• The Korean Journal of Nuclear Medicine Technology
• /
• v.23 no.2
• /
• pp.43-50
• /
• 2019

Purpose The principle of nuclear medicine test is divided into two main categories: competition(radioimmunoassay, RIA) and noncompetitive reaction(Immunoradiometric assay, IRMA). It is known that the curve fitting method, which is commonly used in inspection field, uses Spline interpolation in RIA method and Linear interpolation method in IRMA method. Among them, the insulin test using the IRMA test showed a significant difference, especially at low concentrations, despite the same algorithm of linear interpolation between fully automated radio immunoassay analyzers. In this study, we aim to obtain results from applying two different of algorithm using fully automated radio immunoassay analyzers including Gamma pro, Gamma 10, Cobra, and SR300. Materials and Methods A total of 30 test samples were selected for the test of TSH, ferritin, C-peptide, and insulin serum levels. Test was performed by IRMA method. We compared the difference in the results of applying the linear interpolation method and the spline interpolation method to Gamma Pro, Gamma 10, Cobra, and SR300 equipment. Results Two-way ANOVA was used for statistical analysis. The significance level was applied as P <0.05. The results of TSH, ferritin, C-peptide, and insulin tests were compared between the fully automated radio immunoassay analyzers. There was a significant difference between ferritin, C-peptide, and insulin serum levels(P<0.001). TSH didn't show any significant different between the devices(P=0.29). In the difference between linear and spline interpolation, there was no significant difference between insulin test(P=0.08), TSH test(P=0.81), and Ferritin test(P=0.06). However, C-peptide test showed a significant difference(P=0.03). Especially, the insulin test showed significant difference in lower ranges. As a result of comparing and analyzing the difference between the two interpolation methods, the devices in the low concentration group showed significant difference(P<0.001). Conclusion In case of new equipment in the laboratory it is necessary to recognize that there is a difference in the curve fitting method for each automated radio immunoassay analyzers in the low concentration area when the principle of inspection is IRMA method.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.6
• /
• pp.38-44
• /
• 2019

A Convolution Neural Network(CNN) model was utilized to detect surface cracks in asphalt concrete pavements. The CNN used for this study consists of five layers with 3×3 convolution filter and 2×2 pooling kernel. Pavement surface crack images collected by automated road surveying equipment was used for the training and testing of the CNN. The performance of the CNN was evaluated using the accuracy, precision, recall, missing rate, and over rate of the surface crack detection. The CNN trained with the largest amount of data shows more than 96.6% of the accuracy, precision, and recall as well as less than 3.4% of the missing rate and the over rate.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.34 no.1
• /
• pp.60-67
• /
• 2014

In this study, an automated cable non-destructive testing(NDT) system was developed to monitor the steel cables that are a core component of cable-stayed bridges. The magnetic flux leakage(MFL) method, which is suitable for ferromagnetic continuum structures and has been verified in previous studies, was applied to the cable inspection. A multi-channel MFL sensor head was fabricated using hall sensors and permanent magnets. A wheel-based cable climbing robot was fabricated to improve the accessibility to the cables, and operating software was developed to monitor the MFL-based NDT research and control the climbing robot. Remote data transmission and robot control were realized by applying wireless LAN communication. Finally, the developed element techniques were integrated into an MFL-based cable NDT system, and the field applicability of this system was verified through a field test at Seohae Bridge, which is a typical cable-stayed bridge currently in operation.

• Journal of Advanced Navigation Technology
• /
• v.25 no.5
• /
• pp.377-383
• /
• 2021

In this paper, we describe the case of automatic test equipment development for hardware verification of stores management computer mounted on aircraft. Recently, the required functions of aircraft have been diversified and the related technologies of avionics equipment have developed, and the types and quantity of interfaces required for avionics equipment have increased. In addition to the existing old stores, the stores management computer also needs to control the interface in large quantities as the requirements for the new stores are added. For this reason, the time and manpower required for the inspection of avionic equipment are also increasing, and if the test process of avionic equipment can be automated and unmanned, more efficient inspection system operation will be possible. Therefore, this paper introduces the case of designing test software and test scenario to automate the structural design contents and verification process of test equipment required for the verification of hardware function of stores management computer.

