• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.4
• /
• pp.332-339
• /
• 2014

Face classification has wide applications in security and surveillance. However, this technique presents various challenges caused by pose, illumination, and expression changes. Face recognition with long-distance images involves additional challenges, owing to focusing problems and motion blurring. Multiple frames under varying spatial or temporal settings can acquire additional information, which can be used to achieve improved classification performance. This study investigates the effectiveness of multi-frame decision-level fusion with photon-counting linear discriminant analysis. Multiple frames generate multiple scores for each class. The fusion process comprises three stages: score normalization, score validation, and score combination. Candidate scores are selected during the score validation process, after the scores are normalized. The score validation process removes bad scores that can degrade the final output. The selected candidate scores are combined using one of the following fusion rules: maximum, averaging, and majority voting. Degraded facial images are employed to demonstrate the robustness of multi-frame decision-level fusion in harsh environments. Out-of-focus and motion blurring point-spread functions are applied to the test images, to simulate long-distance acquisition. Experimental results with three facial data sets indicate the efficiency of the proposed decision-level fusion scheme.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.6
• /
• pp.679-686
• /
• 2020

In this thesis, an enhanced method for the feature extraction of vocal source signals and score combination using an MCE-Based weight estimation of the score of multiple feature vectors are proposed for the performance improvement of speaker recognition systems. The proposed feature vector is composed of perceptual linear predictive cepstral coefficients, skewness, and kurtosis extracted with lowpass filtered glottal flow signals to eliminate the flat spectrum region, which is a meaningless information section. The proposed feature was used to improve the conventional speaker recognition system utilizing the mel-frequency cepstral coefficients and the perceptual linear predictive cepstral coefficients extracted with the speech signals and Gaussian mixture models. In addition, to increase the reliability of the estimated scores, instead of estimating the weight using the probability distribution of the convectional score, the scores evaluated by the conventional vocal tract, and the proposed feature are fused by the MCE-Based score combination method to find the optimal speaker. The experimental results showed that the proposed feature vectors contained valid information to recognize the speaker. In addition, when speaker recognition is performed by combining the MCE-based multiple feature parameter scores, the recognition system outperformed the conventional one, particularly in low Gaussian mixture cases.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.945-956
• /
• 2021

In this study, we considered the algorithm of Legendre and Legendre (2012), which derives canonical correspondence analysis from weighted principal component analysis. And, it was proved that the canonical correspondence analysis based on the weighted principal component analysis is exactly the same as Ter Braak's (1986) canonical correspondence analysis based on the Gaussian response model. Ter Braak (1986)'s canonical correspondence analysis derived from a Gaussian response curve that can explain the abundance of species in ecology well uses the basic assumption of the species packing model and then conducts generalized linear model and canonical correlation analysis. It is derived by way of binding. However, the algorithm of Legendre and Legendre (2012) is calculated in a method quite similar to Benzecri's correspondence analysis without such assumptions. Therefore, if canonical correspondence analysis based on weighted principal component analysis is used, it is possible to have some flexibility in using the results. In conclusion, this study shows that the two methods starting from different models have the same site scores, species scores, and species-environment correlations.

• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.237-248
• /
• 2009

The probability of default on the credit evaluation study is represented as a linear combination of two distributions of default and non-default, and the distribution of the probability of default are generally known in most cases. Except the well-known Kolmogorov-Smirnov statistic for testing the identity of two distribution, Kuiper, Cramer-Von Mises, Anderson-Darling, and Watson test statistics are introduced in this work. Under the assumption that the population distribution is known, modified Cramer-Von Mises, Anderson-Darling, and Watson statistics are proposed. Based on score data generated from various probability density functions of the probability of default, the modified test statistics are discussed and compared.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.4
• /
• pp.31-40
• /
• 2022

This study assumed two key parameters that describe smoothness. Previously, the number of studies that defined smoothness with the concept of curvature has been limited. The parameters were the two physical characteristics of curvature: the number of waves and the amplitude of them. The purpose of the study was to observe which of the two parameters influences the perception of smoothness and to examine the relationship between the sense of aesthetic beauty and smoothness. The visual stimuli used in the study were transformed three-dimensional spheres, based on the combination of the three levels of the two parameters of nine distinct conditions. We analyzed the three responses that measured the preference of each visual stimuli, the familiarity score, and the smoothness evaluation score, each with the linear mixed model whose fixed effects were the two parameters and random effects were the participants' individual differences. Nearly the eighty percent of the variance of the smoothness evaluation score was explained by the linear model with the two key parameters and their interaction. The physical characteristics of a viewed object were far more significant than individual differences such as personality factors and the manner of art appreciation. In conclusion, the study examined the perception of smoothness based on the change of the physical characteristics of a shape. The study further recognized the relationship between smoothness and the aesthetic preference. No significant influence of the participants' individual difference such as gender, the degree in fine arts, personality factors, and the manner of art appreciation was observed. The amplitude of waves, rather than the number of them, was far more significant to the perception of smoothness.

