• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.26 no.6_spc
• /
• pp.651-659
• /
• 2016

Condition monitoring (CM) encounters a large data problem due to sensors that measure vibration data with a continuous, and sometimes, high sampling rate. In this study, compressive sensing approaches for condition monitoring are proposed to demonstrate the efficiency in handling a large amount of data and to improve the damage detection capability of the current condition monitoring process. Compressive sensing is a novel sensing/sampling paradigm that takes much fewer samples compared to traditional sampling methods. For the experiments a built-in rotating system was used and all data were compressively sampled to obtain compressed data. Optimal signal features were then selected without the reconstruction process and were used to detect and classify damage. The experimental results show that the proposed method could improve the data processing speed and the accuracy of condition monitoring of rotating systems.

• Journal of the Korean earth science society
• /
• v.42 no.4
• /
• pp.445-458
• /
• 2021

Gravity Recovery and Climate Experiment (GRACE) gravimeter satellites observed the Earth gravity field with unprecedented accuracy since 2002. After the termination of GRACE mission, GRACE Follow-on (GFO) satellites successively observe global gravity field, but there is missing period between GRACE and GFO about one year. Many previous studies estimated terrestrial water storage (TWS) changes using hydrological models, vertical displacements from global navigation satellite system observations, altimetry, and satellite laser ranging for a continuity of GRACE and GFO data. Recently, in order to predict TWS changes, various machine learning methods are developed such as artificial neural network and multi-linear regression. Previous studies used hydrological and climate data simultaneously as input data of the learning process. Further, they excluded linear trends in input data and GRACE/GFO data because the trend components obtained from GRACE/GFO data were assumed to be the same for other periods. However, hydrological models include high uncertainties, and observational period of GRACE/GFO is not long enough to estimate reliable TWS trends. In this study, we used convolutional neural networks (CNN) method incorporating only climate data set (temperature, evaporation, and precipitation) to predict TWS variations in the missing period of GRACE/GFO. We also make CNN model learn the linear trend of GRACE/GFO data. In most river basins considered in this study, our CNN model successfully predicts seasonal and long-term variations of TWS change.

• Journal of Ocean Engineering and Technology
• /
• v.32 no.6
• /
• pp.511-517
• /
• 2018

This paper proposes a vision-based underwater laser scanner with separate structures for an underwater camera and a line laser projector. Because the two devices can be adaptively placed regardless of the features of the unmanned underwater vehicle (UUV), the scanner has significant advantages in relation to its availability and flexibility. Position calibration between the underwater camera and laser projector guarantees a 3D measuring performance with high accuracy. To verify the proposed underwater laser scanner, a test-bed system was manufactured, which consisted of the laser projector, camera, Pan&Tilt, and Attitude and Heading Reference System (AHRS). A camera-laser calibration test and simple 3D reconstruction test were performed in a water tank and the experimental results are reported.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2381-2398
• /
• 2024

The construction of virtual indoor spaces is crucial for the development of metaverses, virtual production, and other 3D content domains. Traditional methods for creating these spaces are often cost-prohibitive and labor-intensive. To address these challenges, we present a pipeline for generating digital twins of real indoor environments from RGB-D camera-scanned data. Our pipeline synergizes space structure estimation, 3D object detection, and the inpainting of missing areas, utilizing deep learning technologies to automate the creation process. Specifically, we apply deep learning models for object recognition and area inpainting, significantly enhancing the accuracy and efficiency of virtual space construction. Our approach minimizes manual labor and reduces costs, paving the way for the creation of metaverse spaces that closely mimic real-world environments. Experimental results demonstrate the effectiveness of our deep learning applications in overcoming traditional obstacles in digital twin creation, offering high-fidelity digital replicas of indoor spaces. This advancement opens for immersive and realistic virtual content creation, showcasing the potential of deep learning in the field of virtual space construction.

• Imaging Science in Dentistry
• /
• v.40 no.4
• /
• pp.171-178
• /
• 2010

Purpose : The purpose of this study is to develop a surgical stent for dental implant procedure that can be easily applied and affordable by using cone beam computerized tomography (CBCT). Materials and Methods : Aluminum, Teflon-PFA (perfluoroalkoxy), and acetal (polyoxymethylene plastic) were selected as materials for the surgical stent. Among these three materials, the appropriate material was chosen using the CBCT images. The surgical stent, which could be easily placed into an oral cavity, was designed with chosen material. CBCT images of the new surgical stent on mandible were obtained using Alphard-3030 dental CT system (Asahi Roentgen Co., Ltd., Kyoto, Japan). The point of insertion was prescribed on the surgical stent with the multiplanar reconstruction software of OnDemand3D (CyberMed Inc., Seoul, Korea). Guide holes were made at the point of insertion on the surgical stent using newly designed guide jig. CBCT scans was taken for the second time to verify the accuracy of the newly designed surgical stent. Results : Teflon-PFA showed radiologically excellent image characteristics for the surgical stent. High accuracy and reproducibility of implantation were confirmed with the surgical stent. Conclusion : The newly designed surgical stent can lead to the accurate implantation and achieve the clinically predictable result.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.39 no.2
• /
• pp.139-144
• /
• 2012

