Ensemble variational Bayes tensor factorization for super resolution of CFRP debond detection

Lü, Peng; Gao, Bin; Feng, Qizhi; Yang, Yang; Woo, Wai Lok; Tian, Gui Yun

doi:10.1016/j.infrared.2017.07.012

Cited by 21 publications

(20 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The PPT algorithm, based on the Fourier Transform (FT), provides both phase and amplitude information which can enhance defect detectability and reduce noise. On the other hand, ARDVB [23], EVBTF [24] and SVD-RARX [25] have been selected from the state-of-the-art unsupervised detection algorithms. It can be seen from Figure 5, EVBTF and the proposed method shows reasonable results by processing original thermal sequences.…”

Section: Results Analysismentioning

confidence: 99%

“…To achieve automatic crack detection and identification for the experimental data from the ECPT system, a blind source separation algorithm was reported [20]. Methods based on sparse decomposition exhibited their robustness for both man-made specimens and samples with natural defects [21][22][23][24]. These methods assume that regions with defects are areas with the highest sparsity, while the low-rank matrix, which is considered as background, is separated to extract sparse components.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pattern Deep Region Learning for Crack Detection in Thermography Diagnosis System

Gao

et al. 2018

Metals

Self Cite

View full text Add to dashboard Cite

Eddy Current Pulsed Thermography is a crucial non-destructive testing technology which has a rapidly increasing range of applications for crack detection on metals. Although the unsupervised learning method has been widely adopted in thermal sequences processing, the research on supervised learning in crack detection remains unexplored. In this paper, we propose an end-to-end pattern, deep region learning structure to achieve precise crack detection and localization. The proposed structure integrates both time and spatial pattern mining for crack information with a deep region convolution neural network. Experiments on both artificial and natural cracks have shown attractive performance and verified the efficacy of the proposed structure.

show abstract

Section: Results Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Pattern Deep Region Learning for Crack Detection in Thermography Diagnosis System

Gao

et al. 2018

Metals

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [39], Feng et al proposed a hybrid algorithm based on the TSR and region growing technique for the task of debond detection in the CFRP composites. In [40], Peng et al proposed a multilayer architecture utilizing the ensemble variation based tensor factorization (EVBTF). The algorithm is tested for debond detection in CFRP composites.…”

Section: Introductionmentioning

confidence: 99%

“…In order to show its efficacy, the algorithm is conducted for debond defects detection in a different structure of CFRP composites. The visual analysis along with F-score [40] comparison are presented with generally used OPTNDT algorithms. In addition, the proposed algorithm is validated on the synthetic data with different noise configurations.…”

Section: Introductionmentioning

confidence: 99%

Ensemble Joint Sparse Low-Rank Matrix Decomposition for Thermography Diagnosis System

Ahmed

Gao

Woo

et al. 2021

IEEE Trans. Ind. Electron.

Self Cite

View full text Add to dashboard Cite

“…Pulsed phase thermography (PPT) [9][10] is a method for transforming time domain features into frequency domain, extracting defect information from frequency domain information, and eliminating noise. According to the sparsity characteristic of the infrared heat map with defects, Gao B et al [11][12][13] proposed an adaptive variational Bayesian method. These methods assume that the defect areas are sparse distributed and the background is low rank matrix.…”

Section: Introductionmentioning

confidence: 99%

Temporal and spatial deep learning network for infrared thermal defect detection

Luo

Gao

Woo

et al. 2019

NDT & E International

Self Cite

121

View full text Add to dashboard Cite

Most common types of defects for composite are debond and delamination. It is difficult to detect the inner defects on a complex shaped specimen by using conventional optical thermography nondestructive testing (NDT) methods. In this paper, a hybrid of spatial and temporal deep learning architecture for automatic thermography defects detection is proposed. The integration of cross network learning strategy has the capability to significantly minimize the uneven illumination and enhance the detection rate. The probability of detection (POD) has been derived to measure the detection results and this is coupled with comparison studies to verify the efficacy of the proposed method. The results show that visual geometry group-Unet (VGG-Unet) cross learning structure can significantly improve the contrast between the defective and non-defective regions. In addition, investigation of different feature extraction methods in which embedded in deep learning is conducted to optimize the learning structure. To investigate the efficacy and robustness of the proposed method, experimental studies have been carried out for inner debond defects on both regular and irregular shaped carbon fiber reinforced polymer (CFRP) specimens.

show abstract

Ensemble variational Bayes tensor factorization for super resolution of CFRP debond detection

Cited by 21 publications

References 27 publications

Pattern Deep Region Learning for Crack Detection in Thermography Diagnosis System

Pattern Deep Region Learning for Crack Detection in Thermography Diagnosis System

Ensemble Joint Sparse Low-Rank Matrix Decomposition for Thermography Diagnosis System

Temporal and spatial deep learning network for infrared thermal defect detection

Contact Info

Product

Resources

About