• Journal of the Korea Society of Computer and Information
• /
• v.19 no.12
• /
• pp.11-19
• /
• 2014

In this paper, we design an intelligent agent system with the data mining module and information fusion module as the core components of the system and investigate the possibility for the medical expert system. In the data mining module, fuzzy neural network, OFUN-NET analyzes multi-view data and produces fuzzy cluster knowledge base. In the information fusion module and application module, they serve the diagnosis result with possibility degree and useful information for diagnosis, such as uncertainty decision status or detection of asymmetry. We also present the experiment results on the BI-RADS-based feature data set selected form DDSM benchmark database. They show higher classification accuracy than conventional methods and the feasibility of the system as a computer aided diagnosis system.

• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.100-108
• /
• 2023

In this study, a neural network model inspired by a one-dimensional convolution U-net is developed to automatically accelerate edge localized mode (ELM) detection from big diagnostic data of fusion devices and increase the detection accuracy regardless of the hyperparameter setting. This model recognizes the input signal patterns and overcomes the problems of existing detection algorithms, such as the prominence algorithm and those of differential methods with high sensitivity for the threshold and signal intensity. To train the model, 10 sets of discharge radiation data from the KSTAR are used and sliced into 11091 inputs of length 12 ms, of which 20% are used for validation. According to the receiver operating characteristic curves, our model shows a positive prediction rate and a true prediction rate of approximately 90% each, which is comparable to the best detection performance afforded by other algorithms using their optimized hyperparameters. The accurate and automatic ELM-burst detection methodology used in our model can be beneficial for determining plasma properties, such as the ELM frequency from big data measured in multiple experiments using machines from the KSTAR device and ITER. Additionally, it is applicable to feature detection in the time-series data of other engineering fields.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2001.05a
• /
• pp.16-19
• /
• 2001

Rapid Prototyping (RP) and Rapid Tooling (RT) were introduced to reduce time-to-market and cost by shortening not only the development phase but also the production phase of the manufacturing process. RP generally builds up a prototype layer by layer, rapidly generating a fully three-dimensional free form shape. RT enables the manufacture of production tools. The integration of RP and RT has the potential for rapid net shaping of thee-dimensional parts, which have geometrical complexity. In this study, net shaping techniques for making three-dimensional parts using RP and RT are described and a sample part are shown. A three-dimensional metal part is manufactured by a new RP process, Variable Lamination Manufacturing by using Expandable Polystyrene Foam (VLM-S), and its application to RT for making a clover punch. In addition, we discussed the technology fusion between metal forming md RP/RT.

• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.754-771
• /
• 2021

In the task of continuous dimension emotion recognition, the parts that highlight the emotional expression are not the same in each mode, and the influences of different modes on the emotional state is also different. Therefore, this paper studies the fusion of the two most important modes in emotional recognition (voice and visual expression), and proposes a two-mode dual-modal emotion recognition method combined with the attention mechanism of the improved AlexNet network. After a simple preprocessing of the audio signal and the video signal, respectively, the first step is to use the prior knowledge to realize the extraction of audio characteristics. Then, facial expression features are extracted by the improved AlexNet network. Finally, the multimodal attention mechanism is used to fuse facial expression features and audio features, and the improved loss function is used to optimize the modal missing problem, so as to improve the robustness of the model and the performance of emotion recognition. The experimental results show that the concordance coefficient of the proposed model in the two dimensions of arousal and valence (concordance correlation coefficient) were 0.729 and 0.718, respectively, which are superior to several comparative algorithms.