The purpose of this study was to evaluate the accuracy and reproducibility of measuring the size of unerupted permanent tooth via cone beam computed tomography(CBCT). Ten children were scanned with dental CBCT, and 3-dimensional reconstruction of the dentitions were generated CBCT. Mesio-distal dimension and buccolingual dimension of the teeth were made directly on the model with a high-precision digitalcaliper and on the CBCT by using three-dimensional dental imaging software. Reliability and accuracy were assessed by using intraclass correlation and paired $t$-tests. ($p$ <0.05) The results were as follows : 1. Intraclass correlations were above 0.9 for Both the CBCT and the model measurements, showinghigh reliability. 2. Although there were high correlation values(r=0.91) between CBCT and model messurement methods, comparisons between the CBCT and model messurement methods showed a statistically significant difference($p$ <0.05). 3. The CBCT measurements tended to slightly underestimate by 0.2 mm. But, the systematic difference of CBCT measurements were clinically acceptable Therefore, CBCT measurement method can be used to measure the size of unerupted teeth in a sufficiently accurate way.

• Advances in aircraft and spacecraft science
• /
• v.4 no.3
• /
• pp.281-296
• /
• 2017

The axial-injection end-burning hybrid rocket proposed twenty years ago by the authors recently recaptured the attention of researchers for its virtues such as no ${\zeta}$ (oxidizer to fuel mass ratio) shift during firing and good throttling characteristics. This paper is the first report verifying these virtues using a laboratory scale motor. There are several requirements for realizing this type of hybrid rocket: 1) high fuel filling rate for obtaining an optimal ${\zeta}$; 2) small port intervals for increasing port merging rate; 3) ports arrayed across the entire fuel section. Because these requirements could not be satisfied by common manufacturing methods, no previous researchers have conducted experiments with this kind of hybrid rocket. Recent advances in high accuracy 3D printing now allow for fuel to be produced that meets these three requirements. The fuel grains used in this study were produced by a high precision light polymerized 3D printer. Each grain consisted of an array of 0.3 mm diameter ports for a fuel filling rate of 98% .The authors conducted several firing tests with various oxidizer mass flow rates and chamber pressures, and analysed the results, including ${\zeta}$ history, using a new reconstruction technique. The results show that ${\zeta}$ remains almost constant throughout tests of varying oxidizer mass flow rates, and that regression rate in the axial direction is a nearly linear function of chamber pressure with a pressure exponent of 0.996.

• Korean Journal of Remote Sensing
• /
• v.38 no.2
• /
• pp.199-213
• /
• 2022

The unsupervised domain adaptation can solve the impractical issue of repeatedly collecting high-quality training data every year for annual crop classification. This study evaluates the applicability of deep learning-based unsupervised domain adaptation models for crop classification. Three unsupervised domain adaptation models including a deep adaptation network (DAN), a deep reconstruction-classification network, and a domain adversarial neural network (DANN) are quantitatively compared via a crop classification experiment using unmanned aerial vehicle images in Hapcheon-gun and Changnyeong-gun, the major garlic and onion cultivation areas in Korea. As source baseline and target baseline models, convolutional neural networks (CNNs) are additionally applied to evaluate the classification performance of the unsupervised domain adaptation models. The three unsupervised domain adaptation models outperformed the source baseline CNN, but the different classification performances were observed depending on the degree of inconsistency between data distributions in source and target images. The classification accuracy of DAN was higher than that of the other two models when the inconsistency between source and target images was low, whereas DANN has the best classification performance when the inconsistency between source and target images was high. Therefore, the extent to which data distributions of the source and target images match should be considered to select the best unsupervised domain adaptation model to generate reliable classification results.

• Geophysics and Geophysical Exploration
• /
• v.14 no.4
• /
• pp.335-341
• /
• 2011

Most sensing and instrumentation systems should have very higher sampling rate than required data rate not to miss important information. This means that the system can be inefficient in some cases. This paper introduces two new research areas about information acquisition with high accuracy from less number of sampled data. One is Compressed Sensing technology (which obtains original information with as little samples as possible) and the other is Super-Resolution technology (which gains very high-resolution information from restrictively sampled data). This paper explains fundamental theories and reconstruction algorithms of compressed sensing technology and describes several applications to geophysical exploration. In addition, this paper explains the fundamentals of super-resolution technology and introduces recent research results and its applications, e.g. FRI (Finite Rate of Innovation) and LIMS (Least-squares based Iterative Multipath Super-resolution). In conclusion, this paper discusses how these technologies can be used in geophysical exploration systems.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.2
• /
• pp.58-65
• /
• 1997

A new automatic classification method with high and stable accuracy for time-series image data is presented in this paper. This method is based on prior condition that a classified map of the target area already exists, or at least one of the time-series image data had been classified. The classified map is used as a reference map to specify training areas of classification categories. The new automatic classification method consists of five steps, i.e., extraction of training data using reference map, detection of changed pixels based upon the homogeneity of training data, clustering of changed pixels, reconstruction of training data, and classification as like maximum likelihood classifier. In order to evaluate the performance of this method qualitatively, four time-series Landsat TM image data were classified by using this method and a conventional method which needs a skilled operator. As a results, we could get classified maps with high reliability and fast throughput, without a skilled operator.