• 한국건설관리학회논문집
• /
• 제24권5호
• /
• pp.35-43
• /
• 2023

공동주택 하자 분쟁의 증가와 함께, 하자관리의 중요성 또한 커지고 있다. 그러나 기존의 연구는 '공용 부분'에 초점을 맞추어 진행되었다. 또한 하자관리의 주체인 '관리사무소'를 위한 시스템 연구도 부족한 실정이다. 이는 관리사무소의 하자관리 능력의 부족과 관리 품질의 저하를 초래한다. 따라서, 본 논문에서는 관리사무소를 위한 기계학습 기반의 하자 정보 관리 시스템을 제안한다. OCR과 NLP 모듈을 사용하여 관리상의 불편한 점을 해소하는 것을 목표로 한다. OCR을 통해 수기로 작성된 하자 정보를 디지털 문서로 변환한다. 이후 언어모델을 이용하여 사용자가 지정한 양식과 함께 하자 정보를 재생성한다. 최종적으로 생성된 텍스트를 데이터베이스에 저장하고 이를 기반으로 통계적 분석을 실행한다. 이러한 일련의 과정을 통해, 관리사무소의 하자관리 역량을 향상할 수 있도록 돕고, 의사결정을 지원할 수 있을 것으로 기대한다.

• 한국컴퓨터정보학회논문지
• /
• 제25권3호
• /
• pp.19-26
• /
• 2020

일반적으로 품질 관리는 많은 제조 공정, 특히 주조 또는 용접과 관련된 공정의 기본 구성 요소가 된다. 그러나 사람이 일일이 수동으로 품질 관리 절차를 하는 것은 종종 시간이 걸리고 오류가 발생하기 쉽다. 최근 고품질 제품에 대한 요구를 만족시키기 위해 지능형 육안 검사 시스템의 사용이 생산 라인에서 필수적이 되고 있다. 본 논문에서는 이를 위해 딥 러닝 기반의 ShuffleDefectNet 결함 감지 시스템을 제안하고자 한다. 제안된 결함 검출 시스템은 NEU 데이터 세트의 결함 검출에 대한 여러 최신 성능들보다 높은 평균 정확도 99.75% 정도를 얻는다. 이 논문에서 여러 다른 트레이닝 데이터로부터 최상의 성능을 탐지하고 탐지 성능을 관찰하였다. 그 결과 ShuffleDefectNet의 전체 아키텍처를 사용할 때 정확성과 속도가 크게 향상됨을 알 수 있었다.

• 반도체디스플레이기술학회지
• /
• 제21권2호
• /
• pp.68-73
• /
• 2022

In this study, we produced the surface defect detection model to automatically detect defect bone plates using a deep learning algorithm. Bone plates with a width and a length of 50 mm are most used for fracture treatment. Normal bone plates and defective bone plates were printed on the 3d printer. Normal bone plates and defective bone plates were photographed with 1,080 pixels using the webcam. The total quantity of collected images was 500. 300 images were used to learn the defect detection model. 200 images were used to test the defect detection model. The mAP(Mean Average Precision) method was used to evaluate the performance of the surface defect detection model. As the result of confirming the performance of the surface defect detection model, the detection accuracy was 96.3 %.

• Journal of Biosystems Engineering
• /
• 제26권2호
• /
• pp.169-176
• /
• 2001

Defect of apple was depreciated the product value and causes storage disease seriously. To detect the defect of ‘Fuji’apple with machine vision system, the optical characteristics of defect should be investigated. In this research, absorbance spectra of defect were acquired by spectrophotometer in the range of visible and NIR region(400∼1,100nm) and L*a*b* color values were also acquired by colorimeter. NIR machine vision system was constructed with B&W camera, frame grabber, 16 tungsten-halogen lamps, variable focal length lens and NIR bandpass filter which was mounted to lens outward. Average gray values of defect at 15 NIR wavelength were acquired and the significant NIR wavelength was selected by comparing Mahalanobis distance between sound and defective apple. As the result of Mahalanobis distance analysis, the significant wavelength to discriminate the defectives in ‘Fuji’apple were found to be 720nm for scab and 970nm for bruise and cuts and 920nm was also effective regardless of defective types.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2004년도 추계학술대회 논문집
• /
• pp.585-589
• /
• 2004

