• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.461-469
• /
• 2022

This paper presents a novel training data extraction approach using semi-supervised learning (SSL)-based classification without the analyst intervention for time-series land-cover mapping. The SSL-based approach first performs initial classification using initial training data obtained from past images including land-cover characteristics similar to the image to be classified. Reliable training data from the initial classification result are then extracted from SSL-based iterative classification using classification uncertainty information and class labels of neighboring pixels as constraints. The potential of the SSL-based training data extraction approach was evaluated from a classification experiment using unmanned aerial vehicle images in croplands. The use of new training data automatically extracted by the proposed SSL approach could significantly alleviate the misclassification in the initial classification result. In particular, isolated pixels were substantially reduced by considering spatial contextual information from adjacent pixels. Consequently, the classification accuracy of the proposed approach was similar to that of classification using manually extracted training data. These results indicate that the SSL-based iterative classification presented in this study could be effectively applied to automatically extract reliable training data for time-series land-cover mapping.

• Korean Journal of Ecology and Environment
• /
• v.33 no.3 s.91
• /
• pp.230-243
• /
• 2000

A large-scale community data were patterned by utilizing an unsupervised learning algorithm in artificial neural networks. Data for benthic macroinvertebrates in streams of South Korea reported in publications for 12 years from 1984 to 1995 were provided as inputs for training with the Kohonen network. Taxa included for the training were 5 phylum, 10 class, 26 order, 108 family and 571 species in 27 streams. Abundant groups were Diptera, Ephemeroptera, Trichoptera, Plecoptera, Coleoptera, Odonata, Oligochaeta, and Physidae. A wide spectrum of community compositions was observed: a few tolerant taxa were collected at polluted sites while a high species richness was observed at relatively clean sites. The trained mapping by the Kohonen network effectively showed patterns of communities from different river systems, followed by patterns of communities from different environmental disturbances. The training by the proposed artificial neural network could be an alternative for organizing community data in a large-scale ecological survey.

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.1965-1974
• /
• 2021

Satellite remote sensing approach can be actively used for forest monitoring. Especially, it is much meaningful to utilize Korea multi-purpose satellites, an independently operated satellite in Korea, for forest monitoring of Korea, Recently, several studies have been performed to exploit meaningful information from satellite remote sensed data via machine learning approaches. The forest information produced through machine learning approaches can be used to support the efficiency of traditional forest monitoring methods, such as in-situ survey or qualitative analysis of aerial image. The performance of machine learning approaches is greatly depending on the characteristics of study area and data. Thus, it is very important to survey the best model among the various machine learning models. In this study, the performance of deep neural network to classify artificial or natural forests was analyzed in Samcheok, Korea. As a result, the pixel accuracy was about 0.857. F1 scores for natural and artificial forests were about 0.917 and 0.433 respectively. The F1 score of artificial forest was low. However, we can find that the artificial and natural forest classification performance improvement of about 0.06 and 0.10 in F1 scores, compared to the results from single layered sigmoid artificial neural network. Based on these results, it is necessary to find a more appropriate model for the forest type classification by applying additional models based on a convolutional neural network.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2010.09a
• /
• pp.790-799
• /
• 2010

The Upper Devonian Grossmont Formation in Alberta, Canada reserves an estimated 50 billion cubic meters of bitumen and possess about 1/6 of the total bitumen resources in northern Alberta. However, unlike the overlying Athabasca oil sands, non conventional bitumen resources has not been developed as yet. The carbonate rocks of Grosmont Formation have been subject to various stages of diagenesis, including dolomatization and karstification with a strong effect on the distribution of porosity and permeability, which resulted in highly heterogeneous reservoirs. An extensive fracture logging and mapping was performed on total of six boreholes located in the study area to explore the characteristics of fracture geometry system and the subsurface structures of carbonates reservoir that holds bitumen. Fractal dimension was used as a measure of the statistical homogeneity of the fractured rock masses. The applicability of random Cantor dust, Dc, as a fractal parameter was examined systematically. The statistical homogeneity of fractured carbonates rock masses was investigated in the study area. The structural domains of the rock masses were delineated depthwise according to estimated Dc. The major fracture orientation was dominated by horizontal beddings having dip of $0-20^{\circ}$. Also, fractures having high dip angles existed with relatively low frequency. Three dimensional fracture network modeling for each structural domain has been performed based on fracture orientation and intensity, and some representative conceptual models for carbonates reservoir in the study area has been proposed. The developed subsurface conceptual models will be used to capture the geomechanical characteristics of the carbonates reservoir.

• The Journal of Engineering Geology
• /
• v.27 no.3
• /
• pp.305-312
• /
• 2017

The most effective way to distinguish subsurface interfaces that produce various geophysical responses is through the integration of multiple geophysical methods, with each method detecting both a complementary and unique set of distinct physical properties relating to the subsurface. In this study, shallow seismic reflection (SSR) and ground penetrating radar (GPR) surveys were conducted at the Cheongju-Gadeok site of the Korea National Groundwater Monitoring Network to map the water table, which was measured at 12 m depth during the geophysical surveys. The water table proved to be a good target reflector in both datasets, as the abrupt transition from the overlying unsaturated weathered rock to the underlying saturated weathered rock yielded large acoustic impedance and dielectric constant contrasts. The two datasets were depth converted and integrated into a single section, with the SSR and GPR surveys conducted to ensure subsurface imaging at approximately the same wavelength. The GPR data provided detailed information on the upper ~15 m of the section, whereas the SSR data imaged structures at depths of 10-45 m. The integrated section thus captured the full depth coverage of the sandy clay, water table, weathered rock, soft rock, and hard rock structures, which correlated well with local drillcore and water table observations. Incorporation of these two geophysical datasets yielded a synthetic section that resembled a simplified aquifer model, with the best-fitting seismic velocity, dielectric constant, and porosity of the saturated weathered layer being $v_{seismic}=1000m/s$, ${\varepsilon}_r=16$, and ${\phi}=0.32$, respectively.

