• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.4_1
• /
• pp.319-326
• /
• 2014

The accurate geo-referencing processes that apply ground control points is prerequisite for effective end use of HRSI (High-resolution satellite imagery). Since the conventional control point acquisition by human operator takes long time, demands for the automated matching to existing reference data has been increasing its popularity. Among many options of reference data, the airborne LiDAR (Light Detection And Ranging) data shows high potential due to its high spatial resolution and vertical accuracy. Additionally, it is in the form of 3-dimensional point cloud free from the relief displacement. Recently, a new matching method between LiDAR data and HRSI was proposed that is based on the image projection of whole LiDAR data into HRSI domain, however, importing and processing the large amount of LiDAR data considered as time-consuming. Therefore, we wmotivated to ere propose a local LiDAR chip generation for the HRSI geo-referencing. In the procedure, a LiDAR point cloud was rasterized into an ortho image with the digital elevation model. After then, we selected local areas, which of containing meaningful amount of edge information to create LiDAR chips of small data size. We tested the LiDAR chips for fully-automated geo-referencing with Kompsat-2 and Kompsat-3 data. Finally, the experimental results showed one-pixel level of mean accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.7
• /
• pp.221-228
• /
• 2020

Part-of-Speech (POS) tagged corpus is a collection of electronic text in which each word is annotated with a tag as the corresponding POS and is widely used for various training data for natural language processing. The training data generally assumes that there are no errors, but in reality they include various types of errors, which cause performance degradation of systems trained using the data. To alleviate this problem, we propose a novel method for detecting errors in the existing POS tagged corpus using the classifier of XGBoost and cross-validation as evaluation techniques. We first train a classifier of a POS tagger using the POS-tagged corpus with some errors and then detect errors from the POS-tagged corpus using cross-validation, but the classifier cannot detect errors because there is no training data for detecting POS tagged errors. We thus detect errors by comparing the outputs (probabilities of POS) of the classifier, adjusting hyperparameters. The hyperparameters is estimated by a small scale error-tagged corpus, in which text is sampled from a POS-tagged corpus and which is marked up POS errors by experts. In this paper, we use recall and precision as evaluation metrics which are widely used in information retrieval. We have shown that the proposed method is valid by comparing two distributions of the sample (the error-tagged corpus) and the population (the POS-tagged corpus) because all detected errors cannot be checked. In the near future, we will apply the proposed method to a dependency tree-tagged corpus and a semantic role tagged corpus.

• Journal of agriculture & life science
• /
• v.43 no.2
• /
• pp.47-53
• /
• 2009

This research is intended to study the induction net of chrysanthemum used in its greenhouse and requiring lots of time for manual work and review the structural safety of a bed and existing greenhouses after designing the bed of a net which increases cropping period by sharply reducing the time of work and provides the bed of induction nets for cultivating chrysanthemum with its existing and new greenhouse. The review of sectional and biodynamic properties in 15 kinds of materials has revealed that the pipes of ${\phi}38.1{\times}1.7t$and ${\phi}38.1{\times}2.0t$ didn't exceed stress ratio but did 10mm drooping allowance. For this reason, the pipe of ${\phi}48.1{\times}1.5t$ net both stress ratio and drooping allowance. For the safety, the middle chamber should be designed into Truss type owing to bed load, wind load, and snow load when the bed of an induction net is installed in the middle chamber. When installing the middle chamber with a truss type, the greenhouse of chrysanthemum in Geochang area needs proper reinforcements because the stress in colullllls and wind proof walls exceeds stress allowance regardless of the installation of a bed.

