• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2009년도 춘계학술발표대회
• /
• pp.35-38
• /
• 2009

The paper presents an effective method to detect fire in video surveillance and monitoring system. The main contribution of this work is that we successfully use the Hidden Markov Models in the process of detecting the fire with a few preprocessing steps. First, the moving pixels detected from image difference, the color values obtained from the fire flames, and their pixels clustering are applied to obtain the image regions labeled as fire candidates; secondly, utilizing massive training data, including fire videos and non-fire videos, creates the Hidden Markov Models of fire and non-fire, which are used to make the final decision that whether the frame of the real-time video has fire or not in both temporal and spatial analysis. Experimental results demonstrate that it is not only robust but also has a very low false alarm rate, furthermore, on the ground that the HMM training which takes up the most time of our whole procedure is off-line calculated, the real-time detection and alarm can be well implemented when compared with the other existing methods.

• 대한두경부종양학회지
• /
• 제16권1호
• /
• pp.33-36
• /
• 2000

Objectives: Angiocentric T-cell lymphoma of the head and neck is an angiocentric and angiodestructive lymphoreticular proliferative disorder. It has been treated with various treatment modalities, but its prognosis is poor and the treatment modality is controversial. We performed this study to suggest a treatment modality with improved results. Materials and Methods: We studied 40 cases of pathologically confirmed angiocentric T-cell lymphoma from July 1984 to December 1996, 35 cases of which showed complete response after initial treatment. All the patients were divided into two groups according to treatment modality. 15 cases received radiotherapy alone (Group I) and 20 cases received radiotherapy after five cycles of CHOP-Bleo chemotherapy(Group II). We analyzed the subsites of tumor, stage, treatment modality and treatment outcome and causes of failure for each group, and compared the three-year no evidence of disease(NED) between the two groups. Results: The three-year NED of a combined chemoradiotherapy was higher than that of a radiotherapy alone (p=0.0478). The three-year NED according to groups and stage were as follows: Group I=6/15(40.0%), stage IE=5/10(50.0%), stage IIE=1/5(20%), Group II=13/20(65.0%), stage IE=9/13(69.2%), stage IIE=4/7(57.1%). Radiotherapy alone is not well effective for the nasal cavity lymphoma extended to paranasal sinus and the palate. Conclusion: We are unable to provide clear guidelines for treatment, but recommend the initial treatment with oral alkylating agents and steroids followed by radiotherapy for Ann Arbor stage II tumors and stage I of the palate lymphoma and the nasal cavity lymphoma extended to paranasal sinus.

• 한국세라믹학회지
• /
• 제33권8호
• /
• pp.889-894
• /
• 1996

This study was subjected to the stabilization of heavy metals using DSp cement. Heavy metal Cr and Pb ions were mixed with cement paste and hydration behavior and leaching property by heavy metal were exami-ned. It was found that, Cr ion accelerated the early hydration of the cement and has no accelerating effect in later hydration period. However Pb ion retarded the hydration of the cement for a early hydration periods. As a result of leaching test the quantity of leachant has a very low value and the influence of leached heavy metal effected on the environments is very weak.

• Journal of the Korean Data and Information Science Society
• /
• 제21권3호
• /
• pp.575-580
• /
• 2010

The Fisher information matrix plays a significant role in statistical inference in connection with estimation and properties of variance of estimators. In this paper, the parameter space of the t-distribution using its Fisher's matrix is de ned. The ${\alpha}$-scalar curvatures to parameter space are calculated.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.167.5-167
• /
• 2001

Single point attitude determination method provides an optimal attitude minimizing the Wahba loss function. However, for the insufficient number of measurement vectors, the conventional single point methods has no unique solution. Thus, we introduce the sequential method to and an optimal attitude minimizing the windowed loss function. In this paper, this function is de ned as the sum of square errors for all measurement vectors within the axed sliding window. For simple implementation, the proposed algorithm is rewritten as a recursive form. Moreover, the covariance matrix is derived and expressed as a recursive form. Finally, we apply this algorithm to the attitude determination system with three LOS measurement sensors.

