• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.139-149
• /
• 2019

In this paper, we present auto-annotation tool and synthetic dataset using 3D CAD model for deep learning based object detection. To be used as training data for deep learning methods, class, segmentation, bounding-box, contour, and pose annotations of the object are needed. We propose an automated annotation tool and synthetic image generation. Our resulting synthetic dataset reflects occlusion between objects and applicable for both underwater and in-air environments. To verify our synthetic dataset, we use MASK R-CNN as a state-of-the-art method among object detection model using deep learning. For experiment, we make the experimental environment reflecting the actual underwater environment. We show that object detection model trained via our dataset show significantly accurate results and robustness for the underwater environment. Lastly, we verify that our synthetic dataset is suitable for deep learning model for the underwater environments.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.47-55
• /
• 2001

Usually estimating the means is used for statistical inference. However depending the purpose of survey, sometimes totals will give the better and more meaningful in statistical inference than the means. Here in this study, we dealt with the unemployment population of small areas with using 4 different small area estimation methods: Direct, Synthetic, Composite, Bayes estimation. For all the estimates considered in this study, the average of absolute bias and men square error were obtained in the Monte Carlo Study which was simulated using data from 1998 Economic Active Population Survey in Korea.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1858-1872
• /
• 2023

With the advance of radar technologies, the availability of synthetic aperture radar (SAR) images increases. To improve application of SAR images, a management system for SAR images is proposed in this paper. The system provides trainable land cover classification module and display of SAR images on the map. Users of the system can create their own classifier with their data, and obtain the classified results of newly captured SAR images by applying the classifier to the images. The classifier is based on convolutional neural network structure. Since there are differences among SAR images depending on capturing method and devices, a fixed classifier cannot cover all types of SAR land cover classification problems. Thus, it is adopted to create each user's classifier. In our experiments, it is shown that the module works well with two different SAR datasets. With this system, SAR data and land cover classification results are managed and easily displayed.

• Nuclear Engineering and Technology
• /
• v.53 no.11
• /
• pp.3563-3579
• /
• 2021

Measured decay heat data of light water reactor (LWR) spent nuclear fuel (SNF) assemblies are adopted to train machine learning (ML) models. The measured data is available for fuel assemblies irradiated in commercial reactors operated in the United States and Sweden. The data comes from calorimetric measurements of discharged pressurized water reactor (PWR) and boiling water reactor (BWR) fuel assemblies. 91 and 171 measurements of PWR and BWR assembly decay heat data are used, respectively. Due to the small size of the measurement dataset, we propose: (i) to use the method of multiple runs (ii) to generate and use synthetic data, as large dataset which has similar statistical characteristics as the original dataset. Three ML models are developed based on Gaussian process (GP), support vector machines (SVM) and neural networks (NN), with four inputs including the fuel assembly averaged enrichment, assembly averaged burnup, initial heavy metal mass, and cooling time after discharge. The outcomes of this work are (i) development of ML models which predict LWR fuel assembly decay heat from the four inputs (ii) generation and application of synthetic data which improves the performance of the ML models (iii) uncertainty analysis of the ML models and their predictions.

• Herbal Formula Science
• /
• v.29 no.4
• /
• pp.285-295
• /
• 2021

Objective : This systematic review was conducted to investigate the effect of herbal medicine on Helicobacter pylori(H. pylori) infection compared to amoxicillin included synthetic drugs. Methods : Relevant randomized controlled trials(RCTs) which were published prior to December 26, 2020, were collected using PubMED database. Risk of bias evaluation and data extraction were done independently by two reviewers, and the third reviewer reassessed mismatching parts. Results : Two RCTs testing two different herbal medicines against synthetic drugs solitary treatment or synthetic drugs with placebo for herbal medicine were included. One study reported that there was no significant difference between the eradication rate of synthetic drugs and the herbal medicine. The other study did not report the eradication rate of the herbal medicine. One study reported histologic severity, the other reported dyspepsia score as efficacy indicators. There was no adverse event reported in all studies. However, the number of included RCTs was too small, the quality of reported data was not enough to verify efficacy of herbal medicine, and there were some methodological problems. Conclusion : It was difficult to conclude that solitary treatment of herbal medicine was as effective as amoxicillin included synthetic drugs for H. pylori infection.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.5
• /
• pp.353-363
• /
• 2012

