Defect Classification

14,000,000 Leading Edge Experts on the ideXlab platform

Scan Science and Technology

Contact Leading Edge Experts & Companies

Scan Science and Technology

Contact Leading Edge Experts & Companies

The Experts below are selected from a list of 11067 Experts worldwide ranked by ideXlab platform

Gui Yun Tian - One of the best experts on this subject based on the ideXlab platform.

  • frequency optimization for enhancement of surface Defect Classification using the eddy current technique
    Sensors, 2016
    Co-Authors: Mengbao Fan, Qi Wang, Binghua Cao, Ali Imam Sunny, Gui Yun Tian
    Abstract:

    Eddy current testing is quite a popular non-contact and cost-effective method for nondestructive evaluation of product quality and structural integrity. Excitation frequency is one of the key performance factors for Defect characterization. In the literature, there are many interesting papers dealing with wide spectral content and optimal frequency in terms of detection sensitivity. However, research activity on frequency optimization with respect to characterization performances is lacking. In this paper, an investigation into optimum excitation frequency has been conducted to enhance surface Defect Classification performance. The influences of excitation frequency for a group of Defects were revealed in terms of detection sensitivity, contrast between Defect features, and Classification accuracy using kernel principal component analysis (KPCA) and a support vector machine (SVM). It is observed that probe signals are the most sensitive on the whole for a group of Defects when excitation frequency is set near the frequency at which maximum probe signals are retrieved for the largest Defect. After the use of KPCA, the margins between the Defect features are optimum from the perspective of the SVM, which adopts optimal hyperplanes for structure risk minimization. As a result, the best Classification accuracy is obtained. The main contribution is that the influences of excitation frequency on Defect characterization are interpreted, and experiment-based procedures are proposed to determine the optimal excitation frequency for a group of Defects rather than a single Defect with respect to optimal characterization performances.

  • Reduction of lift-off effects in pulsed eddy current for Defect Classification
    2011
    Co-Authors: Yunze He, Feilu Luo, Mengchun Pan, Gui Yun Tian
    Abstract:

    Pulsed eddy-current (PEC) testing is an electromagnetic nondestructive testing & evaluation (NDT&E) technique and Defect Classification is one of the most important steps in PEC Defect characterization. With pulse excitation, the PEC response signals contain more features in time domain and rich information in frequency domain. This paper investigates feature extraction techniques for PEC Defect Classification including rising time, differential time to peak and differential time to zero, spectrum amplitude, and differential spectrum amplitude. Experimental study has been undertaken on Al-Mn 3003 alloy samples with artificial surface Defects, sub-surface Defects, and Defects in two-layer structures under different lift-off. Experimental results show that methods are effective to classify the Defects both in single-layer structures and two-layer structures. Comparing the results of different methods, it is found that differential process can eliminate the lift-off in Defect Classification in both time domain and frequency domain. The study can be extended to Defect Classification in complex structures, where lift-off effects are significant.

  • feature extraction and selection for Defect Classification of pulsed eddy current ndt
    Ndt & E International, 2008
    Co-Authors: Gui Yun Tian, Ali Sophian, Tianlu Chen, Pei Wen Que
    Abstract:

    Pulsed eddy current (PEC) is a new emerging nondestructive testing (NDT) technique using a broadband pulse excitation with rich frequency information and has wide application potentials. This technique mainly uses feature points and response signal shapes for Defect detection and characterization, including peak point, frequency analysis, and statistical methods such as principal component analysis (PCA). This paper introduces the application of Hilbert transform to extract a new descending feature point and use the point as a cutoff point of sampling data for detection and feature estimation. The response signal is then divided by the conventional rising, peak, and the new descending points. Some shape features of the rising part and descending part are extracted. The characters of shape features are also discussed and compared. Various feature selection and integrations are proposed for Defect Classification. Experimental studies, including blind tests, show the validation of the new features and combination of selected features in Defect Classification. The robustness of the features and further work are also discussed.

  • wavelet based pca Defect Classification and quantification for pulsed eddy current ndt
    Iet Science Measurement & Technology, 2005
    Co-Authors: Gui Yun Tian, Ali Sophian, David Taylor, J Rudlin
    Abstract:

    A new approach for Defect Classification and quantification by using pulsed eddy current sensors and integration of principal component analysis and wavelet transform for feature based signal interpretation is presented. After reviewing the limitation of current parameters of peak value and its arrival time from pulsed eddy current signals, a two-step framework for Defect Classification and quantification is proposed by using adopted features from principal component analysis and wavelet analysis. For Defect Classification and quantification, different features have been extracted from the pulsed eddy current signals. Experimental tests have been undertaken for ferrous and non-ferrous metal samples with manufactured Defects. The results have illustrated the new approach has better performance than the current approaches for surface and sub-surface Defect Classification. The Defect quantification performance, which is difficult by using current approaches, is impressive.

