• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.136-146
• /
• 2022

Some surface defects make the weak quality of steel materials. To limit these defects, we advocate a one-stage detector model RetinaNet among diverse detection algorithms in deep learning. There are several backbones in the RetinaNet model. We acknowledged two backbones, which are ResNet50 and VGG19. To validate our model, we compared and analyzed several traditional models, one-stage models like YOLO and SSD models and two-stage models like Faster-RCNN, EDDN, and Xception models, with simulations based on steel individual classes. We also performed the correlation of the time factor between one-stage and two-stage models. Comparative analysis shows that the proposed model achieves excellent results on the dataset of the Northeastern University surface defect detection dataset. We would like to work on different backbones to check the efficiency of the model for real world, increasing the datasets through augmentation and focus on improving our limitation.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.2
• /
• pp.356-362
• /
• 2009

The ELectromagnetic Core Imperfection Detector (EL-CID) test was performed on a small generator in the laboratory and a gas turbine generator in the field to assess the fault condition of generator stator core. Artificial defects with six different sizes were introduced in the small generator. The scan results on six defects show a very large increase in the magnitude of fault current compared to that obtained with a healthy core. After the stator core heats up, a thermal imaging camera was used to detect hot spot on the inner surface of the core for comparison. Several faults were found during inspection of the gas turbine generator with the EL-CID. It has been shown that the existence of a fault can be determined by monitoring the magnitude of fault current.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.548-553
• /
• 2011

The lock-in thermography was employed to evaluate the defects in railway bogies. Prior to the actual application on railway bogies, in order to assess the detectability of known flaws, the calibration reference panel was prepared with various dimensions of artificial flaws. The panel was composed of polymer matrix composites, which were the same material with actual bogies. Through lock-in thermography evaluation, the optimal frequency of heat source was determined for the best flaw detection. Based on the defects information, the actual defect assessments on railway bogie were conducted with different types of railway bogies, which were used for the current operation. In summary, it was found that the novel infrared thermography technique could be an effective way for the inspection and the detection of surface defects on bogies since the infrared thermography method provided rapid and non-contact investigation of railway bogies.

• Journal of Welding and Joining
• /
• v.33 no.4
• /
• pp.63-67
• /
• 2015

Non-destructive techniques are used widely in the metal industry in order to control the quality of materials. Eddy current testing(ECT) is one of the most extensively used non-destructive techniques for inspecting electrically conductive materials at very high speeds that does not require any contact between the test piece and the sensor. The New ECT sensor which can detect inner defects was developed regardless the condition of surface. This sensor is verified to do experiment which measure the loss of induced electromotive force. The loss of induced electromotive force was measured in 5.4% and this low frequency ECT device can detect internal defects at depth 20 mm.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.6
• /
• pp.398-406
• /
• 2015

This paper explored the results of experimental investigation on carbon fiber reinforced polymer (CFRP) composite sample with thermal wave technique. The thermal wave technique combines the advantages of both conventional thermal wave measurement and thermography using a commercial Infrared camera. The sample comprises the artificial inclusions of foreign material to simulate defects of different shape and size at different depths. Lock-in thermography is employed for the detection of defects. The temperature field of the front surface of sample was observed and analysed at several excitation frequencies ranging from 0.562 Hz down to 0.032 Hz. Four-point methodology was applied to extract the amplitude and phase of thermal wave's harmonic component. The phase images are analyzed to find qualitative and quantitative information about the defects.

• Smart Structures and Systems
• /
• v.33 no.1
• /
• pp.55-64
• /
• 2024

This paper proposes a fusion imaging-based coating-defect classification method for steel structures that uses zero-shot learning. In the proposed method, a halogen lamp generates heat energy on the coating surface of a steel structure, and the resulting heat responses are measured by an infrared (IR) camera, while photos of the coating surface are captured by a charge-coupled device (CCD) camera. The measured heat responses and visual images are then analyzed using zero-shot learning to classify the coating defects, and the estimated coating defects are visualized throughout the inspection surface of the steel structure. In contrast to older approaches to coating-defect classification that relied on visual inspection and were limited to surface defects, and older artificial neural network (ANN)-based methods that required large amounts of data for training and validation, the proposed method accurately classifies both internal and external defects and can classify coating defects for unobserved classes that are not included in the training. Additionally, the proposed model easily learns about additional classifying conditions, making it simple to add classes for problems of interest and field application. Based on the results of validation via field testing, the defect-type classification performance is improved 22.7% of accuracy by fusing visual and thermal imaging compared to using only a visual dataset. Furthermore, the classification accuracy of the proposed method on a test dataset with only trained classes is validated to be 100%. With word-embedding vectors for the labels of untrained classes, the classification accuracy of the proposed method is 86.4%.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.6 no.1
• /
• pp.45-50
• /
• 1997

It is not clearly known how surface defects or inclusions of a medium carbon steel affect a fatigue strength. In this study, we used SM35C specimens with spheroidized cementite structure to eliminate dependence of micro structure of fatigue crack. The investigation was carried out by behavior of crack closure at non-propagation crack and effect of the fatigue limit according to the artificial defects size. Experimental findings are obtained as follows : (1) Fatigue crack initiation point of medium carbon steel with spheroidized cementite structure is at the surface defects. (2) Non-propagating crack length of smooth specimen is equal to the critical size of defect. (3) Considering the opening and closure behavior of fatigue crack, the defect shape results in various crack opening displacement, while it does not affects the fatigue limit level of medium carbon steel with spheroidized cementite structure. (4) The critical length of the non-propagation crack of smooth specimen is the same as critical size of defect in transient area which determines threshold condition in steel with spheroidized cementite structure.

