Deep learning-based computer vision anomaly detection algorithms are widely utilized in various fields. Especially in the manufacturing industry, the difficulty in collecting abnormal data compared to normal data, and the challenge of defining all potential abnormalities in advance, have led to an increasing demand for unsupervised learning methods that rely on normal data. In this study, we conducted a comparative analysis of deep learning-based unsupervised learning algorithms that define and detect abnormalities that can occur when transparent contact lenses are immersed in liquid solution. We validated and applied the unsupervised learning algorithms used in this study to the existing anomaly detection benchmark dataset, MvTecAD. The existing anomaly detection benchmark dataset primarily consists of solid objects, whereas in our study, we compared unsupervised learning-based algorithms in experiments judging the shape and presence of lenses submerged in liquid. Among the algorithms analyzed, EfficientAD showed an AUROC and F1-score of 0.97 in image-level tests. However, the F1-score decreased to 0.18 in pixel-level tests, making it challenging to determine the locations where abnormalities occurred. Despite this, EfficientAD demonstrated excellent performance in image-level tests classifying normal and abnormal instances, suggesting that with the collection and training of large-scale data in real industrial settings, it is expected to exhibit even better performance.

강풍, 폭우 등 이상기후의 대형화와 빈도 증가로 인해 나무가 부러지거나 쓰러지는 훼손이 증가하고 있으나 나무 내부의 공동, 부후 등 구조적 결함은 육안조사로 판별이 어렵기 때문에 예측을 통한 사전대응에 한계가 있다. 비파괴 음파단층촬영은 나무에 미치는 물리적 훼손을 최소화하면서 내부결함을 추정하는 방법으로 내부결함 진단에 효율적이 나 수종별 정확도에 차이가 발생하기 때문에 현장적용 전 측정결과의 신뢰성 분석이 선행되어야 한다. 이번 연구는 우리나라 대표 수종인 소나무와 은행나무 노거수를 대상으로 음파단층촬영의 신뢰성 검증을 위해 침입성 드릴저항 측정을 교차 적용하여 목재 내부결함을 측정하고 평가결과를 비교하였다. 두 집단 간 결함부 측정 평균값에 대한 t검정 결과 소나무는 통계적으로 유의한 차이가 없는 반면, 은행나무는 유의성에 차이가 있었다. 선형회귀분석 결과 두 수종 모두 드릴저항그래프의 결함이 증가할 때 음파단층영상 결함이 증가하는 양의 상관관계를 보였다.

Visual inspection is regarded as one of the critical activities for quality control in a manufacturing company. it is thus important to improve the performance of detecting a defective part or product. There are three probable working modes for visual inspection: fully automatic (by automatic machines), fully manual (by human operators), and semi-automatic (by collaboration between human operators and automatic machines). Most of the current studies on visual inspection have been focused on the improvement of automatic detection performance by developing a better automatic machine using computer vision technologies. However, there are still a range of situations where human operators should conduct visual inspection with/without automatic machines. In this situation, human operators’performance of visual inspection is significant to the successful quality control. However, visual inspection of components assembled into a mobile camera module belongs to those situations. This study aims to investigate human performance issues in visual inspection of the components, paying more attention to human errors. For this, Abstraction Hierarchy-based work domain modeling method was applied to examine a range of direct or indirect factors related to human errors and their relationships in the visual inspection of the components. Although this study was conducted in the context of manufacturing mobile camera modules, the proposed method would be easily generalized into other industries.

The number of pixel defects including bright and black dots on a panel is one of the critical factors determining the quality of TFT-LCD. Since pixel defects on the TFT-LCD panels are sometimes unavoidable, manufacturers have to inspect the panels so that any panel with an unacceptable number of defects will not be delivered to the buyers. However, the buyers demand for the manufacturers to meet different pixel defects tolerances (acceptable number of pixel defects on a TFT-LCD panel) around central(tight) and peripheral(loose) inspection zones. The disagreement in quality standard among different buyers also cause confusions in screening non-confirmative products and unstable yield of production. Few research has focused on the effects of defect locations on a TFT-LCD panel on their detection probabilities and the rational division of defect inspection zones. In this research, experiments were conducted to find the detection probabilities of black dot defects with respect to their varying locations on a TFT-LCD. It is proposed a rational division of inspection zone on a TFT-LCD panel on the basis of detection probabilities of the defects. With these division of inspection zones and the mean defect detection probability within each zone, it is expected to establish a more reasonable pixel defects tolerances.

This research aims to detect delamination-like damage from asphalt-concrete interface using contactless ultrasonic technique. FEM simulation and experiments demonstrate that surface waves and S mode are strongly present when there is delamination-like damage. It is believed that the measurement of S mode by non-contact ultrasonic system will allow one to determine the existence of delamination-like damage from bridge decks.

The Infrared-Thermography Technique is widely used for nondestructive evaluation (NDE) of structures due to its convenience. Particularly, various techniques have been tried to detect imperfections such as surface defects and welding defects of steel structures. This paper is a part of a study to detect defects in weld zone in steel structures using Infrared Thermography techniques. Finally, defects were detected by comparing infrared thermal image and surface temperature data of defected area with healthy area.