• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.31-41
• /
• 2022

Steel is one of the most fundamental components to mechanical industry. However, the quality of products are greatly impacted by the surface defects in the steel. Thus, researchers pay attention to the need for surface defects detector and the deep learning methods are the current trend of object detector. There are still limitations and rooms for improvements, for example, related works focus on developing the models but don't take into account real-time application with practical implication on industrial settings. In this paper, a real-time application of steel surface defects detection based on YOLOv4 is proposed. Firstly, as the aim of this work to deploying model on real-time application, we studied related works on this field, particularly focusing on one-stage detector and YOLO algorithm, which is one of the most famous algorithm for real-time object detectors. Secondly, using pre-trained Yolov4-Darknet platform models and transfer learning, we trained and test on the hot rolled steel defects open-source dataset NEU-DET. In our study, we applied our application with 4 types of typical defects of a steel surface, namely patches, pitted surface, inclusion and scratches. Thirdly, we evaluated YOLOv4 trained model real-time performance to deploying our system with accuracy of 87.1 % mAP@0.5 and over 60 fps with GPU processing.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.5D
• /
• pp.821-829
• /
• 2006

Traffic load comprise primary input to pavement design causing pavement damage. therefore it should be proceeded suitable traffic load distribution modeling for pavement design and analysis. Traffic load have been represented by equivalent single axle loads (ESALs) which convert mixed traffic stream into one value for design purposes. But there are some limit to apply ESALs to other roads because it is empirical value developed as part of the original AASHO(American Association of State Highway Officials) road test. There have been many efforts to solve these problems. Several leading country have implemented M-E(Mechanistic-Empirical) design procedures based on mechanical concept. As a result, they established traffic load quantification method using load distribution model known as Axle Load Spectra. This paper details Axle Load Spectra and presents axle load distribution model based on normal mixture distribution function using truck load data collected by WIM system installed in national highway. Axle load spectra and axle load distribution model presented in this paper could be useful for basic data when making traffic load quantification plan for pavement design, overweight vehicle permit plan and pavement maintenance cost plan.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6C
• /
• pp.385-393
• /
• 2006

Soil nailing is a reinforcement method used for stabilizing excavated walls or slopes. Due to its much advantages such as ease of construction and economical efficiency, use of soil nailing is increased. However, the soil nail has much disadvantages for use in urban area. The soil nail needs to be installed inevitably beyond private land boundary, which causes rent for use. For this reason, removable soil nailing system was developed. However, the removal rate of this system is just about 50￠¦70%. To resolve this problem, the Fiber Reinforced Plastic (FRP) soil nailing system which does not need to be removed and allows for the installation beyond private land, is developed. In this paper, through theoretical and experimental studies in laboratory and field such as prototype tests, pullout tests, we evaluate the stability and behavior characteristics of the FRP soil nailing system. And, numerical analyses using FLAC2D were performed with respect to various soil conditions, where prototype test for excavation wall and pullout tests were carried out. As a result of this study, the FRP soil nailing systems show similar behavior characteristics with those of removable soil nailing system. Finally, considering the serviceability and mechanical stability of FRP soil nailing systems, it is enough to be used as a good alternative of general soil nailing system.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.6A
• /
• pp.625-639
• /
• 2009

Experimental and analytical studies are performed on the mechanical behavior of concrete-filled tubular(CFT) truss girders for different f/L ratios. Bending tests are conducted on two CFT truss girder specimens to determine fundamental structural characteristics such as the strength and deformation properties. Nonlinear material models for CFT members subjected to an axial compressive force are compared in this paper by using the nonlinear finite element program, ABAQUS. Previous researchers have proposed several nonlinear stress-strain models of confined concrete. In this study, the nonlinear analyses are performed applying several stress-strain models for confined concrete proposed by Mander, Sakino, Han, Susantha and Ellobody, and the results are compared with the experimental results in terms of load-deflection and load-strain relationships. Based on the comparisons of the load-deflection relationships, the models proposed by Mander and Susantha provide a maximum load about 12.0~13.8% higher and that by Sakino gives a maximum load about 7.6% higher than the experimental results. The models proposed by Han and Ellobody give a maximum load only about 0.2~1.2% higher than the test results, showing the best agreement among the proposed stress-strain models. However, the load-strain relations predicted by the existing models generally provide conservative results exhibiting larger strains than the experimental data.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.11 no.4
• /
• pp.433-439
• /
• 2023