• 동아시아식생활학회지
• /
• 제4권2호
• /
• pp.93-108
• /
• 1994

Kmich is the traditional vegetable fermented food. Like other traditional foods. it was not known when the kimchi was first processed. In the chinese historical litera tures, Salt and varlous vegetables were used in Korea from the time of kokuryo and shila Dynasty. Brined vegetable product might be therefore, processed at that time for the winter. But the product was highly salted and any seasonings were not added at all. Chinese cabbage and ned pepper were used for kimchi processing in 15-16 centureis. Vanious type of Kimchi simllar to present products were processed thereafter.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 제37회 하계학술대회 논문집 C
• /
• pp.1370-1371
• /
• 2006

An electrostatic discharge (ESD) protection device, so called, N-type extended drain silicon controlled rectifier (NEDSCR) device, was analyzed for high voltage I/O applications. A conventional NEDSCR device shows typical SCR-like characteristics with extremely low snapback holding voltage. This may cause latchup problem during normal operation. However, a modified NEDSCR device with proper junction / channel engineering demonstrates itself with both the excellent ESD protection performance and the high latchup immunity.

• 대한두경부종양학회지
• /
• 제16권1호
• /
• pp.58-63
• /
• 2000

Objectives: To demonstrate effective diagnostic method and proper management of recurrent thyroid cancer through to compare treatment and surveillance of $I^{131}$ scanning detected recurrence and clinically detected recurrence. Material and Methods: We retrospectively analyzed clinical information about 46 patients who has recurrent thyroid cancer of 298 patients who have been primarily operated due to thyroid cancer in PMC at the over 10 years between 1986 and 1995. We examine incidence of recurrence due to pathologic types, site of recurrence, disease free interval, detection method of recurrence, and also treatment and progression of recurrence. A patients in which the clinical examination was entirely negative and the $I^{131}$ scan demonstrated either a new area of $I^{131}$ uptake or an increased area of concentration, compared to the previous scan, was designated as a recurrence detected by $I^{131}$ scan only. Recurrences that were obviously by physical examination or chest x-ray, etc were considered clinically detected recurrence, regardless of the the results of the thyroid scan. Results: Mean of disease tree interval(DFI) is 36months. When mean DFI of $I^{131}$ scan detected recurrence is 28months, whereas mean DFI of clinically detected recurrence is 47months. In statiscal analysis, p-value is 0.043 as significantly. In progression of recurrent patient, NED is 28case, AWD is Sease, DOD is 13case. Among the 13case, scan detected recurrence is lease of 20 patients(5%), whereas clinically detected recurrence is l2case of 26 patient(46%). In statiscal analysis, p-value is 0.003 as significantly. Conclusion: Early detection of the recurrent thyroid cancer by $I^{131}$ scanning leads to good progress compare with detection by clinical examination. NED: No Evidence of Disease AWD : Alive With Disease DOD : Dead Of Disease DOC: Dead of Other Cause

• 한국융합학회논문지
• /
• 제11권6호
• /
• pp.1-6
• /
• 2020

본 연구에서는 컴퓨터 비전의 분야 중 하나인 문자 인식에 관한 연구를 수행했다. 대표적인 문자인식 기법 중 하나인 광학식 문자 판독 기법의 경우 일정한 규격과 서식에서 벗어나게 되면 인식률이 떨어진다는 한계점이 있다. 따라서 본 연구에서는 딥 러닝 기법을 적용해 이러한 문제점을 해결하고자 한다. 또한 기존의 문자 인식 연구의 경우 대부분 영어 및 숫자 인식에 국한되어 있다. 따라서 본 연구는 한글 인식을 위한 딥 러닝 기반 문자 인식 알고리즘을 제시한다. 알고리즘은 1-NED 평가 방법에서 0.841의 점수를 얻었으며, 이는 영어 인식 결과와 비슷한 수치이다. 본 연구를 통해 딥 러닝 기반 한글 인식 알고리즘의 성능을 확인할 수 있으며, 이를 통해 향후 연구방향에 대해 제시한다.

• 항공우주기술
• /
• 제12권2호
• /
• pp.30-39
• /
• 2013

본 논문에서는 정지궤도복합위성 기상탑재체의 보정 과정 및 영상 복사성능 품질 측정 방법에 대한 사항들을 기술하였다. MTF(Modulation Transfer Function), SNR(Signal-To-Noise Ratio), NEdT(Noise Equivalent Delta Temperature) 및 Dynamic Range 등 위성용 가시광선 및 적외선 센서의 복사성능과 관련된 품질 측정을 위한 주요 파라미터들에 초점을 맞추었다. 지상시험과정에서 측정한 이들 파라미터들의 기준 값들을 발사 후 궤도상 시험기간에 측정한 값들을 비교 검토하는 작업이 영상품질 측정의 핵심적인 과정이다. 천리안위성의 궤도상 시험기간 동안 이루어진 기상탑재체의 영상 품질 분석 과정의 사례들을 바탕으로 마련된 정지궤도 복합위성의 기상탑재체 영상의 복사성능 검정 방법에 대한 논의도 기술하였다.