• The KIPS Transactions:PartD
• /
• v.11D no.5
• /
• pp.1049-1062
• /
• 2004

With the popularization of a World Wide Web (WWW), the quantity of web information has been increased. Therefore, an efficient searching system is needed to offer the exact result of diverse Information to user. Due to this reason, it is important to extract and analysis of user requirements in the distributed information environment. The conventional searching method used the only keyword for the web searching. However, the searching method proposed in this paper adds the context information of keyword for the effective searching. In addition, this searching method extracts keywords by the new keyword extraction method proposed in this paper and it executes the web searching based on a keyword mining profile generated by the extracted keywords. Unlike the conventional searching method which searched for information by a representative word, this searching method proposed in this paper is much more efficient and exact. This is because this searching method proposed in this paper is searched by the example based query included content information as well as a representative word. Moreover, this searching method makes a domain keyword list in order to perform search quietly. The domain keyword is a representative word of a special domain. The performance of the proposed algorithm is analyzed by a series of experiments to identify its various characteristic.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.240-240
• /
• 2011

The first neutral beam injector (NBI-1) has been developed for the Korea Superconducting Tokamak Advanced Research (KSTAR) tokamak. A first long pulse ion source (LPIS-1) has been installed on the NBI-1 for an auxiliary heating and current drive of KSTAR core plasmas. Performance of ion and neutral beam extractions in the LPIS-1 was investigated initially on the KSTAR NBI-1 system, prior to the neutral beam injection into the main plasmas. The ion source consists of a JAEA magnetic bucket plasma generator with multi-pole cusp fields and a set of KAERI prototype-III tetrode accelerators with circular apertures. The inner volume of plasma generator and accelerator column in the LPIS-1 is approximately 123 liters. Final design requirements for the ion source were a 120 kV/ 65 A deuterium beam and a 300 s pulse length. The extraction of ion beams was initiated by the formation of arc plasmas in the LPIS-1, called as an arc-beam extraction method. A stable ion beam extraction of LPIS-1 has been achieved up to an 100 kV/42 A for a 4 s pulse length and an 80 kV/25 A for a 14 s pulse length. Optimum beam perveance of 1.21 microperv has been found at an accelerating voltage of 80 kV. Neutralization efficiency has been measured by using a water flow calorimetry (WFC) method of calorimeter and an operation of bending magnet. The full-energy species of ion beams have been detected by using the diagnostic method of optical multichannel analyzer (OMA). An arc efficiency of the LPIS was 0.6~1.1 A/kW depending on the operating conditions of arc discharge.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1111-1123
• /
• 2023

In recent times, semantic image segmentation methods using deep learning models have been widely used for monitoring changes in surface attributes using remote sensing imagery. To enhance the performance of various UNet-based deep learning models, including the prominent UNet model, it is imperative to have a sufficiently large training dataset. However, enlarging the training dataset not only escalates the hardware requirements for processing but also significantly increases the time required for training. To address these issues, transfer learning is used as an effective approach, enabling performance improvement of models even in the absence of massive training datasets. In this paper we present three transfer learning models, UNet-ResNet50, UNet-VGG19, and CBAM-DRUNet-VGG19, which are combined with the representative pretrained models of VGG19 model and ResNet50 model. We applied these models to building extraction tasks and analyzed the accuracy improvements resulting from the application of transfer learning. Considering the substantial impact of learning rate on the performance of deep learning models, we also analyzed performance variations of each model based on different learning rate settings. We employed three datasets, namely Kompsat-3A dataset, WHU dataset, and INRIA dataset for evaluating the performance of building extraction results. The average accuracy improvements for the three dataset types, in comparison to the UNet model, were 5.1% for the UNet-ResNet50 model, while both UNet-VGG19 and CBAM-DRUNet-VGG19 models achieved a 7.2% improvement.

• Proceedings of the Korean Geotechical Society Conference
• /
• 1999.10a
• /
• pp.405-414
• /
• 1999

The auger and screw piles have known an important evolution during the last decade. Besides the large success of augercast (CFA) piling systems, new systems have been developed combining, to a variable extent, the classical extraction auger with especially designed displacement tools in order to develop screw piles with partial or total lateral soil displacement. These last developments cover the whole range of lateral soil displacement and are more difficult than ever to compare. The authors present the latest evolutions in auger piling systems and compare them with respect to penetration performances, bearing capacities and amount of spoil generated. A special focus is given to a new efficient system: the OMEGA(H) pile in use in Korea since 1997. The results of the Hongcheon site are presented where this R system was applied for a new investment of the Korean National Housing Corporation (KNHC). This first important experience, with the execution of some 1,500 Omega piles with diameter 410 mm, is presented. The piles were installed through loose silty sands down to very dense sands and layers of gravel. The results of full-scale load tests are analysed and show the conformity with requirements of the clients.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.18 no.3
• /
• pp.185-195
• /
• 2002