In this study, the bonding performance of repair materials was evaluated on concrete repair surface to develop concrete repair materials based on UHPC (Ultra High Performance Concrete) which has high mechanical and durability performance. The ten test variables were applied considering the roughness and wet condition of the concrete surface subject to repair, the addition of polymer, and the use PP and PVA fibers in repair materials. The addition of the polymer caused a significant decrease in strength, which was thought to be due to the effect of the additional super plasticizer used to adjust workability. Also, flow was reduced by up to 13.8 % with the use of plastic-based fibers. As a result of evaluating the bond strength of the repair material considering the condition of the surface subject to repair, it was thought that in the case of using UHPC-based repair material, high bonding performance could be secured without any additional surface treatment as long as the surface of the base material was sound. In addition, UHPC-based repair materials showed high bonding performance even when the attachment surface was wet. In the future, research will be conducted on shot-crete application and gradient pouring for the development of UHPC-based repair materials, and continuous improvement in the repair material mixing property will be carried out to ensure economic efficiency and performance as a concrete structural repair material.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.5
• /
• pp.323-330
• /
• 2023

The development of an analysis model that reflects the microstructure characteristics of polyvinyl alcohol (PVA) fiber-reinforced cementitious composites, which have a highly complex microstructure, enables synergy between efficient material design and real experiments. PVA fiber orientations are an important factor that influences the mechanical behavior of PVA fiber-reinforced cementitious composites. Owing to the difficulty in distinguishing the gray level value obtained from micro-CT images of PVA fibers from adjacent phases, fiber segmentation is time-consuming work. In this study, a micro-CT test with a voxel size of 0.65 ㎛³ was performed to investigate the three-dimensional distribution of fibers. To segment the fibers and generate training data, histogram, morphology, and gradient-based phase-segmentation methods were used. A U-net model was proposed to segment fibers from micro-CT images of PVA fiber-reinforced cementitious composites. Data augmentation was applied to increase the accuracy of the training, using a total of 1024 images as training data. The performance of the model was evaluated using accuracy, precision, recall, and F1 score. The trained model achieved a high fiber segmentation performance and efficiency, and the approach can be applied to other specimens as well.

• Journal of Biomedical Engineering Research
• /
• v.44 no.6
• /
• pp.404-413
• /
• 2023

In this research, we studied the development of a SUS304 microneedle array based on microfabrication technology and the applicability of Parylene-C thin film, a medical polymer material. First of all, four materials commonly used in the field of medical engineering (SUS304, Ti, PMMA, and PEEK) were selected and a 5 ㎛ Parylene-C thin film was deposited. The applicability of Parylene-C coating to each material was confirmed through SEM analysis, contact angle measurement, surface roughness(Ra) measurement, and adhesion test according to ASTM standards for each specimen. Parylene-C thin film was deposited based on chemical vapor deposition (CVD), and a 5 ㎛ Parylene-C deposition process was established through trial and error. Through characteristic experiments to confirm the applicability of Parylene-C, SUS304 material, which is the easiest to apply Parylene-C coating without pretreatment was selected to develop a microneedle array based on CNC micromachining technology. The CNC micromachining process was divided into a total of 5 steps, and a microneedle array consisting of 19 needles with an inner diameter of 200 ㎛, an outer diameter of 400 ㎛, and a height of 1.4 mm was designed and manufactured. Finally, a 5 ㎛ Parylene-C coated microneedle array was developed, which presented future research directions in the field of microneedle-based drug delivery systems.

