• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.189-200
• /
• 2022

Ground subsidence on urban roads is a social issue that can lead to human and property damages. Therefore, it is crucial to detect underground cavities in advance and repair them. Underground cavity detection is mainly performed using ground penetrating radar (GPR) surveys. This process is time-consuming, as a massive amount of GPR data needs to be interpreted, and the results vary depending on the skills and subjectivity of experts. To address these problems, researchers have studied automation and quantification techniques for GPR data interpretation, and recent studies have focused on deep learning-based interpretation techniques. In this study, we described a hyperbolic event detection process based on deep learning for GPR data interpretation. To demonstrate this process, we implemented a series of algorithms introduced in the preexisting research step by step. First, a deep learning-based YOLOv3 object detection model was applied to automatically detect hyperbolic signals. Subsequently, only hyperbolic signals were extracted using the column-connection clustering (C3) algorithm. Finally, the horizontal locations of the underground cavities were determined using regression analysis. The hyperbolic event detection using the YOLOv3 object detection technique achieved 84% precision and a recall score of 92% based on AP50. The predicted horizontal locations of the four underground cavities were approximately 0.12 ~ 0.36 m away from their actual locations. Thus, we confirmed that the existing deep learning-based interpretation technique is reliable with regard to detecting the hyperbolic patterns indicating underground cavities.

• Journal of Multimedia Information System
• /
• v.6 no.2
• /
• pp.81-86
• /
• 2019

Deep learning is emerging as one of the best tool in processing data related to medical imaging. In our research work, we have proposed a deep learning based framework CNN (Convolutional Neural Network) for the classification of dysplastic tissue images. The CNN has classified the given images into 4 different classes namely normal tissue, mild dysplastic tissue, moderate dysplastic tissue and severe dysplastic tissue. The dataset under taken for the study consists of 672 tissue images of epithelial squamous layer of oral cavity captured out of the biopsy samples of 52 patients. After applying the data pre-processing and augmentation on the given dataset, 2688 images were created. Further, these 2688 images were classified into 4 categories with the help of expert Oral Pathologist. The classified data was supplied to the convolutional neural network for training and testing of the proposed framework. It has been observed that training data shows 91.65% accuracy whereas the testing data achieves 89.3% accuracy. The results produced by our proposed framework are also tested and validated by comparing the manual results produced by the medical experts working in this area.

• Restorative Dentistry and Endodontics
• /
• v.23 no.1
• /
• pp.296-303
• /
• 1998

The purpose of this study was to assess the effects of cavity designs on the marginal leakage of class 5 glass ionomer restorations. The five cavity designs were as follows ; notch shape(A group), notch shape with groove(B group), combined lesion(C group), combined lesion with groove and deep chamfer margin(D group) and combined lesion with groove and shoulder(E group), and each design had 10 cavities. After the cavities were restored with GIC, they were immersed in 0.5% basic fuchsin solution for 6 hours. The specimens were washed thoroughly and sectioned longitudinally in a buccolingual direction through the center of the restorations. The degree of marginal leakage was measured as the extent of dye penetration under the stereoscope. The results of this study were as follows 1. The enamel margins of all groups showed lesser leakage than dentin/cementum margins(p<0.05). 2. The combined lesion(C group) showed more leakage than notch shape(A group), but there was no siginificant difference(p>0.05). 3. In the notch shape, there was no influence on the marginal leakage by the groove preparation. 4. In the combined lesion, marginal leakage was decreased by the groove preparation and marginal modification.

• Journal of IKEEE
• /
• v.18 no.1
• /
• pp.19-24
• /
• 2014

We report stable, wideband, flat-topped, 10 GHz optical frequency comb generation from a semiconductor-based mode-locked ring laser with an intra-cavity high finesse Fabry-Perot etalon. We demonstrate a stable 10 GHz comb with greater than 200 lines within a spectral power variation below 1 dB, which is the largest value obtained from a similar mode-locked laser in our knowledge. Greater than 20 dB of the spectral peak to deep ratio at 0.02 nm resolution, ~92 femtosecond timing jitter over 1 kHz to 1 MHz range, and non-averaged time traces of pulses confirm very stable optical frequency comb lines.

