• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1077-1094
• /
• 2016

Apache Spark is a fast and general-purpose cluster computing package. It provides a new abstraction named resilient distributed dataset, which is capable of support for fault tolerance while keeping data in memory. This type of abstraction results in a significant speedup compared to legacy large-scale data framework, MapReduce. In particular, Spark framework is suitable for iterative machine learning applications such as logistic regression and K-means clustering, and interactive data querying. Spark also supports high level libraries for various applications such as machine learning, streaming data processing, database querying and graph data mining thanks to its versatility. In this work, we introduce the concept and programming model of Spark as well as show some implementations of simple statistical computing applications. We also review the machine learning package MLlib, and the R language interface SparkR.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.5
• /
• pp.243-248
• /
• 2014

This study presents an investigation into the ways to detect the areas of object movement with Kinect's Depth Frame, which is capable of receiving 3D information regardless of external light sources. Applied to remove noises along the boundaries of objects among the depth information received from sensors were the blurring technique for the x and y coordinates of pixels and the frequency filter for the z coordinate. In addition, a clustering filter was applied according to the changing amounts of adjacent pixels to extract the areas of moving objects. It was also designed to detect fast movements above the standard according to filter settings, being applicable to mobile robots. Detected movements can be applied to security systems when being delivered to distant places via a network and can also be expanded to large-scale data through concerned information.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.44 no.5
• /
• pp.506-515
• /
• 2011

To elucidate demersal organism community structure, abundance, and biomass by period and region in the Uljin marine ranching area, otter trawl surveys were conducted at six stations from February 2008 to December 2009. During the survey period, a total of 118 species was collected in the Uljin marine ranching area, including 76 pisces, 27 mollusca and 15 crustaceans. The average abundance of individuals was 806 ind./$m^2$ and average biomass was 31,853 g/m2. The numbers of species, species richness, individuals per area and biomass per area were higher in 2009 than in 2008. Cluster and MDS analysis, based on a Bray-Curtis similarity matrix of fourth root transformed data of number of species and individuals per area, showed division into two different groups of demersal organisms in 2008 (Group 1) and one demersal organism community in 2009 (Group 2). From cluster and MDS analysis using the similarity of the demersal community among six stations, the Uljin marine ranching area was divided into Group A (St. 2, St. 4, St. 5 and St. 6) and Group B (St. 1 and St. 3).

• Korean Journal of Remote Sensing
• /
• v.36 no.5_3
• /
• pp.1007-1011
• /
• 2020

Recently, natural and anthropogenic disasters have rapidly increased due to the on-going climate change and various human activities. Remote sensing (RS) technology enables the continuous monitoring and rapid detection of disastrous events thanks to its advantages covering vast areas at high temporal resolution. Moreover, RS technology has been very actively used in disaster monitoring and assessment since cluster- and micro-satellites and drones were introduced and became popular. In this special issue, nine papers were introduced, including the processing and applications of remote sensing data for monitoring, assessment, and prediction of various natural disasters. These papers are expected to serve as useful references for disaster management in the future.

• Journal of environmental and Sanitary engineering
• /
• v.10 no.2
• /
• pp.36-45
• /
• 1995

Physicochemical environments and phytoplankton community in the nakdong river middle basin were investigated during a period of 1 year from Nov. 1993 to Oct. 1994. The ranges of Physicochemical environment factors in the nakdong river middle basin were $3.6~27.4{\circ}C$ for temperature, 7.1~9.3 for pH, $7.1~12.7mg/{\ell}$for DO, $0.7~2.1mg/{\ell}$ for BOD, $31~52mg/{\ell}$ for alkalinity, $2.101~3.3mg/{\ell}$ for T-N, $0.052~0.099mg/{\ell}$ for T-p, $5.4~92.3mg/m^{3}$ for chlorophyll-a. The number of Phytoplankton was lowest in Jan. that is, 244ce11s/ml and that was highest in Sep. 1201 cells/ml. The increase of dominant genera were correlated with the total nomber of phytoplanktons. Bacillariaceae, chlorophyceae, cyanophyceae, and euglenophlrceae were to 74.5%, 20%, 3.5%, and 2.2% respectively. Bacillariaceae were abundant during all the year round except for Jun, Jul, Aug. Chlorophyceae were plentiful in Jun, Jul, Aug. Tase-and odor-causing algae and filter clogging algae were 72% in Nov. and 93.5% in Jun. They were abundant all the year round except for winter, so they influenced on the treatment of tap water system. They belong to the general asterionella, cyclotella, melosira, nitzschia of bacillariaceae and actinastrum, pediastrum of chlorophyceae. Dominace index was low in winter and high in umber. Diversity index and equitability index were low in summer and high in winter. Therefore, the structure of phytoplankton community of the nakdong river middle basin was different form season to season.

