• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.166-179
• /
• 2021

Recently, with the development of technology, computer vision research based on the human visual system has been actively conducted. Saliency maps have been used to highlight areas that are visually interesting within the image, but they can suffer from low performance due to external factors, such as an indistinct background or light source. In this study, existing color, brightness, and contrast feature maps are subjected to multiple shape and orientation filters and then connected to a fully connected layer to determine pixel intensities within the image based on location-based weights. The proposed method demonstrates better performance in separating the background from the area of interest in terms of color and brightness in the presence of external elements and noise. Location-based weight normalization is also effective in removing pixels with high intensity that are outside of the image or in non-interest regions. Our proposed method also demonstrates that multi-filter normalization can be processed faster using parallel processing.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.29 no.2
• /
• pp.113-121
• /
• 2011

In this paper, we proposed the method automatically to match areal feature based on similarity using spatial information. For this, we extracted candidate matching pairs intersected between two different spatial datasets, and then measured a shape similarity, which is calculated by an weight sum method of each matching criterion automatically derived from CRITIC method. In this time, matching pairs were selected when similarity is more than a threshold determined by outliers detection of adjusted boxplot from training data. After applying this method to two distinct spatial datasets: a digital topographic map and street-name address base map, we conformed that buildings were matched, that shape is similar and a large area is overlaid in visual evaluation, and F-Measure is highly 0.932 in statistical evaluation.

• Korean Journal of Cognitive Science
• /
• v.12 no.1_2
• /
• pp.35-42
• /
• 2001

In this paper. a document is represented as a weighted graph called a text relationship map. In the graph. a node represents a vector of nouns in a sentence, an edge completely connects other nodes. and a weight on the edge is a value of the similarity between two nodes. The similarity is based on the word overlap between the corresponding nodes. The importance of a node. called an aggregate similarity in this paper. is defined as the sum of weights on the links connecting it to other nodes on the map. In this paper. we present a Korean text summarization system using the aggregate similarity. To evaluate our system, we used two test collection, one collection (PAPER-InCon) consists of 100 papers in the field of computer science: the other collection (NEWS) is composed of 105 articles in the newspapers and had built by KOROlC. Under the compression rate of 20%. we achieved the recall of 46.6% (PAPER-InCon) and 30.5% (NEWS) and the precision of 76.9% (PAPER-InCon) and 42.3% (NEWS).

• Journal of Digital Convergence
• /
• v.15 no.6
• /
• pp.229-236
• /
• 2017

An efficient frequent pattern algorithm is essential for mining association rules as well as many other mining tasks for convergence with its application spread over a very broad spectrum. Models for mining pattern have been proposed using a FP-tree for storing compressed information about frequent patterns. In this paper, we propose a centroid frequent pattern growth algorithm which we called "CAWFP-Growth" that enhances he FP-Growth algorithm by making the center of weights and frequencies for the itemsets. Because the conventional constraint of maximum weighted support is not necessary to maintain the downward closure property, it is more likely to reduce the search time and the information loss of the frequent patterns. The experimental results show that the proposed algorithm achieves better performance than other algorithms without scarifying the accuracy and increasing the processing time via the centroid of the items. The MapReduce framework model is provided to handle large amounts of data via a pseudo-distributed computing environment. In addition, the modeling of the proposed algorithm is required in the fully distributed mode.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.2
• /
• pp.101-112
• /
• 2003

This paper designs and implements the management system that can calibrate the contamination vulnerability of the ground water, using an object oriented data model. Geographic-objects are specified by features extracted from an applicable geographic domain, and geographic-fields are defined by chemical factors extracted from each driven water. To show the topological relationships among the geographic-objects and the geographic-fields, this paper attach the weight and the ratio of the drastic model to chemical factors represented on the land use digital map and the ground water digital map. The geographic feature class, administrative boundary class, land use class and driven water class consist of a class composition hierarchy for evaluating the convenient contamination vulnerability calibration with spatial relationships among the well objects. Therefore, this management system for evaluating the contamination vulnerability can also contribute to the application of other natural environments.

