• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.409-418
• /
• 2021

The land cover map is a very important data that is used as a basis for decision-making for land policy and environmental policy. The land cover map is mapped using remote sensing data, and the classification results may vary depending on the acquisition time of the data used even for the same area. In this study, to overcome the classification accuracy limit of single-period data, multi-series satellite images were used to learn the difference in the spectral reflectance characteristics of the land surface according to seasons on a U-Net model, one of the deep learning algorithms, to improve classification accuracy. In addition, the degree of improvement in classification accuracy is compared by comparing the accuracy of single-period data. Seoul, which consists of various land covers including 30% of green space and the Han River within the area, was set as the research target and quarterly Sentinel-2 satellite images for 2020 were aquired. The U-Net model was trained using the sub-class land cover map mapped by the Korean Ministry of Environment. As a result of learning and classifying the model into single-period, double-series, triple-series, and quadruple-series through the learned U-Net model, it showed an accuracy of 81%, 82% and 79%, which exceeds the standard for securing land cover classification accuracy of 75%, except for a single-period. Through this, it was confirmed that classification accuracy can be improved through multi-series classification.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1771-1777
• /
• 2023

With the advancements of deep learning, many semantic segmentation-based methods for land cover classification have been proposed. However, existing deep learning-based models only use image information and cannot guarantee spatiotemporal consistency. In this study, we propose a land cover classification model using geographical coordinates. First, the coordinate features are extracted through the Coordinate Hash Encoder, which is an extension of the Multi-resolution Hash Encoder, an implicit neural representation technique, to the longitude-latitude coordinate system. Next, we propose an architecture that combines the extracted coordinate features with different levels of U-net decoder. Experimental results show that the proposed method improves the mean intersection over union by about 32% and improves the spatiotemporal consistency.

• Journal of the Korea Concrete Institute
• /
• v.22 no.1
• /
• pp.131-141
• /
• 2010

The objective of this study was to propose a development length equation for GFRP bars. A total of 104 modified pullout tests were completed while the test variables were embedment length (15, 30, $45d_b$), net cover thickness ($0.5{\sim}2.0d_b$), top-cast bar effect, different GFRP bar types (K2KR, K3KR and AsUS), and bar diameters (10, 13, 16 mm). Average bond stresses were determined based on modified pullout test results. Two variable linear regression analysis was performed of the average bond stresses. Utilizing 5% fractile concept, a conservative development length design equation was derived. The design equation derived in this study was compared to the ACI 440 committee equation. The cross-comparison revealed that the current equation resulted in shorter development lengths than those determined by the ACI 440 equation when the net cover thickness was large (greater than $1.0d_b$). On the other hand, when the net cover thickness was small (equal to or less than $1.0d_b$), the development lengths required by the current equation were larger than those by the ACI equation. The bond stresses were significantly influenced by the cover thicknesses. The current equation results in development lengths that are more economical when the cover thickness is large, and more conservative lengths when the cover thickness is small than the ACI 440 committee equation.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.3
• /
• pp.46-51
• /
• 2002

Recently, High capacity alternator are used fur some special equipments in industry. But, several serious problem are occured, especially, broken coil of rotor, caused by crash with stator on rotating the rotor. Although added coil cover thor protect coil of rotor, coil cover is broken. In this study purpose 2 step for corrected that problems. First, three dimensional finite element method far investigate what is most important point. For that purpose, performed stress analysis of coil and coil cover that modeling and finite analysis by ANSYS software. Second, Apply prestress when winding the coil on stator to modify direction of net force. Vector analysis is used for determine corrected prestress. Result of the analysis and prestress are reviewed.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.10 no.1
• /
• pp.153-167
• /
• 2008

High temperature and natural sun light are considered as the core conditions for high quality and large quantity of Phellinus baumii production. However still now on there has been a mistake of excessively cutting off the natural light by spreading the closing nets on the mushroom cultivating house. For an example there are many houses where the closing nets under the roofs be extended to cover the sides of the houses, which way prevents the mushrooms in the houses from receiving sufficient natural sun light and getting the temperature sufficiently to grow so that the quantity and quality of the produced mushrooms are lowered even though the mushrooms can grow in those conditions. In order to avoid this mistake, the closing nets must be placed on the roofs of the houses only without dropping them to cover the sides. Further more when the closing nets are placed triply at the beginning stage of Phellinus baumii's growth in the house, the nets restrain the internal temperature of the house going up and intercept the natural bright light flowing into the house so that the growing tardiness occur to the Phellinus baumii. Therefore the roof only must have been covered by the closing net for 65% cutting off the light until May, and then covered by double folded the net for June, triple folded the net for July and August, double folded the net for September, and the single net for October. When the ventilation in the house has been maintained until the house tightly balloon out through controling lifting force of internal air, the Phellinus baumii can grow well while the bed logs themselves aren't dried out. Marketing is also very much important as well as increasing quality and quantity of Phellinus baumii production.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1663-1676
• /
• 2022

Recently, a number of deep-learning based land cover segmentation studies have been introduced. Some studies denoted that the performance of land cover segmentation deteriorated due to insufficient training data. In this study, we verified the improvement of land cover segmentation performance through data augmentation. U-Net was implemented for the segmentation model. And 2020 satellite-derived landcover dataset was utilized for the study data. The pixel accuracies were 0.905 and 0.923 for U-Net trained by original and augmented data respectively. And the mean F1 scores of those models were 0.720 and 0.775 respectively, indicating the better performance of data augmentation. In addition, F1 scores for building, road, paddy field, upland field, forest, and unclassified area class were 0.770, 0.568, 0.433, 0.455, 0.964, and 0.830 for the U-Net trained by original data. It is verified that data augmentation is effective in that the F1 scores of every class were improved to 0.838, 0.660, 0.791, 0.530, 0.969, and 0.860 respectively. Although, we applied data augmentation without considering class balances, we find that data augmentation can mitigate biased segmentation performance caused by data imbalance problems from the comparisons between the performances of two models. It is expected that this study would help to prove the importance and effectiveness of data augmentation in various image processing fields.

