• 한국정보통신학회논문지
• /
• 제8권3호
• /
• pp.721-726
• /
• 2004

영상의 이진화는 문자 인식, 영상 분석 등 다양한 영상 처리 분야의 전처리 과정으로 자주 적용되고 있다. 영상 이진화는 임계치의 설정에 따라 처리 성능이 좌우되며, 대부분의 기존 이진화 방법은 밝기 값의 히스토그램을 사용하여 평균 밝기 값이나 히스토그램의 골짜기를 임계치로 설정한다. 이와 같은 방법은 양봉의 특징을 보이지 않거나 특정 영상을 추출하려는 경우에는 적절한 임계치를 얻기 어렵다. 따라서 본 논문에서는 그레이 스케일 영상에서 밝기 값을 여러 구간으로 분할하여 각 구간의 밝기 평균값을 구하고, 두 개의 구간에 대해 평균값 사이의 거리를 각 구간에서 평균값과 양극과의 거리 비율로 나누어서 계산된 값을 두 개의 구간을 합친 새로운 구간의 임계치로 설정한다. 최종적으로 하나의 구간이 생성될 때까지 구간 통합과 임계값 계산을 반복함으로써 이진화 임계값을 산출한다. 제안된 이진화 방법의 성능을 평가하기 위하여 다양한 종류의 영상에 적용한 결과, 기존의 이진화 방법들보다 효율적인 것을 확인하였다.

• Journal of applied mathematics & informatics
• /
• 제26권5_6호
• /
• pp.1169-1183
• /
• 2008

In this paper, a method to find the weighted possibilistic variance and moments about the mean value of fuzzy numbers via applying a difuzzification using minimizer of the weighted distance between two fuzzy numbers is introduced. In this way, we obtain the nearest weighted point with respect to a fuzzy number, this main result is a new and interesting alternative justification to define of weighted mean of a fuzzy number. Considering this point and the weighted distance quantity, we introduce the weighted possibilistic mean (WPM) value and the weighted possibilistic variance(WPV) of fuzzy numbers. This paper shows that WPM is the nearest weighted point to fuzzy number and the WPV of fuzzy number is preserved more properties of variance in probability theory so that it can simply introduce the possibilistic moments about the mean of fuzzy numbers without problem. The moments of fuzzy numbers play an important role to estimate of parameters, skewness, kurtosis in many of fuzzy times series models.

• 한국안광학회지
• /
• 제11권3호
• /
• pp.181-185
• /
• 2006

각막곡률반경은 RGP 렌즈를 착용한 대학생(1980~1984년)을 대상으로 측정하였다. 대상자는 187명으로 여자(167), 남자(17) 그리고 원추각막환자(3)이었다. 각막곡률반경은 전체대상자의 평균치가 7.77mm이었고, 좌안과 우안의 측정치 역시 7.77mm이었다. 눈의 수평, 수직경선의 측정치는 7.88mm, 7.65mm이었다. 남자와 여자의 경우는 모두가 수평이 수직경선보다 동일하게 0.22mm 크게 조사되었다. 성별에 따른 곡률반경의 차이는 남자의 경우 평균치가 7.84mm으로 여자의 경우보다 0.08mm 정도 크게 측정되었고, 좌, 우안의 크기는 차이가 없었다. 원추각막환자는 곡률반경의 평균치가 6.86mm으로 정상인 조사대상자의 평균치보다 0.91mm 작게 조사되어 차이가 큰 것으로 나타났다.

• 대한전자공학회논문지SP
• /
• 제48권4호
• /
• pp.21-29
• /
• 2011

사전에 정의된 피부 색상 정보를 이용한 기존 피부 검출 방법들은 배경과 피부 영역을 분할하는 단계에서 사용되는 임계값을 실험을 통하여 주관적 관점에서 결정하였다. 또한 기존 방법들은 배경 환경과 조명 환경에 따라 각각 다른 임계값을 설정하였다. 이러한 기존 방법들은 반복 실험을 통하여 추정된 임계값에 따라 성능이 좌우되는 단점이 제시되었다. 제시된 기존 방법들의 단점을 극복하기 위하여 본 논문은 mean shift 알고리즘 기반의 히스토그램 근사화를 이용한 피부 영역 검출 방법을 제안한다. 제안하는 방법은 CbCr 컬러공간에서의 표준 피부색상과 유사도를 비교하여 생성된 입력 영상의 피부맵(skin-map)의 히스토그램에서 mean shift 방법을 이용하여 각각 밝기 영역별로 수렴하는 극대점을 능동적으로 찾아서 배경 영역과 피부영역으로 분할한다. 히스토그램은 픽셀의 명도값에 따라 누적되는 불연속 함수의 형태를 가지므로 베이지 곡선(Bezier curve) 기법을 이용하여 연속 가우시안 함수로 근사화된다. 따라서 제안하는 방법은 기존 방법에서처럼 수동적으로 임계값을 설정하는 방법을 사용하지 않고 mean shift 기법을 이용하여 능동적으로 영역 분할점인 극대점을 찾아서 피부 영역을 검출한다. 제안된 방법은 실험을 통하여 강인하고 효율적으로 피부 영역을 검출하였다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2003년도 춘계학술대회
• /
• pp.255-260
• /
• 2003

Most of the fatigue test were achieved in fully reversed condition that mean stress is zero. But, generally, mean stress can not be zero. This research was achieved the bending fatigue test that changes mean stress to spring steel and was studied relation with the mean stress and the fatigue life. The method to search effect about mean stress was extended S-N graph in two cases. One method was extended S-N graph using modified Miner's rule with considering damage. Another was extended using tendency of S-N graph. The exponential value(${\alpha}$) of mean stress-alternating stress equation is converged between the Goodman's and the Gerber's value even if fatigue life increases.