• Clinics in Shoulder and Elbow
• /
• v.7 no.1
• /
• pp.23-29
• /
• 2004

It has been reported that rotator cuff tear have good response to arthroscopic or open repair even if the range being so wide. However, the majority of this literature regarding the diagnosis and treatment of tear focused on lesion of the supraspinatus and infraspinatus tendons. But involvement of the subscapularis tendon with rotator cuff tear should be thought to be less common and poorer to open operative repair. Furthermore, some europian author have stated that the rotator cuff tear including the subscapularis tendon are sufficiently distinct in their clinical presentation and prognosis as to merit separate consideration of their diagnosis and treatment. The purpose of this study is to evaluate result of arthroscopic or open repair in patient with rotator cuff tear that include the subscapularis tendon. Of the 128 rotator cuff repairs performed from 1998 through 2003, 12 had a tear that include the subscapularis tendon in combination with the supraspinatus (8 cases) and infraspinatus (4 cases). Mean duration of symptoms before surgical treatment was 6 months (range 3 to 12 months). All 12 patient demonstrated a positive lift off sign. Shoulder function was assessed using the Constant- Murley score, which ranges from 30 to 58. Pain was assessed using a linear visual analogue scale range from 0 to 10. Postoperative Constant score range from 40 to 64 (average 47.8). Pain score improved from 5.5 to 8.5, but there are postoperative pain improvement on nothing in 5 patient. The overall result for 12 patient were satisfy in 2, fair in 5 and dissatisfy in 5: Therefore satisfactory result were noted only in 16 % of this overall group. In conclusion, we have failed to make good result in patient with rotator cuff tear that included the subscapularis tendon. At the result, outcome after surgical repair of this type of rotator tear is comparatively inferior to the result of operative repair of rotator cuff not involved the subscapularis tendon.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1119-1127
• /
• 2009

In PGA golf, total prize money and average score are good response variable related to golf skills such as driving distance, green in regulation and putts per green in regulation. But it's not easy to predict the winner of coming tournament. Thus I applied Neural Networks which has pretty good advantages for non-linear complex modeling to binary data. In neural network architectures, I applied NRBF and MLP architecture model for binary data which represent who had a win or not.

• Journal of Environmental Impact Assessment
• /
• v.21 no.5
• /
• pp.813-821
• /
• 2012

Environmental and public health concern for the emission of air pollutants from burn-up process in waste incineration plants located in the vicinity of living environment was increased during the past decade. The purpose of this study was to suggest of the simple and rapid method of priority setting model for the decision of full-scale public health assessment. This method was consists of total 5-step. Step 1 was "secure the satellite map" and we can use the satellite map which serves from the website such as NAVER Co. Step 2 was "drawing mesh on the map" for catch the point of occupation of environmental sensitivity facilities, and step 3 was "identification and sorting of the facilities", Step 4 was "setting of weight" using the "weighted linear combination (WLC) method". Finally, all facility was sorted by score. As a result, we can set a priority of 145 facilities based on 177 facilities which managed in local government. Facilities in Seoul metropolitan area was high rank in priority list. On the other side, Facilities located at the country or rural area was low rank because of low occupation of the house and the environmental sensitivity facilities such as kindergarten, elementary school, and hospital. In this study, we suggested simple and rapid method that using for screening procedure of public health assessment.

• Journal of Animal Science and Technology
• /
• v.52 no.3
• /
• pp.167-174
• /
• 2010

The objectives of this study were to quantify the combination values of the principal components and factors calculated using body measurements of Hanwoo (Korean Cattle) and estimate their heritabilities. The technique of multivariate analysis was used to reduce a large number of variables to a smaller number of new variables and characterize cattle according to body shape. The analyses were performed using 1,979 cattle at 12 months of age and 936 cattle at 24 months of age. The data for the analyses was obtained from progeny tests performed on Korean Cattle for 6 years from 2003 to 2008. The phenotypic correlations among these traits were estimated to range from 0.32 to 0.90 at 12 months of age and from 0.21 to 0.82 at 24 months of age. The first principal components (PC1s) indicated a weighed average of overall body measurements, accounting for 99.91% of the total variation for both periods of test. The two first PCs had positive coefficients for all body measurements. The major sources of PC, such as chest girth (CG), body length (BL), rump height (RH), and wither height (WH) were similar for both test periods. The heritabilities for PC1, the first factor score (FS1), and the second factor score (FS2) were estimated by multivariate REML method. The estimated heritabilities for PC1, FS1, and FS2 were 0.33, 0.38, and 0.40, respectively, at 12 months of age and 0.26, 0.76, and 0.58 at 24 months of age. Further studies are needed to determine whether the heritabilities of FS1 and FS2 at 24 months of age were overestimated.

• Journal of KIISE
• /
• v.42 no.7
• /
• pp.901-909
• /
• 2015

In this paper, to build a parallel corpus between Korean and English in Wikipedia. We proposed a method to find similar sentences based on language resources and topic modeling. We first applied language resources(Wiki-dictionary, numbers, and online dictionary in Daum) to match word sequentially. We construct the Wiki-dictionary using titles in Wikipedia. In order to take advantages of the Wikipedia, we used translation probability in the Wiki-dictionary for word matching. In addition, we improved the accuracy of sentence similarity measuring method by using word distribution based on topic modeling. In the experiment, a previous study showed 48.4% of F1-score with only language resources based on linear combination and 51.6% with the topic modeling considering entire word distributions additionally. However, our proposed methods with sequential matching added translation probability to language resources and achieved 9.9% (58.3%) better result than the previous study. When using the proposed sequential matching method of language resources and topic modeling after considering important word distributions, the proposed system achieved 7.5%(59.1%) better than the previous study.