  • Defect Classification using a new feature for pulsed eddy current sensors
    Ndt & E International, 2005
    Co-Authors: Gui Yun Tian, Ali Sophian
    Abstract:

    The objective of this study is to identify Defects such as surface cracks, subsurface Defects and metal losses using feature based pulsed eddy current sensors. A new feature, termed as the rising point, related to the propagation time of electromagnetic waves in metallic targets is proposed for Defect Classification. Experimental studies of the validation, robustness of the new feature of rising time are reported. In addition to other features, Defects can be detected and quantified robustly and lift-off can also be derived from the rising time. Conclusion and further work are derived on the basis of the findings.

Ali Sophian - One of the best experts on this subject based on the ideXlab platform.

  • feature extraction and selection for Defect Classification of pulsed eddy current ndt
    Ndt & E International, 2008
    Co-Authors: Gui Yun Tian, Ali Sophian, Tianlu Chen, Pei Wen Que
    Abstract:

    Pulsed eddy current (PEC) is a new emerging nondestructive testing (NDT) technique using a broadband pulse excitation with rich frequency information and has wide application potentials. This technique mainly uses feature points and response signal shapes for Defect detection and characterization, including peak point, frequency analysis, and statistical methods such as principal component analysis (PCA). This paper introduces the application of Hilbert transform to extract a new descending feature point and use the point as a cutoff point of sampling data for detection and feature estimation. The response signal is then divided by the conventional rising, peak, and the new descending points. Some shape features of the rising part and descending part are extracted. The characters of shape features are also discussed and compared. Various feature selection and integrations are proposed for Defect Classification. Experimental studies, including blind tests, show the validation of the new features and combination of selected features in Defect Classification. The robustness of the features and further work are also discussed.

  • wavelet based pca Defect Classification and quantification for pulsed eddy current ndt
    Iet Science Measurement & Technology, 2005
    Co-Authors: Gui Yun Tian, Ali Sophian, David Taylor, J Rudlin
    Abstract:

    A new approach for Defect Classification and quantification by using pulsed eddy current sensors and integration of principal component analysis and wavelet transform for feature based signal interpretation is presented. After reviewing the limitation of current parameters of peak value and its arrival time from pulsed eddy current signals, a two-step framework for Defect Classification and quantification is proposed by using adopted features from principal component analysis and wavelet analysis. For Defect Classification and quantification, different features have been extracted from the pulsed eddy current signals. Experimental tests have been undertaken for ferrous and non-ferrous metal samples with manufactured Defects. The results have illustrated the new approach has better performance than the current approaches for surface and sub-surface Defect Classification. The Defect quantification performance, which is difficult by using current approaches, is impressive.

  • Defect Classification using a new feature for pulsed eddy current sensors
    Ndt & E International, 2005
    Co-Authors: Gui Yun Tian, Ali Sophian
    Abstract:

    The objective of this study is to identify Defects such as surface cracks, subsurface Defects and metal losses using feature based pulsed eddy current sensors. A new feature, termed as the rising point, related to the propagation time of electromagnetic waves in metallic targets is proposed for Defect Classification. Experimental studies of the validation, robustness of the new feature of rising time are reported. In addition to other features, Defects can be detected and quantified robustly and lift-off can also be derived from the rising time. Conclusion and further work are derived on the basis of the findings.

  • Defect Classification using a new feature for pulsed eddy current sensors
    Ndt & E International, 2005
    Co-Authors: Gui Yun Tian, Ali Sophian
    Abstract:

    The objective of this study is to identify Defects such as surface cracks, subsurface Defects and metal losses using feature based pulsed eddy current sensors. A new feature, termed as the rising point, related to the propagation time of electromagnetic waves in metallic targets is proposed for Defect Classification. Experimental studies of the validation, robustness of the new feature of rising time are reported. In addition to other features, Defects can be detected and quantified robustly and lift-off can also be derived from the rising time. Conclusion and further work are derived on the basis of the findings.

Jochen G Hofstaetter - One of the best experts on this subject based on the ideXlab platform.

Lucian B Solomon - One of the best experts on this subject based on the ideXlab platform.

Tianlu Chen - One of the best experts on this subject based on the ideXlab platform.

  • feature extraction and selection for Defect Classification of pulsed eddy current ndt
    Ndt & E International, 2008
    Co-Authors: Gui Yun Tian, Ali Sophian, Tianlu Chen, Pei Wen Que
    Abstract:

    Pulsed eddy current (PEC) is a new emerging nondestructive testing (NDT) technique using a broadband pulse excitation with rich frequency information and has wide application potentials. This technique mainly uses feature points and response signal shapes for Defect detection and characterization, including peak point, frequency analysis, and statistical methods such as principal component analysis (PCA). This paper introduces the application of Hilbert transform to extract a new descending feature point and use the point as a cutoff point of sampling data for detection and feature estimation. The response signal is then divided by the conventional rising, peak, and the new descending points. Some shape features of the rising part and descending part are extracted. The characters of shape features are also discussed and compared. Various feature selection and integrations are proposed for Defect Classification. Experimental studies, including blind tests, show the validation of the new features and combination of selected features in Defect Classification. The robustness of the features and further work are also discussed.

Defect Classification - 15 days free trial to Access Experts & Abstracts