• Korean Journal of Construction Engineering and Management
• /
• v.20 no.6
• /
• pp.89-97
• /
• 2019

Recently, BIM has been actively introduced in construction projects. In particular, the introduction of BIM in cost estimating process is expected to improve the accuracy and efficiency of the cost estimate. However, the quantity calculation and BoQ documents preparation process still require manual work. Although the BIM model may support quantification process, it is still problematic that the level of detail of the model must be very detailed to meet the items in BoQ. To address this, it is necessary to analyse the LOD criteria and to analyze the extent to which quantity can be computed according to the level of detail in the BIM model. For the analysis of LOD-based work items, the work item grades were divided into A,B, and C. In this study, the ratio and cost of each item that can be calculated at the LOD level in the detailed design phase are reviewed for each type of work, and the method for improving the quantity calculation using BIM is proposed. In the LOD 300 stage(Detailed design stage), the largest number of items in the class B, the major improvement class, are window and glass work. In addition, the most expensive type of work was analyzed by reinforced concrete work. In the future, it is necessary to suggest appropriate improvement way for items with high item ratios and items with high cost ratios. The results of this study are expected to be used as a BIM-based cost estimation or as basic data for improving the current BoQ system.

• Proceedings of KOSOMES biannual meeting
• /
• 2008.05a
• /
• pp.117-120
• /
• 2008

As we can see at the oil spill occident on the coast of Taean, the viscosity of the spilled oil becomes thicker as time goes by. It becomes waste oil with the form of tar. It moves to other areas following a tide. When the temperature of the water goes up, the viscosity becomes lighter and forms oil film. It repeats the process spreading to and polluting extensive areas where the tide reaches. People have used hand nets to collect waste oil of tar at the sea. But it is very difficult for them to collect the tar type waste oil spread on large areas before it reaches to the beach. This paper tried to find a way to collect the tar type waste oil efficiently. It used absorption mat of boom type that uses the attachment characteristics of the tar type waste oil and floating waste recovery device of net type. It tested the possibility of using the devices at the oil spill accident on the beach of Taean. The result showed that net type recovery device was much more efficient in collecting the waste oil in short time than the hand net system of people. It confirmed that the net type recovery device could be used to collect tar type waste oil.

• Korean Journal of Heritage: History & Science
• /
• v.47 no.3
• /
• pp.160-171
• /
• 2014

The provision of support for the art of making these items needs to be considered with the focus on the following factors: the local situation of the areas where such traditional handicrafts are still made, the craftspeople involved in their production, and their communities. So far, discussion about how to reinvigorate traditional handicrafts, including those mentioned above, has been concentrated on the measures taken to promote them as part of the handicrafts industry and the allocation of the government's budget for important intangible cultural heritages. The government runs a traditional handicrafts management system and provides financial support for the craftspeople and spaces for exhibiting their work. This form of support has led to systematic management of traditional handicrafts and heightened public interest in cultural heritage, as well as publicizing the country's traditional crafts, but has made little progress in the following areas: the fostering of young people willing to learn traditional skills, diversification of the types of skills to be maintained, or establishment of the networks of collaboration among the craftspeople. The most important aspect among the efforts mentioned above is to maintain cultural traditions that are unique to each region by encouraging local craftspeople to engage in their work with a solid sense of pride backed up by financial support. This study was carried out in connection with the need to reinvigorate the art of making tanggeon (horsehair crown), manggeon (horsehair headband), and gat (black horsehair hat), which few people wear as they are used only for ornamental purposes nowadays. This study examined the circumstances surrounding the artisans engaged in the production of horsehair handicrafts prior to their designation as a cultural heritage, and the changes that occurred in the local communities associated with their production after the designation, in order to assess the status of inheritance of this tradition.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.6
• /
• pp.251-256
• /
• 2019

A capsule endoscopy is a medical device that can capture an entire digestive organ from the esophagus to the anus at one time. It produces a vast amount of images consisted of about 8~12 hours in length and more than 50,000 frames on a single examination. However, since the analysis of endoscopic images is performed manually by a medical imaging specialist, the automation requirements of the analysis are increasing to assist diagnosis of the disease in the image. Among them, this study focused on automatic detection of polyp images. A polyp is a protruding lesion that can be found in the gastrointestinal tract. In this paper, we propose a weighted-image generation method to enhance the polyp image learning by multi-scale analysis. It is a way to extract the suspicious region of the polyp through the multi-scale analysis and combine it with the original image to generate a weighted image, that can enhance the polyp image learning. We experimented with SVM and RF which is one of the machine learning methods for 452 pieces of collected data. The F1-score of detecting the polyp with only original images was 89.3%, but when combined with the weighted images generated by the proposed method, the F1-score was improved to about 93.1%.

