• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.2059-2064
• /
• 2004

Texture analysis is an important role for automatic visual insfection. This paper presents an application of wavelet extension and Gray level co-occurrence matrix (GLCM) for detection of defect encountered in textured images. Texture characteristic in low quality images is not to easy task to perform caused by noise, low frequency and small dimension. In order to solve this problem, we have developed a procedure called wavelet image extension. Wavelet extension procedure is used to determine the frequency bands carrying the most information about the texture by decomposing images into multiple frequency bands and to form an image approximation with higher resolution. Thus, wavelet extension procedure offers the ability to robust feature extraction in images. Then the features are extracted from the co-occurrence matrices computed from the sub-bands which performed by partitioning the texture image into sub-window. In the detection part, Mahalanobis distance classifier is used to decide whether the test image is defective or non defective.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.463-464
• /
• 2011

Currently, different kinds of diagnosis and inspection technologies are applied to prevent the internal mechanical transformation of transformers. For example, examination of internal Partial Discharge of transformer, analysis of transformer oil gas, and measurement Frequency Response Analyzer(FRA) are used to diagnose defect. Especially, diagnosis technique through Frequency Response Analyzer(FRA) has been used and developed from 1960, when it was first introduced, till now to become an important tool to examine presence of defect and to prove quality of machines for the most electric machine producers electric power company in the world. However, diagnosis through FRA is still in introduction level in Korea and the application method for FRA is not established yet. For that reason, study about the application of domestic electric installation according to the FRA is needed. It is expected that the study play an important part in the prevention of defect due to the internal transformation of transformer by introducing measurement theory, providing measurement method, and analyzing application cases.

• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.176-180
• /
• 2001

In this paper, we have studies the images have been reconstructed by using combination of images which has been acquired by the variation of operating frequency. When inner images has been reconstructed, inner image has been superposition by surface state effect. In this case, image enhancement the phase object and enhance the contrast of inner image. In the result of the specimen for the round defect with 2mm diameter, for the types of the depth are 1.5mm, 2mm, 2.5mm, and 3mm, it has been show that the shape of defect has better than before this processing and phase contrast grow large twice. And we have constructed the acoustic microscope by using quadrature detector that is able simultaneously to acquired the amplitude and phase of the reflected signal. Father more we have studied the reconstruction method of the amplitude and phase images and the enhancement method of the defect images' contrast.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.61-68
• /
• 2018

In general, defects cause a lot of maintenance costs and serious damage to various stakeholders, such as the owners, contractors or occupants of apartments. For this reason, a systematic and efficient defect management method is needed to minimize defect disputes. This paper derives a defect classification framework and proposes a defect risk assessment model for different types of defects. For this purpose, 6,000 defect items are allocated to the defect classification framework; these items are associated with 34 apartment projects over ten years old. As a result of this analysis, it was confirmed that the defect risks are concentrated in the areas of RC and finishing work. Based on these results, it is necessary to prevent the major risks of defects according to their priority. Based on this research, it is judged that further research to develop a method of managing the risks of defects may be necessary.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.56-64
• /
• 2019

Various defects that occur in the maintenance stage are connected to all kinds of wasted resources and economic losses as additional investments are made. Residents are harmed temporally, materially, and psychologically, and businesses suffer not only monetary losses but also reduced credit ratings. The aim of this study was to increase the efficiency of quality management and minimize defect disputes by estimating the importance of the defect type considering the defect frequency and severity in apartment buildings. For this, 7,548 defect items for 48 apartment buildings were examined. The analysis confirmed that defects are concentrated on RC, finishing and MEP work. In addition, defects with high importance are identified as broken, incorrect installation, missing tasks, and water problems. In addition, the exterior wall/roof, the Internal wall, ceiling, and floor, which are constructed in the field, are more important than the furniture and MEP equipment installed in the field.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.359-366
• /
• 2021

A number of construction companies in Korea invest considerable human and financial resources to construct a system for managing apartment defect data and for categorizing repair tasks. Thus, this study proposes machine learning models to automatically classify defect complaint text-data into one of the sub categories of 'finishing work' (i.e., one of the defect repair tasks). In the proposed models, we employed two word representation methods (Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF)) and two machine learning classifiers (Support Vector Machine, Random Forest). In particular, we conducted both binary- and multi- classification tasks to classify 9 sub categories of finishing work: home appliance installation work, paperwork, painting work, plastering work, interior masonry work, plaster finishing work, indoor furniture installation work, kitchen facility installation work, and tiling work. The machine learning classifiers using the TF-IDF representation method and Random Forest classification achieved more than 90% accuracy, precision, recall, and F1 score. We shed light on the possibility of constructing automated defect classification systems based on the proposed machine learning models.

