• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.305-310
• /
• 2010

In this paper, we propose a heuristic algorithm for motion correspondence based on a point series similarity. A point series is a sequence of points which are sorted in the ascending order of their x-coordinate values. The proposed algorithm clusters the points of a previous frame based on their local adjacency. For each group, we construct several potential point series by permuting the points in it, each of which is compared to the point series of the following frame in order to match the set of points through their similarity based on a proximity constraint. The longest common subsequence between two point series is used as global information to resolve the local ambiguity. Experimental results show an accuracy of more than 90% on two image sequences from the PETS 2009 and the CAVIAR data sets.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.3
• /
• pp.282-291
• /
• 2004

There has been many researches in TopicMap and RDF which are approach to handle data efficiently with metadata. However, No researches has been performed to service and implement except for presentation and description. In this paper, We suggest the caching mechanism to support an efficient access of knowledgemap and practical knowledgemap service with implementation of TopicMap system. First, We propose a method to navigate Knowledgemap efficiently that includes advantage of former methods. Then, To transmit TopicMap efficiently, We suggest caching mechanism for knowledgemap. This method is that user will be able to navigate knowledgemap efficiently in the viewpoint of human, not application. Therefor the mechanism doesn't cash topics by logical or physical locality but clustering by information and characteristic value of TopicMap. Lastly, we suggest replace mechanism by using graph structure of TopicMap for efficiency of transmission.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.282-289
• /
• 2008

Although there were some technological developments in improving the collaborative filtering, they have yet to fully reflect the actual relation of the items. In this paper, we propose the recommendation system using associative web document classification by word frequency and ${\alpha}$-cut to address the short comings of the collaborative filtering. The proposed method extracts words from web documents through the morpheme analysis and accumulates the weight of term frequency. It makes associative rules and applies the weight of term frequency to its confidence by using Apriori algorithm. And it calculates the similarity among the words using the hypergraph partition. Lastly, it classifies related web document by using ${\alpha}$-cut and calculates similarity by using adjusted cosine similarity. The results show that the proposed method significantly outperforms the existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.11
• /
• pp.2089-2094
• /
• 2017

Altmetrics is the measurement index and quantitative data to complement the traditional indicators based on the citation. Altmetrics research has acquired greater importance in the past few years, partly due to the complement to the traditional bibliometrics. This paper aims to reveal the research status and trends in altmetrics research. A total of 187 articles from 2005 to 2017 are obtained and analyzed, illustrating a steady rise (S-mode) in altmetrics research since 2005. Using social network analysis and co-word analysis, the author cooperation network and keyword co-occurrence network are developed. The core scientists and eight international research groups are discovered, reflecting that researchers in this field have a low degree of cooperation. Four topics of altmetrics research are discovered by hierarchical clustering. The results can be useful for the advanced research of altmetrics.

• The KIPS Transactions:PartD
• /
• v.11D no.7 s.96
• /
• pp.1357-1366
• /
• 2004

As the importance or XML is increasing to manage information and exchange data efficiently in the web, there are on going works about structural integration and retrieval. The XML. document with the defined structure can retrieve the structure through the DTD or XML schema, but the existing method can't apply to XML. documents which haven't the structure information. Therefore. in this paper we propose a new clus-tering technique at a basic research which make it possible to retrieve structure fast about the XML documents that haven't the structure information. We first estract the feature of frequent structure from each XML document. And we cluster based on the similar structure by con-sidering the frequent structure as representative structure of the XML document, which makes it possible to retrieve the XML document raster than dealing with the whole documents that have different structure. And also we perform the structure retrieval about XML documents based on the clusters which is the group of similar structure. Moreover, we show efficiency of proposed method to describe how to apply the structure retrieval as well as to display the example of application result.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.6
• /
• pp.321-328
• /
• 2017

