Using active thermography to inspect pin-hole defects in anti-reflective coating with k-mean clustering

Wang, Hongjin; Hsieh, Sheng‐Jen; Zhou, Xunfei; Peng, Bo; Singh, Bhavana

doi:10.1016/j.ndteint.2015.08.006

Cited by 19 publications

(5 citation statements)

References 28 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SPSS 26.0 was used to analyse the statistical data. K-means clustering analysis, a widely used centroid-based clustering algorithm (Wang et al, 2015), was applied to divide stroke families into different clusters based on the FRAS-C scores of patients and caregivers at T0. This was done by ensuring that each FRAS-C score belonged to a cluster with the nearest mean.…”

Section: Discussionmentioning

confidence: 99%

Dyadic profiles of family resilience among patients with first‐episode stroke: A longitudinal study of the first 6 months after stroke

Zhang

Gao

et al. 2022

Journal of Clinical Nursing

View full text Add to dashboard Cite

Background The importance of family resilience in the recovery of stroke patients has been demonstrated in numerous studies. However, little is known about post‐stroke family resilience. Aims To investigate the family resilience of stroke patients from a patient–caregiver dyadic perspective during the first 6 months after stroke. Methods A total of 288 dyads of patients diagnosed with a first‐episode stroke and their principal caregivers were recruited from neurology departments of 7 tertiary hospitals in Shanghai and Shangqiu, China. Family resilience and family function were assessed during hospitalisation and at 1, 3 and 6 months after stroke. K‐means cluster analysis was used to identify different clusters of family resilience based on family resilience of patients and caregivers during hospitalisation. The STROBE guidelines for observational studies were followed. Results Three clusters of family resilience were identified with distinct trajectories: cluster of high resilience (HR), cluster of low resilience (LR) and cluster of discrepant resilience (DR). The level of family function was consistently highest in cluster HR and lowest in cluster with LR at four time points. Most (69.8%) families fell into the cluster with low resilience and low family function. Characteristics such as the Rankin scores and education level of patients, education level of caregivers, family monthly income and living district were different among the three clusters. Conclusions We concluded that family resilience was linked to the family functioning of patients with a first‐episode stroke, however, the levels of resilience in most families were low. Factors, including the education level, family income and stroke severity of patients were revealed to influence the family resilience and its development. Relevance to clinical practice A resilience‐focused approach to family‐related treatment is beneficial for families. Therefore, understanding family resilience among stroke survivors is needed to inform the development of interventions for enhancing the recovery of stroke families.

show abstract

Section: Discussionmentioning

confidence: 99%

Dyadic profiles of family resilience among patients with first‐episode stroke: A longitudinal study of the first 6 months after stroke

Zhang

Gao

et al. 2022

Journal of Clinical Nursing

View full text Add to dashboard Cite

show abstract

“…Infrared cameras are employed in various disciplines for non-destructive evaluation through thermographic techniques. Additionally, several methods 15,16 have been developed to facilitate the automatic identification of specific anomalies. In contrast to many applications where the nature of the defect is predetermined and the input data is relatively unambiguous, monitoring the heat shield of tokamak walls presents a unique challenge.…”

Section: A Ir Diagnostics Installed On West Tokamakmentioning

confidence: 99%

Anomaly classification by inserting prior knowledge into a max-tree based method for divertor hot spot characterization on WEST tokamak

Gorse,

Mitteau,

Marot

2023

Review of Scientific Instruments

View full text Add to dashboard Cite

The divertor of WEST (W Environment in Steady-state Tokamak) is the main component for plasma control and exhaust. It receives high heat fluxes, which can cause damage to plasma facing units above the allowable heat flux. Improving the operation safety on the actively cooled tungsten divertor is being researched in place at WEST, toward providing divertor monitoring solution for ITER. Divertor operation safety relies on detecting, monitoring, and classifying all hot spots on the divertor surface using infrared (IR) cameras. In this paper, a method based on max-tree representation and attributes of IR images is used to classify normal from abnormal strikelines on the divertor. The proposed method requires only high-level prior knowledge of abnormal temperatures and divertor structure but does not require any labeled data, unlike existing methods, such as support vector machines (SVMs) or convolutional neural networks (CNNs). The max-tree classifier method is tested on real IR images from the WEST tokamak and shows that 88% of hot spots are accurately classified with a small enough calculation duration that can be performed between two pulses.

show abstract

“…He et al 2019), independent principal component analysis (Ahmad et al 2019), and sequential forward floating selection (Wang et al 2017). The main methods involved in the feature recognition stage are perceptron (Zeng, Dai, and Mu 2007), support vector machine (Xiao et al 2020), decision tree (Z. F. , and clustering (Wang et al 2015).…”

Section: Current Development Of Iiprmentioning

confidence: 99%

Deep learning for industrial image: challenges, methods for enriching the sample space and restricting the hypothesis space, and possible issue

Liu

Bao

Wang

et al. 2021

International Journal of Computer Integrated Manufacturing

View full text Add to dashboard Cite

Deep learning (DL) is an important enabling technology for intelligent manufacturing. The DLbased industrial image pattern recognition (DLBIIPR) plays a vital role in the improvement of product quality and production efficiency. Although DL technology has been widely used in the field of natural image, industrial image often has some mixed characteristics, such as small sample, imbalance, small target, strong interference, fine-grained, temporality and semantical, which reduce the feasibility and generalization of DLBIIPR. To solve this problem, this paper provides an overview of approaches commonly used in industry by enriching the sample space and limiting the hypothesis space. In order to improve the confidence of front-line workers in using DL models, the explainable deep learning (XDL) methods are reviewed, and a case study is used to verify the effectiveness of XDL.

show abstract

Using active thermography to inspect pin-hole defects in anti-reflective coating with k-mean clustering

Cited by 19 publications

References 28 publications

Dyadic profiles of family resilience among patients with first‐episode stroke: A longitudinal study of the first 6 months after stroke

Dyadic profiles of family resilience among patients with first‐episode stroke: A longitudinal study of the first 6 months after stroke

Anomaly classification by inserting prior knowledge into a max-tree based method for divertor hot spot characterization on WEST tokamak

Deep learning for industrial image: challenges, methods for enriching the sample space and restricting the hypothesis space, and possible issue

Contact Info

Product

Resources

About