• Journal of the Korean Society for information Management
• /
• v.35 no.4
• /
• pp.141-164
• /
• 2018

Recently, as the amount of academic literature has increased rapidly and complex researches have been actively conducted, researchers have difficulty in analyzing trends in previous research. In order to solve this problem, it is necessary to classify information in units of academic papers. However, in Korea, there is no academic database in which such information is provided. In this paper, we propose an automatic classification system that can classify domestic academic literature into multiple classes. To this end, first, academic documents in the technical science field described in Korean were collected and mapped according to class 600 of the DDC by using K-Means clustering technique to construct a learning set capable of multiple classification. As a result of the construction of the training set, 63,915 documents in the Korean technical science field were established except for the values in which metadata does not exist. Using this training set, we implemented and learned the automatic classification engine of academic documents based on deep learning. Experimental results obtained by hand-built experimental set-up showed 78.32% accuracy and 72.45% F1 performance for multiple classification.

• The Journal of Engineering Geology
• /
• v.29 no.2
• /
• pp.153-162
• /
• 2019

Rapid variations in the geometry (i.e., thickness) of the refractor and low velocities affect greatly the imaging of the reflectors of land seismic data. Conventional solutions to obtain the weathering models utilizes first break picking method, which requires time consuming steps and causes the human error in picking the first arrivals. A new interferometric approach (interferometric refraction statics, IRS) which utilizes the first arrival signal (S/N enhanced by refraction convolution stack) instead of first break picking, is tested in this study to the synthetic data from the velocity structure provided by surface geophysics (refraction, MASW) and borehole geophysics (tomography, SPS logging) for the Cheongju granitic bodies. The results of IRS approach are found to be better than the ones from conventional first break picking in terms of continuities and horizontal resolution of the reflectors. The unresolved long-wavelength statics in brute stack are much removed by IRS weathering correction and the overlying refractors (the base of shallow weathering zone) are incidentally delineated in the refraction convolution stack.

• Journal of Korea Water Resources Association
• /
• v.52 no.8
• /
• pp.515-529
• /
• 2019

Geospatial information for river network and watershed boundary have played a fundamental roles in terms of river management, planning and design, hydrological and hydraulic analysis. Irrespective of their importance, the lack of punctual update and improper maintenance in currently available river-related geospatial information systems has revealed inconsistency issues between individual systems and spatial inaccuracy with regard to reflecting dynamically transferring riverine geography. Given that digital elevation models (DEMs) of high spatial resolution enabling to reproduce precise river network are only available adjacent to national rivers, DEMs with poor spatial resolution lead to generate unreliable river network information and thereby reduce their extensible applicabilities. This study first of all evaluated published spatial information available in Korea with respect to their spatial accuracy and consistency, and also provides a methodology and tool to modify existing low resolution of DEMs by means of striation of conventional or digitized river network to replicate input river network in various degree of further delineation. The tool named FSND was designed to be operated in ArcGIS ModelBuilder which ensures to automatically simulate river network striation to DEMs and delineation with different flow accumulation threshold. The FNSD was successfully validated in Seom River basin to identify its replication of given river network manually digitized based on recent aerial photograph in conjunction with a DEM with 30 meter spatial resolution. With the derived accuracy of reproducibility, substantiation of a various order of river network and watershed boundary from the striated DEM posed tangible possibility for highly extending DEMs with low resolution to be capable of producing reliable riverine spatial information subsequently.