• Asian Journal of Turfgrass Science
• /
• v.16 no.2
• /
• pp.75-83
• /
• 2002

The back-filled soil of the New Incheon International Airport construction site was reclaimed with sea sand in near the Young-Jong island. The primary study was carried out from August 1993 to June 1997 to study soil amendment and to select salt resistance turfgrass species. This study dealt with low maintenance area that included most part of open space of airport site. The second experiment, from October 1996 to August 1998, focused on soil amendment and selection of turfgrass species for alongside runway where turf area was maintained. Through two previous studies, propagation methods with zoysiagrass were tested for alongside runway and surrounding areas at 1998. The study of construction methods with zoysiagrass, vegetative propagation showed better results on visual quality and cover rating compare with seeding propagation. However, significant different between vegetative and seeding propagation was not showed on visual quality and drought tolerance after one year of plot establishment. The cover rating by seeding construction methods reached in excess of 70% of limitation suggested by the Incheon International Airport Cooperation. Zoysia net and sprigging net methods were the most suitable where there requires rapid and high rate of ground cover. Seeding propagation should be acceptable to obtain a resonable cover rating where there allows relatively longer period of completion. Therefore, it should be possible to attain a proper rating of ground cover on the site of open space, alongside runway or areas similar to the New Incheon International Airport which is being built on dredged seashore sand. However, the methods of soil amendment, selecting salt tolerance species, and proper construction procedure should be considered at the a time.

• Journal of IKEEE
• /
• v.14 no.4
• /
• pp.305-311
• /
• 2010

Most recent web search engines perform effective keyword-based multimedia contents retrieval by investigating keywords associated with multimedia contents on the Web and comparing them with query keywords. On the other hand, most music and compilation albums provide professional artwork as cover art that will be displayed when the music is played. If the cover art is not available, then the music player just displays some dummy or random images, but this has been a source of dissatisfaction. In this paper, in order to automatically create cover art that is matched with music contents, we propose a music album cover art creation scheme based on music contents analysis and result synthesis. We first (i) analyze music contents and their lyrics and extract representative keywords, (ii) expand the keywords using WordNet and generate various queries, (iii) retrieve related images from the Web using those queries, and finally (iv) synthesize them according to the user preference for album cover art. To show the effectiveness of our scheme, we developed a prototype system and reported some results.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.32 no.3
• /
• pp.257-265
• /
• 1996

The combined hydroacoustic and bottom trawl surveys were conducted in the Cheju southeastern area by the training ship "KAYA" belong to Pukyong National University in July 1995 and the traning ship "NAGASAKI MARl]" belong to Nagasaki University in April 1994, respectively. The main purpose of the investigations was to provide the basic data for the management and the biomass estimation of commercially important demersal fish stocks in this area. Fish samples were collected by bottom trawling from 10 trawl stations randomly selected in the survey area, and the species and length compositions of trawl catches were examined. The fish school target strength for demersal fish aggregations was related to the catchability of trawl net with a 90 mm mesh codend. The most abundant species in the 1995 trawl stations were Japanese flying squid, sword tip squid and red horsehead and that of the 1994 trawl stations Japanese flying squid and blackmouth goosefish. The average weight per cubic meter of trawl catches collected by bottom trawling in the Cheju southeastern area were $1.0791{\times}lO^-4$kg/$m^3$ in the 1994 survey area and $1.3636{\times}lO^-4$kg/$m^3$ in the 1995 survey area, respectively. The catch data by cover net suggest that the efficiency of trawl net could affect the weight normalized target strength values for demersal fish aggregations. That is, the average target strength per unit of weight dropped from - 33.1 dB/kg using the total catch by codend and cover net to - 30.5 dB/kg using only the catch data by codend, and a change of2.6 dB/kg was observed.ange of2.6 dB/kg was observed.

• FLOWER RESEARCH JOURNAL
• /
• v.19 no.1
• /
• pp.1-7
• /
• 2011

This study was carried out to establish a production system of cut flowers and bulbs of oriental hybrid lilies in highland in summer. In study, effect of cultivation type and shading was examined. The cultivation types used in the bulb production experiments were greenhouses with no cover, greenhouses with net screen (net screen house), and greenhouses with EVA roof cover and net screen sides (rain shelter house). Net screen and rain shelter houses had either 0 or 30% shading. Cultivation type had significant effect and shading had a minor effect on bulb production. The greatest yield and quality of produced bulbs were obtained in the rain shelter house with 30% shading. From the results of this study, it is concluded that dispersed production of high quality cut lily flowers and bulbs could be achieved during the summer in highland.