• Smart Structures and Systems
• /
• v.30 no.3
• /
• pp.303-315
• /
• 2022

The degradation of wheel tread may result in serious hazards in the railway operation system. Therefore, timely wheel defect diagnosis of in-service trains to avoid tragic events is of particular importance. The focus of this study is to develop a novel wheel defect detection approach based on the relevance vector machine (RVM) which enables online detection of potentially defective wheels with trackside monitoring data acquired under different running-speed conditions. With the dynamic strain responses collected by a trackside monitoring system, the cumulative Fourier amplitudes (CFA) characterizing the effect of individual wheels are extracted to formulate multiple probabilistic regression models (MPRMs) in terms of multi-kernel RVM, which accommodate both variables of vibration frequency and running speed. Compared with the general single-kernel RVM-based model, the proposed multi-kernel MPRM approach bears better local and global representation ability and generalization performance, which are prerequisite for reliable wheel defect detection by means of data acquired under different running-speed conditions. After formulating the MPRMs, we adopt a Bayesian null hypothesis indicator for wheel defect identification and quantification, and the proposed method is demonstrated by utilizing real-world monitoring data acquired by an FBG-based trackside monitoring system deployed on a high-speed trial railway. The results testify the validity of the proposed method for wheel defect detection under different running-speed conditions.

• Journal of Welding and Joining
• /
• v.19 no.3
• /
• pp.267-273
• /
• 2001

In this study, welding phenomena involved in formation of penetrators during high frequency electric resistance welding were investigated. High speed cinematography of the process revealer that a molten bridge between neighboring skelp edges forms at apex point and travels along narrow gap toward to welding point at a speed ranging from 100 to 400 m/min. The bridge while moving along the narrow gap swept away oxide containing molten metal from the gap, providing oxide-free surface for a forge-welding at upsetting stand frequency of the budge formation, travel distance and speed of the bridge were affected by the heat input rate into strip. The travel distance and its standard deviation were found to have a strong relationship with the weld defect density. Based on the observation, a new mechanism of the penetrator formation during HF ERW process is proposed.

• Journal of Magnetics
• /
• v.18 no.3
• /
• pp.357-360
• /
• 2013

A modified SQI method using magnetic leakage flux (MFL) signal for underground gas pipelines' defect detection and characterization is presented in this paper. Raw signals gathered using MFL signals include many unexpected noises and high frequency signals, uneven background signals, signals caused by real defects, etc. The MFL signals of defect free pipelines primarily consist of two kinds of signals, uneven low frequency signals and uncertain high frequency noises. Leakage flux signals caused by defects are added to the case of pipelines having defects. Even though the SQI (Self Quotient Image) is a useful tool to gradually remove the varying backgrounds as well as to characterize the defects, it uses the division and floating point operations. A modified SQI having low computational complexity without time-consuming division operations is presented in this paper. By using defects carved in real pipelines in the pipeline simulation facility (PSF) and real MFL data, the performance of the proposed method is compared with that of the original SQI.

• Journal of Power System Engineering
• /
• v.14 no.4
• /
• pp.63-67
• /
• 2010

This paper introduces the eddy current signal characteristic of magnetic and non-magnetic gas turbine rotor. In the past, Magnetic particle inspection method was used in magnetic material for qualitative defect evaluation and the ultrasonic test method was used for quantitative evaluation. Nowadays, eddy current method is used in magnetic gas turbine rotor inspection due to advanced sensor design technology. We are studying on the magnetic gas turbine rotor by using eddy current method. We prepared diverse depth specimens made by magnetic and non-magnetic materials. We select optimum frequency according to material standard penetration data and experiment results. We got the signal on magnetic and non-magnetic material about 0.2 mm, 05 mm, 1.0 mm, 1.5 mm 2.0 mm and 2.5 mm depth defects and compare the signal amplitude and signal trend according to defect depth and frequency. The results show that signal amplitudes of magnetic are bigger than non-magnetic material and the trends are similar on every defect depth and frequency. The detection and resolution capabilities of eddy current are more effective in magnetic material than in non-magnetic materials. So, the eddy current method is effective inspection method on magnetic gas turbine rotor. And it has the merits of time saving and simple procedure by elimination of the ultrasonic inspection in traditional inspection method.