• Korean Journal of Ecology and Environment
• /
• v.51 no.1
• /
• pp.16-28
• /
• 2018

Although physiological and ecological characteristics of cyanobacteria have been studied extensively for decades, unknown areas still remain greater than the already known. Recently, the development of omics techniques based on molecular biology has made it possible to view the ecosystem from a new and holistic perspective. The molecular mechanism of toxin production is being widely investigated, by comparative genomics and the transcriptomic studies. Biological interaction between bacteria and cyanobacteria is also explored: how their interactions and genetic biodiversity change depending on seasons and environmental factors, and how these interactions finally affect each component of ecosystem. Bioinformatics techniques have combined with ecoinformatics and omics data, enabling us to understand the underlying complex mechanisms of ecosystems. Particularly omics started to provide a whole picture of biological responses, occurring from all layers of hierarchical processes from DNA to metabolites. The expectation is growing further that algal blooms could be controlled more effectively in the near future. And an important insight for the successful bloom control would come from a novel blueprint drawn by omics studies.

• Journal of Wetlands Research
• /
• v.12 no.3
• /
• pp.67-78
• /
• 2010

Surface sedimentary facies and the change of microphytobenthos distribution in Geunso Bay tidal flat were monitored using remotely sensed data. Sediment distribution was analyzed along with the spectral reflectance based on the in situ data, and the spectral characteristics of the area where microphytobenthos occupied was examined. A medium to low spatial resolution of satellite image was not suitable for the detection of the surface sediments changes in the study area due to its ambiguity in the sedimentary facies boundary, but the seasonal changes of microphytobenthos distribution could be obviously detected. However, area of predominance of sand grains and seagrass distribution could be distinctly identified from a high spatial resolution remote sensing image. From this, it is expected that KOMPSAT-2 satellite images can be applied effectively to the study on the surface sedimentary facies and detailed ecological mapping in a tidal flat.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.4
• /
• pp.311-321
• /
• 2020

With the recent breakthrough in technology for micro-satellite, small satellite, and mega constellation missions, and the cost of accessing space from recycled space launch vehicles sharply lowered. Thus, space development is set to make a transition to a new space age that is different from before. Accordingly, the problem of worsening space environment conditions due to the increase in space debris, an inevitable adjunct to active space development, is also emerging as an international concern. In this paper, the contents and trends of international organizations for the protection of the space environment and space debris mitigation guidelines of each country were reviewed and forecast the future movement of the international community. In addition, it is suggested example of guidelines that fit the reality of Korea and how to apply them.

• Journal of the Institute of Convergence Signal Processing
• /
• v.9 no.3
• /
• pp.188-194
• /
• 2008

We propose the new method for the traffic signs recognition that is one of the DAS(Driving assistance system) in the intelligent vehicle. Our approach estimates a varied degree by using a geometric method from the varied traffic signs in noise, rotation and size, and extracts the recognition symbol from the compensated traffic sign for a recognition by using the sequential color-based clustering. This proposed clustering method classify the traffic sign into the attention, regulation, indication, and auxiliary class. Also, The circular tracing method is used for the final traffic sign recognition. To evaluate the effectiveness of the proposed method, varied traffic signs were built. As a result, The proposed method show that the 95 % recognition rate for a single variation, and 93 % recognition rate for a mixed variation.

• Journal of Digital Convergence
• /
• v.12 no.5
• /
• pp.57-68
• /
• 2014

This article analyses frame transition of personal information leakage in Korea from 1984 to 2014. In order to investigate the transition, we have collected newspaper article's titles. This study adopts classification, text network analysis(by co-occurrence symmetric matrix), and clustering techniques as part of social network analysis. Moreover, we apply definition of centrality in network in order to reveal the main frame formed in each of four periods. As a result, accessibility of personal information is extended from public sector to private sector. The boundary of personal information leakage is expanded to overseas. Therefore it is urgent to institutionalize the protection of personal information from a global perspective.