• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.6
• /
• pp.67-74
• /
• 2006

In this paper, a tiled display system is developed to get a high-resolution image in visualization of large-scale structural analysis data with low-resolution display devices and low-cost cluster computer system. Concerning the hardware system, some of the crucial points are investigated, and a new beam-projector positioner is designed and manufactured to resolve the keystone phenomena which result in distorted image. In the development of tiled display software, Qt and OpenGL are utilized for GUI and rendering, respectively. To obtain the entire tiled image, LAM-MPI is utilized to synchronize the several sub-images produced from each cluster computer node.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.149-158
• /
• 2024

Traditional manual identification of crop leaf diseases is challenging. Owing to the limitations in manpower and resources, it is challenging to explore crop diseases on a large scale. The emergence of artificial intelligence technologies, particularly the extensive application of deep learning technologies, is expected to overcome these challenges and greatly improve the accuracy and efficiency of crop disease identification. Crop leaf disease identification models have been designed and trained using large-scale training data, enabling them to predict different categories of diseases from unlabeled crop leaves. However, these models, which possess strong feature representation capabilities, require substantial training data, and there is often a shortage of such datasets in practical farming scenarios. To address this issue and improve the feature learning abilities of models, this study proposes a deep transfer learning adaptation strategy. The novel proposed method aims to transfer the weights and parameters from pre-trained models in similar large-scale training datasets, such as ImageNet. ImageNet pre-trained weights are adopted and fine-tuned with the features of crop leaf diseases to improve prediction ability. In this study, we collected 16,060 crop leaf disease images, spanning 12 categories, for training. The experimental results demonstrate that an impressive accuracy of 98% is achieved using the proposed method on the transferred ResNet-50 model, thereby confirming the effectiveness of our transfer learning approach.

• Nuclear Engineering and Technology
• /
• v.56 no.8
• /
• pp.3199-3209
• /
• 2024

During the decommissioning of nuclear and particle accelerator facilities, a considerable amount of large-scale radioactive waste may be generated. Accurately defining the activation level of the waste is crucial for proper disposal. However, directly measuring the internal radioactivity distribution poses challenges. This study introduced a novel technology employing machine learning to assess the internal radioactivity distribution based on external measurements. Random radioactivity distribution within a structure were established, and the photon spectrum measured by detectors from outside the structure was simulated using the FLUKA Monte-Carlo code. Through training with spectrum data corresponding to various radioactivity distributions, an evaluation model for radioactivity using simulated data was developed by above Monte-Carlo simulation. Convolutional Neural Network and Transformer methods were utilized to establish the evaluation model. The machine learning construction involves 5425 simulation datasets, and 603 datasets, which were used to obtain the evaluated results. Preprocessing was applied to the datasets, but the evaluation model using raw spectrum data showed the best evaluation results. The estimation of the intensity and shape of the radioactivity distribution inside the structure was achieved with a relative error of 10%. Additionally, the evaluation based on the constructed model takes only a few seconds to complete the process.

• Fire Science and Engineering
• /
• v.20 no.4 s.64
• /
• pp.65-71
• /
• 2006

Some major properties such as a heat release rate have been measured experimentally for the validation of fire model and the clarification of fire phenomena as the study is more rigorous recently. Although the reduced-scale experiment also provides the basic data and the physical understanding in fire study, it is not enough to explain real fire problem directly because there is no exact analogy theory between a real fire and the reduced scale model. Therefore, large cone calorimeter have been built and used in a few foreign countries for the measurement of large scale fire. This paper addressed the theoretical background and the description of key features in the design of the facility. It will be a useful guide for implementation of the large scale cone calorimeter in the future.

• Journal of Hydrogen and New Energy
• /
• v.35 no.1
• /
• pp.14-26
• /
• 2024

Global efforts continue with the goal of transition to a "carbon neutral (net zero)" society with zero carbon emissions by 2050. For this purpose, the technology of water electrolysis is being developed, which can store electricity generated from renewable energies in large quantities and over a long period of time as hydrogen. Recently, various research and large-scale projects on 'green hydrogen', which has no carbon emissions, are being conducted. In this paper, a comparison of water electrolysis technologies was carried out and, based on data provided by the International Energy Agency (IEA), large-scale water electrolysis demonstration projects were analyzed by classifying them by technology, power supply, country and end user. It is expected that through the analysis of large-scale water electrolysis demonstration projects, research directions and road maps can be provided for the development/implementation of commercial projects in the future.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2000.11a
• /
• pp.41-44
• /
• 2000

With the advent of DNA microarray and "chip" technologies, gene expression in an organism can be monitored on a genomic scale, allowing the transcription levels of many genes to be measured simultaneously. Functional interpretation of massive expression data and linking such data to DNA sequences have become the new challenges to bioinformatics. I will us yeast cell cycle expression data analysis as an example to demonstrate how special database and computational methods may be used for extracting functional information, I will also briefly describe a novel clustering algorithm which has been applied to the cell cycle data.

