• Journal of the Korea Society of Computer and Information
• /
• v.14 no.11
• /
• pp.73-79
• /
• 2009

In this paper, a method is proposed that enhances the performance of the facial expression recognition using template matching of Local Binary Pattern(LBP) histogram. In this method, the face image is segmented into blocks, and the LBP histogram is constructed to be used as the feature of the block. Block dissimilarity is calculated between a block of input image and the corresponding block of the model image. Image dissimilarity is defined as the weighted sum of the block dissimilarities. In conventional methods, the block weights are assigned by intuition. In this paper a new method is proposed that optimizes the weights from training samples. An experiment shows the recognition rate is enhanced by the proposed method.

• The KIPS Transactions:PartB
• /
• v.13B no.7 s.110
• /
• pp.643-652
• /
• 2006

Decision trees are mainly used for the classification and prediction in data mining. The distribution of spatial data and relationships with their neighborhoods are very important when conducting classification for spatial data mining in the real world. Spatial decision trees in previous works have been designed for reflecting spatial data characteristic by rating Euclidean distance. But it only explains the distance of objects in spatial dimension so that it is hard to represent the distribution of spatial data and their relationships. This paper proposes a decision tree based on spatial entropy that represents the distribution of spatial data with the dispersion and dissimilarity. The dispersion presents the distribution of spatial objects within the belonged class. And dissimilarity indicates the distribution and its relationship with other classes. The rate of dispersion by dissimilarity presents that how related spatial distribution and classified data with non-spatial attributes we. Our experiment evaluates accuracy and building time of a decision tree as compared to previous methods. We achieve an improvement in performance by about 18%, 11%, respectively.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1998.10a
• /
• pp.597-602
• /
• 1998

The composite systems, consisting of R/C Columns-Steel Beams, are reasonable structures because of their constructional and economical advantages, workability and so on. But, it is difficult to apply the composite systems to actual design due to material dissimilarity and complicate stress flow in the connection. This study aims to propose the hybrid beam-column connection system with structural tee and through experimental research make clear the shear and moment resistance capacity and stress transfer mechanism.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.718-721
• /
• 2008

Novel computational integral imaging technique with enhanced depth sensitivity is proposed. For each lateral position at a given depth plane, the dissimilarity between corresponding pixels of the elemental images is measured and used as a suppressing factor for that position. Experimental and simulation results show that reconstructed depth image on the incorrect depth plane is effectively suppressed.

• Journal of Information Display
• /
• v.11 no.2
• /
• pp.68-75
• /
• 2010

The relation between light sources and screen cells is considered part of the theoretical model of an autostereoscopic 3D display. The geometry of the image and observer regions is presented, including the cases of single and multiple regions. The characteristic function is introduced. Formulas for the geometric parameters are obtained, including areas and angles. Special attention is drawn to the screen location. The method of transforming the formulas between regions is stated. For multiple regions, geometric dissimilarity was found. This allows the model to be applied in finding the geometric characteristics of multiview and integral-imaging 3D displays.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1353-1359
• /
• 2010

Basic objective in cluster analysis is to discover natural groupings of items. In general, clustering is conducted based on some similarity (or dissimilarity) matrix or the original input data. Various measures of similarities between objects are developed. In this paper, we consider a clustering of huge categorical real data set which shows the aspects of time-location-activity of Korean people. Some useful similarity measure for the data set, are developed and adopted for the categorical variables. Hierarchical and nonhierarchical clustering method are applied for the considered data set which is huge and consists of many categorical variables.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2243-2245
• /
• 1998

In this paper, we treat image segmentation as a graph partitioning problem. and use the normalized cut for segmenting the graph. The normalized cut criterion measures both the total dissimilarity between the different graphs as well as the total similarity within the groups. The minimization of this criterion can formulated as a generalized eigenvalues problem. We have applied this approach to segment static image. This criterion can be shown to be computed efficiently by a generalized eigenvalues problem

• Genomics & Informatics
• /
• v.14 no.4
• /
• pp.181-186
• /
• 2016

Glucose tolerance tests have been devised to determine the speed of blood glucose clearance. Diabetes is often tested with the standard oral glucose tolerance test (OGTT), along with fasting glucose level. However, no single test may be sufficient for the diagnosis, and the World Health Organization (WHO)/International Diabetes Federation (IDF) has suggested composite criteria. Accordingly, a single multi-class trait was constructed with three of the fasting phenotypes and 1- and 2-hour OGTT phenotypes from the Korean Association Resource (KARE) project, and the genetic association was investigated. All of the 18 possible combinations made out of the 3 sets of classification for the individual phenotypes were taken into our analysis. These were possible due to a method that was recently developed by us for estimating genomic associations using a generalized index of dissimilarity. Eight single-nucleotide polymorphisms (SNPs) that were found to have the strongest main effect are reported with the corresponding genes. Four of them conform to previous reports, located in the CDKAL1 gene, while the other 4 SNPs are new findings. Two-order interacting SNP pairs of are also presented. One pair (rs2328549 and rs6486740) has a prominent association, where the two single-nucleotide polymorphism locations are CDKAL1 and GLT1D1. The latter has not been found to have a strong main effect. New findings may result from the proper construction and analysis of a composite trait.

• Journal of Plant Biotechnology
• /
• v.31 no.1
• /
• pp.19-23
• /
• 2004

This study was conducted to observe the morphological characteristics and to analyse genetic relationship using RAPD markers in 9 species of Iris. The period of flowering was shown very broad among 9 species of Iris, from April 22 to June 9. The flower stalk was fluctuated from 9.0cm (I. rossii) to 104.0cm (I. ensata). I. ensata has the largest flower size and followed by I. laevigata, I. tectorum, I. sanguinea, I. pseudacorus, I. lactea, I. rossii, I. minutoaurea, I. odaesanensis. Among the 108 amplified bands by PCR with 8 random primers, 107 showed polymorphism and only one showed monomorphism among 9 species of Iris. The average dissimilarity coefficient among 9 species of Iris was 0.252. The range of dissimilarity coefficient was shown as from 0.095 to 0.609. Nine species were divided into three groups. The first group was I. tectorum, and the second group was I. lactea, I. laevigata, I. pseudacorus, I. odaesanensis. I. minutoaurea, I. rossii, I. sanguinea, I. ensata were belong to group three.

• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.1
• /
• pp.43-49
• /
• 2015

The aim of this study was to model unusual association with declarative knowledge by positive affect using ACT-R cognitive architecture. Existing research related with cognitive modeling tends to pay a lot of attention to strong and negative cognitive moderator. Mild positive affect, however, has far-reaching effects on problem solving and decision making. Typically, subjects with positive affect were more likely to respond to unusual associates in a word association task than subjects with neutral affect. In this study, a cognitive model using ACT-R cognitive architecture was developed to show the effect of positive affect on the cognitive organization related with memory. First, we organized the memory structure of stimulus word 'palm' based on published results in a word association task. Then, we decreased an ACT-R parameter that reflects the amount of weighting given to the dissimilarity between the stimulus word and the associate word to represent reorganized memory structure of the model by positive affect. As a result, no significant associate probability difference between model prediction and existing empirical data was found. The ACT-R cognitive architecture could be used to model the effect of positive affect on the unusual association by decreasing (manipulating) the weight of the dissimilarity. This study is useful in conducting model-based evaluation of the effects of positive affect in complex tasks involving memory, such as creative problem solving.