, which gives additional time to the sustain period, hence increases the brightness. The brightness and contrast ratio increase about 20% together comparing to the conventional ramp type selective reset waveform with the driving scheme of 10 subfield ADS method. The driving margin was measured with the line by line addressed pattern on the white test panel of 2inch diagonal size and the discharge gas was Ne+Xe4%, 400torr.

• 정보처리학회 논문지
• /
• 제13권4호
• /
• pp.181-188
• /
• 2024

기존의 비속어 탐지 방법들은 의도적으로 변형된 비속어를 식별하는 데 한계가 있다. 이 논문에서는 자연어 처리의 한 분야인 개체명 인식에 기반한 새로운 방법을 소개한다. 우리는 시퀀스 레이블링을 이용한 비속어 탐지 기법을 개발하고, 이를 위해 한국어 악성 댓글 중 일부 비속어를 레이블링하여 직접 데이터셋을 구축하여 실험을 수행하였다. 또한 모델의 성능을 향상시키기 위하여 거대 언어 모델중 하나인 ChatGPT를 활용해 한국어 혐오발언 데이터셋의 일부를 레이블링을 하는 방식으로 데이터셋을 증강하여 학습을 진행하였고, 이 과정에서 거대 언어 모델이 생성한 데이터셋을 인간이 필터링 하는 것만으로도 성능을 향상시킬 수 있음을 확인하였다. 이를 통해 데이터셋 증강 과정에는 여전히 인간의 관리감독이 필요함을 제시하였다.