• 대한수의학회지
• /
• 제14권2호
• /
• pp.173-177
• /
• 1974

Thirty of healthy non-pregnant Holstein cows being raised in the vicinity of Daejeon, which had been delivered of calves 4 to 6 times since they were imported to Korea, were selected with the purpose of serum chemical analysis. Total protein, albumin, globulin, total cholesterol, urea nitrogen and glucose and A/G ratio were investigated. The results of present investigation were summarized as follows: 1. Ranges and mean values (with standard error) of total protein, albumin and globulin were 6.87~8.58g/100ml, $7.75{\pm}0.60g/100ml$; 2.6~3.5g/100ml(37.1~45.0%), $3.26{\pm}0.27g/100ml$(40.1%); and 3.78~5.48g/100ml(55.1~64.6%), $4.51{\pm}0.43g/100ml$(58.1%), respectively. 2. Range and mean value (with standard error) of A/G ratio were 0.52~0.86 and 0.73~0.093. 3. Range and mean value (with standard error) of glucose were 35~50mg/100ml and $45.2{\pm}9.1mg/100ml$. 4. Range and mean value(with standard error) of total cholesterol were 122~228mg/100ml and $194.6{\pm}11.1mg/100ml$. 5. Serum urea nitrogen showed that the range of 10.5~15.1mg/100ml and the mean value(with standard error) of $12.7{\pm}1.5mg/100ml$.

• Journal of the Korean Data and Information Science Society
• /
• 제26권2호
• /
• pp.535-545
• /
• 2015

In a various range of applications including hydrology, the type-I extreme value distribution has been extensively used as a probabilistic model for analyzing extreme events. In this paper, we introduce methods for estimating the scale parameter of the type-I extreme value distribution. A simulation study is performed to compare the estimators in terms of mean-squared error and bias, and the obtained results are provided.

• Journal of Nutrition and Health
• /
• 제28권8호
• /
• pp.749-758
• /
• 1995

In this study, the food intake, feces and urine of 16 primary school age boys and girls were collected and intake and excretion of sodium and calcium were measured. The boys and girls were 8-12 years old and measurement continued for four weeks during which they maintained their normal living pattern and body weight. Each boy's and girl's daily intake and excretion of sodium and calcium were measured and apparent digestibility and balance were also studied. The results were as follows. 1) Mean daily intake of sodium was 8.52$\pm$0.38g for the boys and 7.31$\pm$0.44g for the girls. The mean value in males was significantly higher than that in females(p<0.05). Mean daily in take of calcium was 411.0$\pm$16.0mg for the boys and 356.5$\pm$15.4mg for the girls. The mean value in males was significantly higher than that in females(p<0.01). 2) Mean daily fecal loss and apparent digestibility of sodium was 0.32$\pm$0.04g and 96% for the boys and 0.52$\pm$0.07g and 93% for the girls. The fecal loss mean value in males was significantly lower than that in females(p<0.05). Mean daily fecal loss and apparent digestibility of calcium was 299.8$\pm$8.3mg and 29% for the boys and 194.1$\pm$14.3mg and 46% for the girls. The fecal loss mean value in males was significantly higer than that in females(p<0.01). 3) Mean daily urinary loss of sodium was 6.55$\pm$0.50g and showed the positive balance of 1.65g for the boys and 5.67$\pm$0.20g and showed the positive balance of 1.12g for the girls. The urinary loss mean values of the two groups were not significantly different. Mean daily urinary loss of calcium was 42.8$\pm$5.1mg and showed the positive balance of 79.4mg for the boys and 25.0$\pm$1.64mg and showed the positive balance of 137.4mg for the girls. The urinary loss mean value in males was significantly higer than that in females(p<0.01).

• Journal of the Korean Data and Information Science Society
• /
• 제17권2호
• /
• pp.421-436
• /
• 2006

Huang proposed a grey-based nearest neighbor approach to predict accurately missing attribute value in 2004. Our study proposes which way to decide the number of nearest neighbors using not only the deng's grey relational grade but also the wen's grey relational grade. Besides, our study uses not an arithmetic(unweighted) mean but a weighted one. Also, GRG is used by a weighted value when we impute missing values. There are four different methods - DU, DW, WU, WW. The performance of WW(Wen's GRG & weighted mean) method is the best of any other methods. It had been proven by Huang that his method was much better than mean imputation method and multiple imputation method. The performance of our study is far superior to that of Huang.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2006년도 PROCEEDINGS OF JOINT CONFERENCEOF KDISS AND KDAS
• /
• pp.55-62
• /
• 2006

Huang proposed a grey-based nearest neighbor approach to predict accurately missing attribute value in 2004. Our study proposes which way to decide the number of nearest neighbors using not only the dong's grey relational grade but also the wen's grey relational grade. Besides, our study uses not an arithmetic(unweighted) mean but a weighted one. Also, GRG is used by a weighted value when we impute a missing values. There are four different methods - DU, DW, WU, WW. The performance of WW(wen's GRG & weighted mean) method is the best of my other methods. It had been proven by Huang that his method was much better than mean imputation method and multiple imputation method. The performance of our study is far superior to that of Huang.