• Journal of Microbiology and Biotechnology
• /
• v.16 no.2
• /
• pp.256-263
• /
• 2006

In the model legume Medicago truncatula, two mutants, sickle and sunn, exhibit morphologically and genetically distinct hypernodulation phenotypes. However, efforts to isolate the single recessive and single semidominant genes for sickle and sunn, respectively, by map-based cloning have so far been unsuccessful, partly due to the absence of clones that enable walks from linked marker positions. To help resolve these difficulties, a new bacterial artificial chromosome (BAC) library was constructed using BamHI-digested genomic fragments. A total of 23,808 clones were collected from ligation mixtures prepared with double-size-selected high-molecular-weight DNA. The average insert size was 116 kb based on an analysis of 88 randomly selected clones using NotI digestion and pulsed-field gel electrophoresis. About 18.5% of the library clones lacked inserts. The frequency of the BAC clones carrying chloroplast or mitochondrial DNA was 0.98% and 0.03%, respectively. The library represented approximately 4.9 haploid M. truncatula genomes. Hybridization of the BAC clone filters with a $C_{0}t-l$ DNA probe revealed that approximately 37% of the clones likely carried repetitive sequence-enriched DNA. An ordered array of pooled BAC DNA was screened by polymerase chain reactions using 13 sequence-characterized molecular markers that belonged to the eight linkage groups. Except for two markers, one to five positive BAC clones were obtained per marker. Accordingly, the sickle- and sunn-linked BAC clones identified herein will be useful for the isolation of these biotechnologically important genes. The new library will also provide clones that fill the gaps between preexisting BAC contigs, facilitating the physical mapping and genome sequencing of M. truncatula.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.5
• /
• pp.178-188
• /
• 2020

This paper proposes a lane recognition CNN network using split-attention network as a backbone to extract feature. Split-attention is a method of assigning weight to each channel of a feature map in the CNN feature extraction process; it can reliably extract the features of an image during the rapidly changing driving environment of a vehicle. The proposed deep neural networks in this paper were trained and evaluated using the Tusimple data set. The change in performance according to the number of layers of the backbone network was compared and analyzed. A result comparable to the latest research was obtained with an accuracy of up to 96.26, and FN showed the best result. Therefore, even in the driving environment of an actual vehicle, stable lane recognition is possible without misrecognition using the model proposed in this study.

• 전자공학회논문지 IE
• /
• v.45 no.4
• /
• pp.47-53
• /
• 2008

A new intra-field deinterlacing algorithm with edge direction vector (EDV) in the image block is introduced. This proposed filter is suitable to the region with high motion or scene change. We first introduce an EDV, which is computed by Sobel mask used edge map, so that filer resolution of the edge direction can be acquired. The proposed EDV oriented deinterlacing system operates by identifying small pixel variations in five orientations, $26.5^{\circ}$, $45^{\circ}$, $90^{\circ}$, $135^{\circ}$, and $153.5^{\circ}$. The EDV values work as inputs of Sobel mask and return edge direction degree and confidence parameters. Based on edge direction degree and confidence parameters the missing pixel is computed. The results of computer simulations demonstrate that the proposed method outperforms a number of intra-field deinterlacing methods in the literature.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1431-1438
• /
• 2014

We analyzed the articles from "Kukje Shinmun" and "Busan Ilbo", which are two local newpapers of Busan Metropolitan City. The articles cover from January 1, 2013 to December 31, 2013. Meaningful pattern inherent in 2889 articles of which the title includes "Busan" and "Traffic" and related data was analyzed. Textmining method, which is a part of datamining, was used for the social network analysis (SNA). HDFS and MapReduce (from Hadoop ecosystem), which is open-source framework based on JAVA, were used with Linux environment (Uubntu-12.04LTS) for the construction of unstructured data and the storage, process and the analysis of big data. We implemented new algorithm that shows better visualization compared with the default one from R package, by providing the color and thickness based on the weight from each node and line connecting the nodes.

• Journal of Korean Society for Geospatial Information Science
• /
• v.10 no.1 s.19
• /
• pp.59-66
• /
• 2002

The rapid change of industrial structure causes to increase distribution cost and requires necessity of physical distribution system urgently. Traffic situation is getting extremely worse and traffic jam has led to increasing expense of physical distribution delivery which dominates 20% of distribution cost. In this situation, the shortest and most suitable path search system is required by modern people who must waste a lot of time for moving with a car or on the street as well as many companies. for these reasons, we developed the shortest-path-searching system applying the dijkstra algorithm which is one of the effective shortest path algorithm to GIS, and it was constructed by considering realistic urban traffic and the pattern of street in a physical situation. Also, this system was developed to be updated weight data automatically, considering the dynamic change of traffic situation such as a traffic information service which will be served in real time. Finally, we designed this system to serve on web by using MapObjects IMS.