As the distributed simulation system has been developed in the defense modeling and simulation, interoperability among heterogeneous simulators is essential to achieve the objectives of the distributed simulation system. However, the interoperability has been a problem to synchronize synthetic environment among many heterogeneous simulators that utilize different environmental data formats. SEDRIS (Synthetic Environmental Data Representation and Interchange Specification) was initiated to solve the problem of the interoperability. SEDRIS provides a standard mechanism to promote data interchange and reusability, and it represents all environmental things from ocean, land and atmosphere to the universe. Although SEDRIS provides various advantages as a standard environmental data format, applying of SEDRIS has been hindered by its broadness and complexity. The main objective of this paper is to explain application development process to improve SEDRIS usability for the atmosphere and ocean. The developed application of this paper provides auto-generation of SEDRIS data with minimum user's input, and the auto-generated SEDRIS data is also visualized and verified by the application.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.5
• /
• pp.475-481
• /
• 2021

The on-board processor of satellite Synthetic Aperture Radar(SAR) digitizes the back-scattered echoes and transmits them to the ground. As satellite SAR image of various operating conditions including broadband and high resolution is required, an enormous amount of SAR data is generated. Decimation filter is used for data compression to improve the transmission efficiency of these data. Decimation filter is implemented with the FIR(Finite Impulse Response) filter and here, the decimation ratio and tap length are constrained by resource requirements of FPGA used for implementation. This paper suggests to use a non-integer ratio decimation filter in order to optimize the data transmission efficiency. Also, it proposes a filter design method that remarkably reduces the resource constraints of the FPGA in-use via applying a polyphase filter structure. The required resources for implementing the proposed filter is analysed in this paper.

• Journal of Aerospace System Engineering
• /
• v.13 no.6
• /
• pp.1-8
• /
• 2019

The synthetic turbulence generation model for inlet boundary conditions of subsonic Backward Facing Step (BFS) was investigated. The average u-velocity and Reynolds stress at inlet boundary follows experimental data. Synthetic Eddy Method (SEM), random noise, and uniform flow conditions were implemented relative to the synthetic turbulence generation method. A three dimensional Large Eddy Simulation (LES) was applied for turbulent flow simulation. Turbulent and mean flow characteristics such as flow reattachment length, velocity profiles, and Reynolds stress profiles of BFS were compared with respect to the turbulent effects.

• Journal of the Semiconductor & Display Technology
• /
• v.17 no.2
• /
• pp.26-31
• /
• 2018

In this paper, surface defects and typical illumination mechanisms for steel plates are analyzed, and then optimum illumination mechanism is selected using discrete wavelet transform (DWT) synthetic images and discriminant measure (DM). The DWT synthetic images are generated using component images decomposed by Haar wavelet transform filter. The best synthetic image according to surface defects is determined using signal to noise ratio (SNR). The optimum illumination mechanism is selected by applying discriminant measure (DM) to the best synthetic images. The DM is applied using the tenengrad-euclidian function. The DM is evaluated as the degree of contrast using the defect boundary information. The performance of the optimum illumination mechanism is verified by quantitative data and intuitive image looks.

• Korean Journal of Remote Sensing
• /
• v.21 no.2
• /
• pp.163-171
• /
• 2005

I have developed a processor for synthetic aperture radar (SAR) raw data compression using range-doppler algorithm for educational purpose. The program realized a generic SAR focusing algorithm so that it can deal with any SAR system if the specification is known. It can run efficiently on a low-cost computer by selecting minimum size out of a whole dataset, and can produce intermediate images during the process. Especially, the program is designed for educational purpose in such a way that Doppler centroid and azimuth ambiguity can be determined graphically by the user. By distributing the source code and the algorithm to public, I intend to maximize the educational effect on understanding and utilizing SAR data. This paper introduces the principle of SAR focusing algorithm embedded on the eSAR processor and shows an example of data processing using ERS-1 raw data.