• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.152-159
• /
• 2024

Cracks in structural materials present a critical challenge to infrastructure safety and long-term durability. Timely and precise crack detection is essential for proactive maintenance and the prevention of catastrophic structural failures. This study introduces an innovative approach to tackle this issue using U-Net deep learning architecture. The primary objective of the intended research is to explore the potential of U-Net in enhancing the precision and efficiency of crack detection across various concrete crack detection under various environmental conditions. Commencing with the assembling by a comprehensive dataset featuring diverse images of concrete cracks, optimizing crack visibility and facilitating feature extraction through advanced image processing techniques. A wide range of concrete crack images were collected and used advanced techniques to enhance their visibility. The U-Net model, well recognized for its proficiency in image segmentation tasks, is implemented to achieve precise segmentation and localization of concrete cracks. In terms of accuracy, our research attests to a substantial advancement in automated of 95% across all tested concrete materials, surpassing traditional manual inspection methods. The accuracy extends to detecting cracks of varying sizes, orientations, and challenging lighting conditions, underlining the systems robustness and reliability. The reliability of the proposed model is measured using performance metrics such as, precision(93%), Recall(96%), and F1-score(94%). For validation, the model was tested on a different set of data and confirmed an accuracy of 94%. The results shows that the system consistently performs well, even with different concrete types and lighting conditions. With real-time monitoring capabilities, the system ensures the prompt detection of cracks as they emerge, holding significant potential for reducing risks associated with structural damage and achieving substantial cost savings.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.34 no.2
• /
• pp.135-140
• /
• 2014

Pressure vessels are used in various industrial fields. If a defect occurs on the inner or outer surface of a pressure vessel, it may cause a massive accident. A defect on the outer surface can be detected by visual inspection. However, a defect on the inner surface is generally impossible to detect with visual inspection. Nondestructive testing can be used to detect this type of defect. Laser speckle shearing interferometry is one nondestructive testing method that can optically detect a defect; its advantages include noncontact, full field, and real time inspection. This study evaluated the detectable size for an internal defect of a pressure vessel. The material of the pressure vessel was ASTM A53 Gr.B. The internal defect was detected when the pressure vessel was loaded by internal pressure controlled by a pneumatic system. The internal pressure was controlled from 0.2 MPa to 0.6 MPa in increments of 0.2 MPa. The results confirmed that an internal defect with a 25 % defect depth could be detected even at 0.2 MPa pressure variation.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.3
• /
• pp.103-110
• /
• 2016

Recently, structural health monitoring system(SHMS) has been appled cable bridges as the effective maintenance tool and the management threshold is considered to assess the status of the bridge in SHMS. The threshold is generally determined by the allowable limit based on design specification because there is no method and standard for threshold calculation. In case of the conventional thresholds, it is difficult to recognize the event, abnormal behavior and gradual damage within the threshold. Therefore, this study reviewed the problem of previous methods and proposed the advanced methodologies based on probabilistic approach for threshold calculation which can be applied to practice work. Gumbel distribution is adopted in order to calculate the threshold for caution and warning states considering the expectations for return periods of 50 and 100 years. The thresholds were individually determined for measurement data and data variation to detect the various abnormal behaviors within allowable range. Finally, it has confirmed that the thresholds by the proposed method is detectable the abnormal behavior of real-time measuring data from SHMS.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_1
• /
• pp.599-605
• /
• 2017

Accurate cloud discrimination in satellite images strongly affects accuracy of remotely sensed parameter produced using it. Especially, cloud contaminated pixel over ocean is one of the major error factors such as Sea Surface Temperature (SST), ocean color, and chlorophyll-a retrievals,so accurate cloud detection is essential process and it can lead to understand ocean circulation. However, static threshold method using real-time algorithm such as Moderate Resolution Imaging Spectroradiometer (MODIS), Advanced Himawari Imager (AHI) can't fully explained reflectance variability over ocean as a function of relative positions between the sun - sea surface - satellite. In this paper, we assembled a reflectance spectral library as a function of Solar Zenith Angle (SZA) and Viewing Zenith Angle (VZA) from ocean surface reflectance with clear sky condition of Advanced Himawari Imager (AHI) identified by NOAA's cloud products and spectral library is used for applying the Dynamic Time Warping (DTW) to detect cloud pixels. We compared qualitatively between AHI cloud property and our results and it showed that AHI cloud property had general tendency toward overestimation and wrongly detected clear as unknown at high SZA. We validated by visual inspection with coincident imagery and it is generally appropriate.