• Nuclear Engineering and Technology
• /
• v.51 no.4
• /
• pp.1142-1153
• /
• 2019

This paper presents the effect of fusion procedure on the fusion performance of the thermal butt fusion in the safety class III buried HDPE piping per various tests performed, including high speed tensile impact, free bend, blunt notched tensile, notched creep, and PENT tests. The suitability of fusion joints and qualification procedures was evaluated by comparing test results from the base material and buttfusion joints. From the notched tensile test result, it was found that the fused joints have much lower toughness than the base material. It was also identified that the notched tensile test is more desirable than the high speed tensile impact and free bend tests presented in the ASME Code Case N-755-3 as a fusion qualification test method. In addition, with regard to the single low-pressure fusion joint performances, the procedure given by the ISO 21307 was determined to be better that the one specified in the Code Case N-755-3.

• Nuclear Engineering and Technology
• /
• v.54 no.6
• /
• pp.2334-2337
• /
• 2022

Deuterium-tritium reaction is the most promising one in term of the highest nuclear fusion cross-section for the reactor. So it is one of urgent issues to develop materials and components that are simultaneously resistant to high heat flux and high energy neutron flux in realization of the fusion energy. 2.45 MeV neutron production was reported in D-D reaction in KSTAR and regarded as beam-target is the dominant process. The feasibility study of KSTAR to wide area neutron source facility is done in term of D-D and D-T reactions from the empirical scaling law from the mixed fast and thermal stored energy and its projection to cases of heating power upgrade and DT reaction is done.

• Nuclear Engineering and Technology
• /
• v.41 no.4
• /
• pp.455-476
• /
• 2009

The history of nuclear fusion research in Korea is rather short compared to that of advanced countries. However, since the mid-1990s, at which time the construction of KSTAR was about to commence, fusion research in Korea has been actively carried out in a wide range of areas, from basic plasma physics to fusion reactor design. The flourishing of fusion research partly owes to the fact that industrial technologies in Korea including those related to the nuclear field have been fully matured, with their quality being highly ranked in the world. Successive pivotal programs such as KSTAR and ITER have provided diverse opportunities to address new scientific and technological problems in fusion as well as to draw young researchers into related fields. The frame of the Korean nuclear fusion program is now changing from a small laboratory scale to a large national agenda. Coordinated strategies from different views and a holistic approach are necessary in order to achieve optimal efficiency and effectiveness. Upon this background, the present paper reflects upon the road taken to arrive at this point and looks ahead at the coming future in nuclear fusion research activities in Korea.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.409-418
• /
• 2021

As the amount of construction for aerial photography increases, the need for automation of quality inspection is emerging. In this study, an experiment was performed to classify or detect clouds in aerial photos using deep learning techniques. Also, classification and detection were performed by including satellite images in the learning data. As algorithms used in the experiment, GoogLeNet, VGG16, Faster R-CNN and YOLOv3 were applied and the results were compared. In addition, considering the practical limitations of securing erroneous images including clouds in aerial images, we also analyzed whether additional learning of satellite images affects classification and detection accuracy in comparison a training dataset that only contains aerial images. As results, the GoogLeNet and YOLOv3 algorithms showed relatively superior accuracy in cloud classification and detection of aerial images, respectively. GoogLeNet showed producer's accuracy of 83.8% for cloud and YOLOv3 showed producer's accuracy of 84.0% for cloud. And, the addition of satellite image learning data showed that it can be applied as an alternative when there is a lack of aerial image data.

• Nuclear Engineering and Technology
• /
• v.50 no.3
• /
• pp.411-415
• /
• 2018

Nuclear fusion may be among the strongest sustainable ways to replace fossil fuels because it does not contribute to acid rain or global warming. In this context, activated cobalt materials in corrosion products for fusion energy are significant in determination of dose levels during maintenance after a coolant leak in a nuclear fusion reactor. Therefore, cross-section studies on cobalt material are very important for fusion reactor design. In this article, the excitation functions of some nuclear reaction channels induced by proton particles on $^{59}Co$ structural material were predicted using different models. The nuclear level densities were calculated using different choices of available level density models in ALICE/ASH code. Finally, the newly calculated cross sections for the investigated nuclear reactions are compared with the experimental values and TENDL data based on TALYS nuclear code.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.