We have implemented a technique to correct the brightness, saturation, and contrast of an image according to the degree of light, and further compensate the backlight. Backlight compensation can be done automatically or manually. For manual backlight compensation, we have to select the region of interest (ROI). ROI can be selected by connecting the outline of the desired object. We make users select the region delicately with the new magnetic lasso tool. The previous lasso tool has a disadvantage that the start point and the end point must be connected. However, the proposed lasso tool has the advantage of selecting the region of interest without connecting the start point and the end point. We can automatically obtain various results of backlight compensation by adjusting the number of k-means clusters for texture extraction and the threshold value for binarization.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.36 no.5
• /
• pp.915-926
• /
• 2016

The purpose of this study is to classify debris flow hazard zones in urbanized areas using multivariate statistical analyses and to suggest customized management strategies to each areal type. Using field survey data set in Seoul, 49 sample debris hazard zones are selected. Clustering and discriminant analyses show that debris flow hazard zones are classified into two types. Surrounding land use and land slope are major factors influencing to the categorization. The results suggest that, by considering the characteristics of each areal type, more customized management strategies for debris flow hazard are necessary. Particularly, in addition to traditional structural measures, non-structural measures including land use and development control for downstream built-up areas should be emphasized in urbanized areas to mitigate human and property damages from debris flow hazard more fundamentally.

• The Journal of Society for e-Business Studies
• /
• v.17 no.4
• /
• pp.39-56
• /
• 2012

Technological advancements in information technology including RFID and mobile technologies have made it feasible to track the customers travel path in a store. The customer travel paths provide valuable implications to understanding the customer behaviors in a store. In our research, we develop a shopping path analysis system to track and analyze the customer travel path. The proposed system consists of RFID systems for collecting the customer paths and analysis system. The analysis system conducts clustering for identifying the distinctive shopping patterns, and analyzes the profile of a grocery, such as congestion rate, visiting rate, and staying time, etc. We show the applicability of our proposed system using the actual data obtained at a grocery in Seoul as a case study.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.1
• /
• pp.225-231
• /
• 2013

In this study, we propose the design of optimized pRBFNNs-based night vision face recognition system using PCA algorithm. It is difficalt to obtain images using CCD camera due to low brightness under surround condition without lighting. The quality of the images distorted by low illuminance is improved by using night vision camera and histogram equalization. Ada-Boost algorithm also is used for the detection of face image between face and non-face image area. The dimension of the obtained image data is reduced to low dimension using PCA method. Also we introduce the pRBFNNs as recognition module. The proposed pRBFNNs consists of three functional modules such as the condition part, the conclusion part, and the inference part. In the condition part of fuzzy rules, input space is partitioned by using Fuzzy C-Means clustering. In the conclusion part of rules, the connection weights of pRBFNNs is represented as three kinds of polynomials such as linear, quadratic, and modified quadratic. The essential design parameters of the networks are optimized by means of Differential Evolution.

• Korean Journal of Agricultural Science
• /
• v.46 no.4
• /
• pp.971-980
• /
• 2019

We profiled the health-promoting bioactive components in nine types of soybean seeds with different seed coat colors (yellow, green, brown, and black) and investigated the effects of different extraction solvents (methanol, ethanol, and water) on their antioxidant activities. The carotenoid and anthocyanin compositions varied greatly by seed color, and the phenolic acids, total phenol, and total flavonoid contents differed by genotype. The carotenoid content was relatively higher in soybean seeds with green and black seed coats than in those with a yellow seed coat while lutein was the most plentiful. The anthocyanin content was considerably higher in the soybean seed with the black seed coat. The results of the DPPH assay showed strong antioxidative activities in the methanol- and water-extracts compared to the ethanol-extract, irrespective of the seed coat colors. Moreover, the soybean seeds with the black seed coat exhibited the highest antioxidant activity among the samples, regardless of the extraction solvent used. Eighteen bioactive compounds were subjected to data-mining processes including principal component analysis and hierarchical clustering analysis. Multivariate analyses showed that brown and black seeds were distinct from the yellow and green seeds in terms of the levels of carotenoids and anthocyanins, respectively. These results help our understanding of the compositional differences in the bioactive components among soybean seeds of various colors, providing valuable information for future breeding programs that seek to enhance the levels of compounds with health benefits.