• Journal of the Korea Safety Management & Science
• /
• v.19 no.1
• /
• pp.15-25
• /
• 2017

The findings were summarized as follows. The safety check by manufacturer showed that 6 of 13 companies are over the average occurrence of defects. It was expected that there would be a difference between manufacturing technology capability and production system of each manufacturer. Consequently, manufacturers should institutionally improve and strengthen certification items for the upward standardization of safety certification before factory. Second, the safety check by year showed that the results of this study accord with those of previous studies on defect time. Consequently, manufacturers should classify the 3-year-old equipment for vehicle-mounted MEWP into a special check subject to do a nondestructive test according to proven results, and also reflect the test in a safety test system to do regular preventive activities of equipment defects. Third, the safety check by part showed that the boom and outrigger parts of vehicle-mounted MEWP have the most defects. Stress concentration resulted in defects as the boom part was most frequently operated in the structural parts for a real work. To prevent this, it is suitable to improve the hardness of boom materials. The outrigger part needs improvement in safety devices with materials. As an outrigger supports the overturning moment of equipment, it is most affected by its load based on the operating radius, resulting in fatigue crack.

• Journal of Internet Computing and Services
• /
• v.24 no.5
• /
• pp.9-16
• /
• 2023

Recently, there is an increasing demand to quickly identify changes in the field by applying Digital-twin for on-site management control and analyzing changes in real-time data transmitted. When technologies such as Digital-twin are applied, early problem response and quick response to situations are possible. However, in order to maximize the advantages of digital twin technology, a method of collecting and managing sensor data that collects field information at an appropriate period is required. need. In addition, it is necessary to consider how sensor information is transmitted to each management domain and how to identify and respond to abnormal situations so that it can be visualized for each management purpose. In this paper, we propose a method for collecting heterogeneous sensor information and related system configuration for on-site inspection management based on digital twin.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.16 no.1
• /
• pp.108-114
• /
• 2012

Purpose: Immunosorbent assay most commonly used in the laboratory and commercial third-party quality control material is a substance that provided by BIO-RAD. However, in Reference Sheet by radioimmunoassay test kit or a measuring device for the mention of Acceptable Range is somewhat lacking. Radioimmunoassay for the inspection of test results by setting Acceptable Range to increase the objectivity of the recommendations on the data accumulated by the manufacturer listed in the Reference Sheet is to be issued. Materials and Methods : In our hospital since 2009 partially BIO-RAD using quality control materials in 2011, excluding certain items, some items were most of the BIO-RAD third-party quality control materials are used. Thus, internal quality control data accumulated BIO-RAD's Unity Real Time program by using the items were measured. Results : BIO-RAD using quality control material items were about 50 of the 20 Point Data averages, standard deviations, variation coefficients were calculated to measure the Acceptable Range of kit, automated immunoassay attributed Roche Elecsys / E170 / cobas e Systems the measures and compared. Conclusions : BIO-RAD QC materials commonly used hospital and peer group by setting the measurement kit, suitable for laboratory equipment for radioimmunoassay of Acceptable Range manufacturer recommendation to increase the objectivity of the test results by national and international recognition for radioimmunoassay should seek to increase.

• Tunnel and Underground Space
• /
• v.33 no.6
• /
• pp.534-546
• /
• 2023

If the disc cutter is excessively worn or damaged, it becomes incapable of rotating and efficiently cutting rockmass. Therefore, it is important to appropriately manage the replacement cycle of the disc cutter based on its degree of wear. In general, the replacement cycle is determined based on the results of manual inspection. However, the manual measurements has issues related to worker safety and may lead to inaccurate measurement results. For these reasons, some foreign countries are developing the real-time measurement system of disc cutter wear by using different sensors. The ultrasonic sensors, eddy current sensors, magnetic sensors, and others are utilized for measuring the wear amount of disc cutters. In this study, the applicability of eddy current sensors for real-time measurement of wear amount in TBM disc cutters was evaluated. The distance measurement accuracy of the eddy current sensor was assessed through laboratory tests. In particular, the accuracy of eddy-current sensor was evaluated in various environmental conditions within the cutterhead chamber. In addition, the measurement accuracy of the eddy current sensor was validated using a 17-inch disc cutter.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.475-478
• /
• 2011

driver for safety always check the status of their vehicle, and it is essential to understand. But if you want to know the status of the driver of the vehicle in specialist referral time and money because it costs the operator shall be paid. Today's rapidly changing IT technology with the development of the various features of your phone to check the status of the vehicle was able to do. However, the car's existing phone system, car diagnostic expertise must be learned because it will reveal the status of the vehicle do not have the expertise to not highlight the need for diagnostic. To reflect these points in smartphone users to easily use their own vehicles at a time to determine the status of a system that is required. In this paper, OBD-II protocol conversion WiFi OBD-II connector, retrieving information from the driver of the vehicle replacement cycle of consumables required vehicle inspection, vehicle problems in real-time diagnostic information to the user ease of use shows the IOS implementation in the automotive supply was implemented based on the smartphone.

• The Korean Journal of Physiology and Pharmacology
• /
• v.24 no.1
• /
• pp.89-99
• /
• 2020

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.