Purpose This article describes the historic and clinical aspects of the determination of the vertical dimension of occlusion and the synoptic procedure of the determination of the vertical dimension of occlusion in complete denture. The determining procedure of the susceptible vertical dimension of occlusion is one of the most important steps in construction of complete denture and prosthodontic treatment. It is considered essential for the improvement and the recovery of facial esthetics and stomatognathic functions. Results Several methods have been suggested for measurement of the vertical dimension of occlusion in the construction of complete denture and the prosthodontic rehabilitation. These range from pre-extraction records to the use of physiologic rest position, swallowing, phonetics, esthetics and facial proportion, etc. But, there is no universally accepted or completely accurate method. There seems to be no significant advantages of one technique other than those of cost, time and equipment requirements, and seems to be in controversial in determining the vertical dimension. Conclusion The vertical dimension of occlusion should be determined and reinspected carefully by dentist for a successful prosthesis with several methods. The more investigations are necessary for more objective and scientific techniques in determining the vertical dimension of occlusion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.643-661
• /
• 2018

This paper proposes a novel algorithm for lane detection based on inverse perspective transformation and Kalman filter. A simple inverse perspective transformation method is presented to remove perspective effects and generate a top-view image. This method does not need to obtain the internal and external parameters of the camera. The Gaussian kernel function is used to convolute the image to highlight the lane lines, and then an iterative threshold method is used to segment the image. A searching method is applied in the top-view image obtained from the inverse perspective transformation to determine the lane points and their positions. Combining with feature voting mechanism, the detected lane points are fitted as a straight line. Kalman filter is then applied to optimize and track the lane lines and improve the detection robustness. The experimental results show that the proposed method works well in various road conditions and meet the real-time requirements.

• Journal of Soil and Groundwater Environment
• /
• v.23 no.5
• /
• pp.1-16
• /
• 2018

Recent development of lithium ion batteries for vehicles industries have led to a boom in lithium exploration and development for the new generation of batteries. One of the cheapest sources of lithium is the brines hosted in the aquifers of the arid intermontane-closed salar basins. Because the resource is a fluid, with the attendant problems of in-aquifer mixing, reorganization, and lower recovery factors compared with most metalliferous and industrial mineral deposits due to reliance on pumping of the brine from wells for extraction, existing codes for filing resource and reserve estimates require new approach for these prospects. Evaluation of brine resources is complex and requires participation of a variety of qualified experts such as hydrogeologists, geologists, geochemists and chemical engineers. The technical reports disclosing the results of these estimates should reflect the inputs of multi-disciplinary approaches. The requirements for brine resource and reserve evaluation, drawing on several examples from the experiences in the Central Andes are reviewed in this paper.

• Proceedings of the Korean Nuclear Society Conference
• /
• 2004.10a
• /
• pp.719-720
• /
• 2004

A multi-megawatt long pulse ion source (LPIS) of neutral beam injector was developed for the KSTAR. Beam extraction experiments of the LPIS were carried out at the neutral beam test stand (NBTS). Design requirements for the ion source were 120 kV/65 A deuterium beam and a 300 s pulse length. A maximum ion density of $9.1310^{11}$ $cm^{-3}$ was measured by using electric probes, and an optimum arc efficiency of 0.46 A/kW was estimated with ion saturation current of the probes, arc power, and total beam area. An arcing problem, caused by the structural defect of decelerating grid supporter, in the third gap was solved by the blocking of backstream ion particles, originated from the plasma in the neutralizer duct, through the unnecessary spaces on the side of grid supporter. A maximum drain power of 1.5 MW (i.e. 70 kV/21 A) with hydrogen was measured for a pulse duration of 0.5 s. Optimum beam perveance was ranged from 0.75 to 0.85. An improved design of accelerator for the effective control of beam particle trajectory should provide higher beam perveance.

• Journal of Korea Multimedia Society
• /
• v.13 no.7
• /
• pp.1088-1094
• /
• 2010

Image authentication algorithms have to discriminate forged contents in the various critical fields of military, medical services, digital documents. They must ensure perceptual invisibility and fragility against malicious attacks. It is desirable that watermarking algorithms support sufficient insertion capacity and blind feature. And, high reliable algorithms that can eliminate false-positive and false-negative errors are needed in the watermark extraction process. In this paper, we control coefficients of high frequency band in a DCT domain and compensate brightness histogram for high reliability. As a result, we found that the proposed algorithm guarantee various requirements such as perceptual invisibility with high PSNR values, fragility, high reliability and blind feature. In addition, experiment results show that the proposed algorithm can be used steganographic applications by sufficient capacity of watermark.

• Smart Structures and Systems
• /
• v.25 no.5
• /
• pp.631-641
• /
• 2020

To meet the urgent requirements of safety surveillance from civil engineering management authorities, this study proposes a refined and efficient approach to generate full-field high-resolution panorama of construction sites using camera-amounted UAV (Unmanned Aerial Vehicle). GPS (Global Position System) information extraction for pre-registration, feature points filtering for efficient registration and optimal seaming line seeking for fusion are performed in sequence to form the full-field panorama generation framework. Advantages of the proposed method are as follows. First, GPS information can sort images for pre-registration, avoiding inefficient repeated pairwise calculations and matching. Second, the feature points are filtered according to the characteristics of the construction site images to reduce the amount of calculation. The proposed framework is validated on a road construction site and results demonstrate that it can generate an accurate and high-quality full-site panorama for the safety supervision in a much efficient manner.