• Journal of The Korean Astronomical Society
• /
• v.48 no.4
• /
• pp.213-228
• /
• 2015

The Horizon Run 4 is a cosmological N-body simulation designed for the study of coupled evolution between galaxies and large-scale structures of the Universe, and for the test of galaxy formation models. Using 6300³ gravitating particles in a cubic box of L_box = 3150 h⁻¹Mpc, we build a dense forest of halo merger trees to trace the halo merger history with a halo mass resolution scale down to M_s = 2.7 × 10¹¹h⁻¹M_⊙. We build a set of particle and halo data, which can serve as testbeds for comparison of cosmological models and gravitational theories with observations. We find that the FoF halo mass function shows a substantial deviation from the universal form with tangible redshift evolution of amplitude and shape. At higher redshifts, the amplitude of the mass function is lower, and the functional form is shifted toward larger values of ln(1/σ). We also find that the baryonic acoustic oscillation feature in the two-point correlation function of mock galaxies becomes broader with a peak position moving to smaller scales and the peak amplitude decreasing for increasing directional cosine μ compared to the linear predictions. From the halo merger trees built from halo data at 75 redshifts, we measure the half-mass epoch of halos and find that less massive halos tend to reach half of their current mass at higher redshifts. Simulation outputs including snapshot data, past lightcone space data, and halo merger data are available at http://sdss.kias.re.kr/astro/Horizon-Run4.

• Imaging Science in Dentistry
• /
• v.50 no.4
• /
• pp.331-337
• /
• 2020

Purpose: As cone-beam computed tomography (CBCT) has become the most widely used 3-dimensional (3D) imaging modality in the dental field, storage space and costs for large-capacity data have become an important issue. Therefore, if 3D data can be stored at a clinically acceptable compression rate, the burden in terms of storage space and cost can be reduced and data can be managed more efficiently. In this study, a deep learning network for super-resolution was tested to restore compressed virtual CBCT images. Materials and Methods: Virtual CBCT image data were created with a publicly available online dataset (CQ500) of multidetector computed tomography images using CBCT reconstruction software (TIGRE). A very deep super-resolution (VDSR) network was trained to restore high-resolution virtual CBCT images from the low-resolution virtual CBCT images. Results: The images reconstructed by VDSR showed better image quality than bicubic interpolation in restored images at various scale ratios. The highest scale ratio with clinically acceptable reconstruction accuracy using VDSR was 2.1. Conclusion: VDSR showed promising restoration accuracy in this study. In the future, it will be necessary to experiment with new deep learning algorithms and large-scale data for clinical application of this technology.

• Geomechanics and Engineering
• /
• v.13 no.6
• /
• pp.893-906
• /
• 2017

Automatic large scale soil model generation is very critical stage for earthquake hazard simulation of urban areas. Manual model development may cause some data losses and may not be effective when there are too many data from different soil observations in a wide area. Geographic information systems (GIS) for storing and analyzing spatial data help scientists to generate better models automatically. Although the original soil observations were limited to soil profile data, the recent developments in mapping technology, interpolation methods, and remote sensing have provided advanced soil model developments. Together with advanced computational technology, it is possible to handle much larger volumes of data. The scientists may solve difficult problems of describing the spatial variation of soil. In this study, an algorithm is proposed for automatic three dimensional soil and velocity model development of southern part of the European side of Istanbul next to Sea of Marmara based on GIS data. In the proposed algorithm, firstly bedrock surface is generated from integration of geological and geophysical measurements. Then, layer surface contacts are integrated with data gathered in vertical borings, and interpolations are interpreted on sections between the borings automatically. Three dimensional underground geology model is prepared using boring data, geologic cross sections and formation base contours drawn in the light of these data. During the preparation of the model, classification studies are made based on formation models. Then, 3D velocity models are developed by using geophysical measurements such as refraction-microtremor, array microtremor and PS logging. The soil and velocity models are integrated and final soil model is obtained. All stages of this algorithm are carried out automatically in the selected urban area. The system directly reads the GIS soil data in the selected part of urban area and 3D soil model is automatically developed for large scale earthquake hazard simulation studies.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.897-898
• /
• 2008

Array comparative genomic hybridization (aCGH) has been used to detect chromosomal regions of amplifications or deletions, which allows identification of new cancer related genes. As aCGH, a large-scale spatial data, contains significant amount of noises in its raw data, it has been an important research issue to segment genomic DNA regions to detect its true underlying copy number aberrations (CNAs). In this study, we focus on applying a segmentation method to multiple data sets. We compare two different threshold values for analyzing aCGH data with CBS method [1]. The proposed threshold values are p-value or $Q{\pm}1.5IQR$ and $Q{\pm}1.5IQR$.