• Restorative Dentistry and Endodontics
• /
• v.27 no.2
• /
• pp.113-121
• /
• 2002

Conventional intraoral radiography continues to be the most widely used image modality for the diagnosis of dental caries. But, conventional intraoral radiography has several shortcomings, including the difficulty of exposing and processing intraoral film of consistently acceptable quality. In addition, radiographic retaking that was the result of processing errors, may result in increased discomfort and radiation dose to the patient. Recently, various digital radiographies substitute for conventional intraoral radiography to overcome these disadvantages. The advantages of digital radiography are numerous. One of advantages Is the elimination of processing errors. In addition, the radiation dose for digital system is approximately 20% to 25% of that required for conventional intraoral radiography Another potential advantage of digital imaging is the ability to perform image quality enhancements such as contrast and density modulation, which may increase diagnostic accuracy. The purpose of this study was to compare the diagnostic ability of artificial proximal defects to conventional intraoral radiography, direct digital image(CDX2000HQ$^{\circledR}$) and indirect digital image(Digora$^{\circledR}$). Artificial defects were made in proximal surfaces of 60 extracted human molars using #1/2, #1, #2 round bur. Five dentists assessed proximal defects on conventional intraoral radiography, direct digital image(CDX2000HQ$^{\circledR}$) and indirect digital image(Digora$^{\circledR}$). ROC(Receiver Operating Characteristic) analysis and Two-way ANOVA test were used for the evaluation of detectability, and following results were acquired. 1. The mean ROC area of conventional intraoral radiography, direct digital image(CDX2000HQ$^{\circledR}$) and indirect digital Image(Digora$^{\circledR}$) were 0.6766, 0.7538, 0.6791(Grade I), 0.7176, 0.7594, 0.7361(Grade II), and 0.7449, 0.7608, 0.7414(Grade III), respectively. 2. Diagnostic ability of direct digital image was higher than other image modalities. But, there was no statistically significant difference among other imaging modalities for Grade I, II, III lesion(p>0.05). In conclusion, when direct and indirect digital system are comparable with conventional intraoral radiography. these systems may be considered an alternative of conventional intraoral radiography for the diagnosis of proximal surface caries.

• Advances in materials Research
• /
• v.5 no.3
• /
• pp.193-203
• /
• 2016

Compared to metal-to-metal tribology, polymer tribology presents further complexity as it is more prone to be influenced by operating conditions. Over the past two decades, progress in the field of wear of polymers has led to the establishment of more refined wear mechanisms. The current paper establishes the link between different load parameters and the wear rate of polymers, based on experimental investigations. A pin-on-plate reciprocating tribometer was used to examine the wear behaviour of polyamide sliding against a steel counterface, under constant and fluctuating loads, in dry conditions. In addition, the influence of controlled imperfections in the polymer surface upon its wear rate were examined, under cyclic and steady loading, in order to better understand surface fatigue wear of polymers. The imposed imperfections consisted of vertical artificial deep crack (slit) perpendicular or parallel to the direction of sliding. The study concludes with the followings findings; in general, wear of polymers shows a significant tendency to the type of applied load. Under cyclic loads, polymers show an increase in wear rate compared to those tested under static loads. Such increase was found to increase with the increase in cyclic load frequency. It is also demonstrated that surface cracks results in higher wear rates, particularly under cyclic loads.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.23 no.11 s.170
• /
• pp.1922-1928
• /
• 1999

Rotary bending fatigue tests were carried out to investigate the influence of artificial defects on fatigue limit in annealed and austempered ductile iron. Obtained main results are as follows : (1) Artificial defect(micro hole type, dia.<0.4 mm) on specimen surface did not bring about a obvious reduction of fatigue limit in austempered ductile iron(ADI) as compared with annealed ductile iron. (2) According to the investigation of $\sqrt{area}_c$ which is the critical defect size to crack initiation at artificial defect, $\sqrt{area}_c$ of ADI is larger than that of annealed ductile iron. This shows that the situation of crack initiation at artificial defect in ADI is more difficult in comparison with annealed ductile iron. (3) One of the reasons for the low rate of crack initiation from artificial defect in ADI is that the resistance of matrix to crack initiation is higher than that of annealed ductile iron. (4) In case that the $\sqrt{area}$ of artificial defect and graphite nodule is the same, the rate of crack initiation from graphite nodule is higher than that from artificial defect. This reason is that the serious ruggedness around graphite nodule is formed by austempering treatment.