• Journal of Communications and Networks
• /
• v.12 no.6
• /
• pp.568-576
• /
• 2010

The paper presents a control architecture aimed at implementing bandwidth optimization combined with call admission control (CAC) over a digital video broadcasting (DVB) return channel satellite terminal (RCST) under quality of service (QoS) constraints. The approach can be applied in all cases where traffic flows, coming from a terrestrial portion of the network, are merged together within a single DVB flow, which is then forwarded over the satellite channel. The paper introduces the architecture of data and control plane of the RCST at layer 2. The data plane is composed of a set of traffic buffers served with a given bandwidth. The control plane proposed in this paper includes a layer 2 resource manager (L2RM), which is structured into decision makers (DM), one for each traffic buffer of the data plane. Each DM contains a virtual queue, which exactly duplicates the corresponding traffic buffer and performs the actions to compute the minimum bandwidth need to assure the QoS constraints. After computing the minimum bandwidth through a given algorithm (in this view the paper reports some schemes taken in the literature which may be applied), each DM communicates this bandwidth value to the L2RM, which allocates bandwidth to traffic buffers at the data plane. Real bandwidth allocations are driven by the information provided by the DMs. Bandwidth control is linked to a CAC scheme, which uses current bandwidth allocations and peak bandwidth of the call entering the network to decide admission. The performance evaluation is dedicated to show the efficiency of the proposed combined bandwidth allocation and CAC.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2002.11a
• /
• pp.520-532
• /
• 2002

A new architecture called the Storage Area Network(SAN) was developed in response to the requirements of high availability of data scalable growth and system performance. In order to use SAN more efficiently, most SAN operating systems support storage virtualization concepts that allow users to view physical storage devices attached to SAN as a large volume virtually. A logical volume manager plays a key role in storage virtualization. In order for mapping managers to process snap-shots and reorganizations, an efficient freespace manager is required, and it affects the overall performance of logical volume. In this thesis, we design and implement a freespace manager for logical volume manager. The freespace manager efficiently allocates physical blocks for logical blocks. Also, it processes space allocation requests for suports snapshots and reorganizations.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.24 no.1
• /
• pp.207-228
• /
• 2013

Due to the advance of scholarly communication, the field of open access has been studied over the last decade. The purpose of this study is to analyze and demonstrate the field of open access via co-word analysis. The data set was collected from Web of Science citation database during the period from January 1998 to July 2012 using the Topic category. A total of 479 journal articles were retrieved and 8,643 noun keywords were extracted from the titles and abstracts. In order to achieve the purpose of this study, network analysis, clustering analysis and multidimensional scaling mapping were used to examine the domain and the sub-domains of open access field. 18 clusters in the network analysis are recognized and 4 clusters are shown in the map of multidimensional scaling. In addition, the centrality analysis in the weighted networks was used to explore the significant keywords in this field. The results of this study are expected to demonstrate and guide the intellectual structure and new approaches of open access field.

• Korean Journal of Microbiology
• /
• v.29 no.6
• /
• pp.392-396
• /
• 1991

Mu d1(Ap lac) bacteriophage can be used to search for genes which are members of a common regulatory network without having to know the functions of the genes in advance. Aim was for obtaining the loci in the SOS network as well as temperature inducible loci. For this purpose, recA441 allele was used. This allele encodes a thermosensitive recA gene product; thus, the recA441 allele can be activated upon temperature upshift without by external DNA damage. Approximately 10, 000 colonies were screened, and then searched for the colonies which expressed .betha.-galactosidase higher level at 42.deg.C than at 30.deg.C. The strains identified fell into two dlasses; (i) ones in which the increased expression was $recA^{+}$ $lexA^{+}$ -dependent, that is, din(damage-inducible) genes which were due to the activation of recA441 allele and (ii) ones in which the increased expression was $recA^{+}$ $lexA^{+}$ -independent and only temperature-inducible, tin genes. Rough mapping position was obtained for these genes.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.11a
• /
• pp.520-532
• /
• 2002

A new architecture called the Storage Area Network(SAN) was developed in response to the requirements of high availability of data, scalable growth, and system performance. In order to use SAN more efficiently, most SAN operating systems support storage virtualization concepts that allow users to view physical storage devices attached to SAN as a large volume virtually. A logical volume manager plays a key role in storage virtualization. In order for mapping managers to process snap-shots and reorganizations, an efficient freespace manager is required, and it affects the overall performance of logical volume. In this thesis, we design and implement a freespace manager for logical volume manager. The freespace manager efficiently allocates physical blocks for logical blocks. Also, it processes space allocation requests for supports snapshots and reorganizations.