• Nuclear Engineering and Technology
• /
• v.53 no.2
• /
• pp.657-665
• /
• 2021

Nondestructive evaluation methods play an important role in ensuring component integrity and safety in many industries. Operator fatigue can play a critical role in the reliability of such methods. This is important for inspecting high value assets or assets with a high consequence of failure, such as aerospace and nuclear components. Recent advances in convolution neural networks can support and automate these inspection efforts. This paper proposes using residual neural networks (ResNets) for real-time detection of corrosion, including iron oxide discoloration, pitting and stress corrosion cracking, in dry storage stainless steel canisters housing used nuclear fuel. The proposed approach crops nuclear canister images into smaller tiles, trains a ResNet on these tiles, and classifies images as corroded or intact using the per-image count of tiles predicted as corroded by the ResNet. The results demonstrate that such a deep learning approach allows to detect the locus of corrosion via smaller tiles, and at the same time to infer with high accuracy whether an image comes from a corroded canister. Thereby, the proposed approach holds promise to automate and speed up nuclear fuel canister inspections, to minimize inspection costs, and to partially replace human-conducted onsite inspections, thus reducing radiation doses to personnel.

• Korean Journal of Construction Engineering and Management
• /
• v.25 no.2
• /
• pp.36-44
• /
• 2024

This paper presents the application of Convolutional Neural Networks (CNNs) and Region of Interest (ROI) techniques for concrete crack analysis. Surfaces of concrete structures, such as beams, etc., are exposed to fatigue stress and cyclic loads, typically resulting in the initiation of cracks at a microscopic level on the structure's surface. Early detection enables preventative measures to mitigate potential damage and failures. Conventional manual inspections often yield subpar results, especially for large-scale infrastructure where access is challenging and detecting cracks can be difficult. This paper presents data collection, edge segmentation and ROI techniques application, and analysis of concrete cracks using Convolutional Neural Networks. This paper aims to achieve the following objectives: Firstly, achieving improved accuracy in crack detection using image-based technology compared to traditional manual inspection methods. Secondly, developing an algorithm that utilizes enhanced Sobel edge segmentation and ROI techniques. The algorithm provides automated crack detection capabilities for non-destructive testing.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2003.05a
• /
• pp.270-273
• /
• 2003

The strain measurement of the panel in the sheet metal forming is essential work which provides experimental data needed to die design, process design, and product inspection. To measure efficiently the complex geometry strain, the 3-dimensional automative strain measurement system, which has high accuracy in theory, but has some 3∼5% errors in practice, is often used. The object of this study is to develop the error compensation technology to eliminate the strain, errors resulted when formed panels are measured using an automated strain measurement system. To achieve the study object, the position error calibration method correcting coordinates of the grid node recognized by a camera using error functions is suggested. Then the position errors were found by calculating the difference in the position of the cube node between real coordinates and measured coordinates in toms of node coordinates and the error calibration equations were derived by regressing the position errors. In order to show the validation of the suggested position error calibration method, finite element analysis and current calibration method was performed for the initial-blankformed.

• Journal of Korean Society for Quality Management
• /
• v.30 no.3
• /
• pp.120-138
• /
• 2002

In order to carry out TQM and 6 Sigma management, it is the key elements to make clear a goal and a mission. Above all, we can succeed In achieving a dramatic change and a great outcome only when we make clear the method, incentives, organization and plans under the clear objective. In order to secure the competitiveness against the external challenge, it is essential to keep the several crucial factors such as CEO's will, the systematic process to measure and manage, monitoring to satisfy customer's needs and an aggressive development of TQM activity to encourage the endeavour of the relentless enhancement, and also a positive effort Is to be made for evaluating all quality culture like training experts internally by an outstanding training program under CEO's firm leadership. This study is carried out to understand that which features and factors of success can exist in a company culture if a company accepts a theoretical basis and concept, the general of TQM and 6 Sigma which are one of a management strategy, and carries out TQM and 6 Sigma for achieving improvement of quality and customer's satisfaction.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.5B
• /
• pp.824-832
• /
• 2010

In construction structure management, the effects of investigation by the professional manager is dependent on the cost, the inspection periods and methods. Therefore, effective and automated maintenance system for the target structure is required. Although some bridge monitoring systems are operating using wire based networks, the performance is not good enough to show sufficient ability as integrated bridge management system. In this paper, we design and implement an integrated bridge management system based on sensor networks. Two expert modules for bridge management and the integrated system management are provided. Moreover, web-based monitoring system is also designed for users at anywhere. The results show that the system is effective and readily available.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.3
• /
• pp.1-7
• /
• 2020

With the increase in cyber attacks, automated IDS using machine learning is being studied. According to recent research, the IDS using the recursive learning model shows high detection performance. However, the simple application of the recursive model may be difficult to reflect the associated session characteristics, as the overlapping session environment may degrade the performance. In this paper, we designed the session management module and applied it to LSTM (Long Short-Term Memory) recursive model. For the experiment, the CSE-CIC-IDS 2018 dataset is used and increased the normal session ratio to reduce the association of mal-session. The results show that the proposed model is able to maintain high detection performance even in the environment where session relevance is difficult to find.