• Proceedings of the KIEE Conference
• /
• 2001.07c
• /
• pp.1890-1892
• /
• 2001

This paper described on the fabrication of microstructures by DRIE(Deep Reactive Ion Etching). SOI(Si-on-insulator) electric devices with buried cavities are fabricated by SDB technology and electrochemical etch-stop. The cavity was fabricated the upper handling wafer by Si anisotropic etch technique. SDB process was performed to seal the fabricated cavity under vacuum condition at -750 mm Hg. In the SDB process, captured air and moisture inside of the cavities were removed by making channels towards outside. After annealing(1000$^{\circ}C$, 60 min.), the SDB SOI structure was thinned by electrochemical etch-stop. Finally, it was fabricated microstructures by DRIE as well as a accurate thickness control and a good flatness.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.193-197
• /
• 2008

Array of micro-dimple on polyoxymethylene (POM) surface was fabricated using Q-switched Nd:YAG laser and its characteristics were studied in terms of heat affected zone (HAZ), dimple geometry, and the effect of specimen surface roughness. Process parameters such as lamp current, process time, and the stream of air in order to minimize HAZ and flow of molten polymer into cavity were extensively studied in this work. Dimple geometry was further investigated by 3-D optical microscopy to provide deep insight into morphological analysis near the dimples. This paper also presents the applicapability of micro-dimples in polymeric tribological system, such as a thrust bearing. Micro-dimples were expected to provide low coefficient of friction and enhanced lubricity at the sliding interface.

• Proceedings of the KACD Conference
• /
• 2001.11a
• /
• pp.584.1-584
• /
• 2001

The purpose of this study was to evaluate the effects of various cavity configuration on bond strength and micro leakage of composite restoration. The specimens were prepared as followed; For control group(C=1), bovine teeth were wet-ground to expose flat dentin surface and in experimental groups, cylindrical cavities, same all2mm deep and 6mm in diameter(C=2.3), 4mm(C=3.0), and 3mm(C=3.7) were prepared.(omitted)

• The Journal of the Korean dental association
• /
• v.57 no.7
• /
• pp.380-391
• /
• 2019

In recent years, with the introduction of various restorative materials, restorations using CAD/CAM equipment have been increasing in the esthetic dentistry. The critical steps in the fabrication of indirect restorations with CAD/CAM equipment are proper cavity preparation and making accurate impressions. The process of tooth preparation for CAD/CAM restoration should include a mechanical understanding of milling. In addition, during tooth preparation, the clinician should be familiar with additional equipment and techniques for obtaining the convenience. In order to obtain an accurate oral scan, the clinician should understand the limitations of the oral scan and be skilled at techniques for obtaining a successful image when making oral scans. This article focused clinical guidelines for the preparation of CAD/CAM restorations and introduced clinical methods for making successful impression of oral scans in narrow and deep tooth cavity areas.

• Explosives and Blasting
• /
• v.42 no.2
• /
• pp.29-41
• /
• 2024

The increasing demand for metallic minerals due to global growth and the continued exploitation of near-surface minerals requires safe and efficient ways to mine ores present in deep mines. In deep mines, stresses concentrated around the cavity increase, which can lead to problems such as induced seismicity and rockbursts. In addition, the transfer of energy from blasting to deeply located faults can cause fault slip, which can lead to earthquakes, and controlling these events is key to deep mining methods. In this technical report, we will introduce the Underhand Closed Bench (UCB) mining method, which can control possible accidents and increase productivity when mining in deep mines.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2003.11a
• /
• pp.299-304
• /
• 2003

Borehole radar is one kind of GPR, but it can be used in deep boreholes, and it has many advantages compared with low frequency borehole EM tools, and surface GPR. We have developed various techniques on borehole radar for environment study. The hardware development includes broadband radar system with the functions of polarimetry and inteferometry. By using these systems, we tested the measurements to applications such as subsurface fracture characterization, subsurface cavity detections. In this paper, we will describe the advantages of the advanced radar technology for environment studies, and show some experiment results.