• Journal of Korea Foundry Society
• /
• v.43 no.6
• /
• pp.279-285
• /
• 2023

In order to predict the mechanical properties of ductile cast iron heat treated in an intercritical temperature range, samples machined from cast iron with a tensile strength of 450 MPa were heat-treated at various intercritical temperatures and air-cooled, after which a microstructural analysis and Brinell hardness test were conducted. As the heat treatment temperature was increased in the intercritical temperature range, the ferrite fraction in the ductile cast iron decreased and the pearlite fraction increased, whereas the nodularity and nodule count did not change considerably from the corresponding values in the as-cast condition. The Brinell hardness values of the heat-treated ductile cast iron increased gradually as the heat treatment temperature was increased. Based on the measured alloy composition, the fraction of each stable phase and the hardness model from the literature, the hardness of the ductile cast iron heat treated in the intercritical temperature range was calculated, showing values very similar to the measured hardness data. In order to check whether it is possible to predict the hardness of heat-treated ductile cast iron by using the phase fraction obtained from thermodynamic calculations, the volumes of graphite, ferrite, and austenite in the alloy were calculated for each temperature condition. Those volume fractions were then converted into areas of each phase for hardness prediction of the heat-treated ductile cast iron. The hardness values of the cast iron samples based on thermodynamic calculations and on the hardness prediction model were similar within an error range up to 27 compared to the measured hardness data.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.6
• /
• pp.250-258
• /
• 2007

Current speech recognition technology s achieved high performance with the development of hardware devices, however it is insufficient for some applications where high reliability is required, such as voice control of powered wheelchairs for disabled persons. For the system which aims to operate powered wheelchairs safely by voice in real environment, we need to consider that non-voice commands such as user s coughing, breathing, and spark-like mechanical noise should be rejected and the wheelchair system need to recognize the speech commands affected by disability, which contains specific pronunciation speed and frequency. In this paper, we propose non-voice rejection method to perform voice/non-voice classification using both YIN based fundamental frequency(F0) extraction and reliability in preprocessing. We adopted a multi-template dictionary and acoustic modeling based speaker adaptation to cope with the pronunciation variation of inarticulately uttered speech. From the recognition tests conducted with the data collected in real environment, proposed YIN based fundamental extraction showed recall-precision rate of 95.1% better than that of 62% by cepstrum based method. Recognition test by a new system applied with multi-template dictionary and MAP adaptation also showed much higher accuracy of 99.5% than that of 78.6% by baseline system.

• Journal of Korean Society of Water and Wastewater
• /
• v.37 no.4
• /
• pp.215-229
• /
• 2023

In this study, non-destructive technologies that can be applied to evaluate the integrity of valve materials, safety against internal pressure caused by corrosion, and the blocking function of large-diameter water valves during operation without requiring specimen collection or manpower entering the inside of the valve were tested to assess the reliability of the technologies and their suitability for field application. The results showed that the condition of the graphite structure inside the valve body can be evaluated directly through the optical microscope in the field without specimen collection for large-diameter water butterfly valves, and the depth of corrosion inside the valve body can be determined by array ultrasound and the tensile strength can be measured by instrumented indentation test. The reliability of each of these non-destructive techniques is high, and they can be widely used to evaluate the condition of steel or cast iron pipes that are significantly smaller in thickness than valves. Evaluation of blocking function of the valves with mixed gas showed that it can be detected even when a very low flow rate of mixed gas passes through the disk along with the water flow. Finally, as a result of evaluating the field applicability of non-destructive technologies for three old butterfly valves installed in the US industrial water pipeline, it was found that it is possible to check the material and determine the suitability of large-diameter water valves without taking samples, and to determine the corrosion state and mechanical strength. In addition, it was possible to evaluate safety through the measurement results, and it is judged that the evaluation of the blocking function using mixed gas will help strengthen preventive response in the event of an accident.