This paper deals with the algorithm development that inspects defects such as Lens Focus, Black Defect, Dim Defect, Color Defect, White Balance, and Line Defect caused by the process of Compact Camera Module (CCM). These days the demand of CCM goes on increasing in various types like PDA, a cellular phone and PC camera every year. However, owing to the defect inspection of CCM by the semiskilled work the average inspection time of CCM takes about 40 to 50 seconds. As time goes by the efficiency takes a sudden turn for the worse because workers must inspect with seeing a monitor directly. In this paper, to solve these problems, we developed the imaging processing algorithm to inspect the defects in captured image of assembled CCM. The performances of the developed inspection system and its algorithm are tested on many samples. Experimental results reveal that the proposed system can focus the lens of CCM within 5s and we can recognize various types of defect of CCM modules with good accuracy and high speed.

• Carbon letters
• /
• 제16권3호
• /
• pp.198-202
• /
• 2015

Carbon-supported Pt catalyst systems containing defect adsorption sites on the anode of direct methanol fuel cells were investigated, to elucidate the mechanisms of H₂ dissociation and carbon monoxide (CO) poisoning. Density functional theory calculations were carried out to determine the effect of defect sites located neighboring to or distant from the Pt catalyst on H₂ and CO adsorption properties, based on electronic properties such as adsorption energy and electronic band gap. Interestingly, the presence of neighboring defect sites led to a reduction of H₂ dissociation and CO poisoning due to atomic Pt filling the defect sites. At distant sites, H₂ dissociation was active on Pt, but CO filled the defect sites to form carbon π-π bonds, thus enhancing the oxidation of the carbon surface. It should be noted that defect sites can cause CO poisoning, thereby deactivating the anode gradually.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.195-198
• /
• 2001

Backside defect of plate structure may grow due to fatigue or overload to cause critical failure during operation, so it is important to detect this kind of defect in line. For this purpose, nondestructive, non-contact and highly sensitive method is required. ESPI and Shearography are considered as useful method to satisfy these requirements. In this paper, the possibility of application of ESPI and Shearography to detect the backside defect of steel plate and to quantify the defect size was tested. For the experiment, some steel plates with defect on the backside were prepared. Experimental results for these plates showed that location and size of defect could be detected correctly by both of ESPI and Shearography.

• 산업경영시스템학회지
• /
• 제27권2호
• /
• pp.10-16
• /
• 2004

Data mining technique is the exploration and analysis, by automatic or semiautomatic means, of large quantities of data in order to discover meaningful patterns and rules. This paper uses a data mining technique for the prediction of defect types in manufacturing Process. The Purpose of this Paper is to model the recognition of defect type Patterns and Prediction of each defect type before it occurs in manufacturing process. The proposed model consists of data handling, defect type analysis, and defect type prediction stages. The performance measurement shows that it is higher in prediction accuracy than logistic regression model.

• 반도체디스플레이기술학회지
• /
• 제3권3호
• /
• pp.23-25
• /
• 2004

The defect formation energy of boron nitride (BN) nanotubes is investigated using molecular-dynamics simulation. Although the defect with tetragon-octagon pairs (4-88-4) is favored in the flat cap of BN nanotubes, BN clusters, and the growth of BN nanotubes, the formation energy of the 4-88-4 defect is significantly higher than that of the pentagon-heptagon pairs (5-77-5) defect in BN nanotubes. The 5-77-5 defect reduces the effect of the structural distortion caused by the 4-88-4 defect, in spite of homoelemental bonds.

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1288-1296
• /
• 2016

The reliable detection of the limited defect in TFT-LCD images is difficult due to the small intensity difference with the background. However, the proposed detection method reliably detects the limited defect by enhancing the TFT-LCD image based on the cumulative histogram and then detecting the defect through the mean and standard deviation of the enhanced image. Notably, an image enhancement using a cumulative histogram increases the intensity contrast between the background and the limited defect, which then allows defects to be detected by using the mean and standard deviation of the enhanced image. Furthermore, through the comparison with the histogram equalization, we confirm that the proposed algorithm suppresses the emphasis of the noise. Experimental comparative results using real TFT-LCD images and pseudo images show that the proposed method detects the limited defect more reliably than conventional methods.