In this paper, the use of a signal to noise ratio (SNR) is proposed for the quantification of the goodness of some selected processing techniques of thermographic images, such as differentiated absolute contrast, skew...In this paper, the use of a signal to noise ratio (SNR) is proposed for the quantification of the goodness of some selected processing techniques of thermographic images, such as differentiated absolute contrast, skewness and kurtosis based algorithms, pulsed phase transform, principal component analysis and thermographic signal reconstruction. A new hybrid technique is also applied (PhAC—Phase absolute contrast), it combines three different processing techniques: phase absolute contrast, pulsed phase thermography and thermographic signal reconstruction. The quality of the results is established on the basis of the values of the parameter SNR, assessed for the present defects in the analyzed specimen, which enabled to quantify and compare their identification and the quality of the results of the employed technique.展开更多
In the ultrasonic nondestructive evaluation of the quality of solid state welded joints, such as friction bonding and diffusion bonding, the main difficulty is the identification of micro defects which are most likel...In the ultrasonic nondestructive evaluation of the quality of solid state welded joints, such as friction bonding and diffusion bonding, the main difficulty is the identification of micro defects which are most likely to emerge in the welding process. The ultrasonic echo on the screen of a commercial ultrasonic detector due to a micro defect is so weak that it is completely masked by noise, and impossible to be pointed out. In the present paper, wavelet analysis (WA) is utilized to process A scan ultrasonic signals from weak bonding defects in friction bonding joints and porosity in diffusion bonding joints. First, perception of WA for engineers is given, which demonstrates the physical mechanism of WA when applied to signal processing. From this point of view, WA can be understood easily and more thoroughly. Then the signals from welding joints are decomposed into a time scale plane by means of WA. We notice that noise and the signal echo attributed to the micro defect occupy different scales, which make it possible to enhance the signal to noise ratio of the signals by proper selection and threshold processing of the time scale components of the signals, followed by reconstruction of the processed components.展开更多
文摘In this paper, the use of a signal to noise ratio (SNR) is proposed for the quantification of the goodness of some selected processing techniques of thermographic images, such as differentiated absolute contrast, skewness and kurtosis based algorithms, pulsed phase transform, principal component analysis and thermographic signal reconstruction. A new hybrid technique is also applied (PhAC—Phase absolute contrast), it combines three different processing techniques: phase absolute contrast, pulsed phase thermography and thermographic signal reconstruction. The quality of the results is established on the basis of the values of the parameter SNR, assessed for the present defects in the analyzed specimen, which enabled to quantify and compare their identification and the quality of the results of the employed technique.
基金This work is financially supported by the Beijing Natural Science Foundation!(No.2 962 0 0 4 )
文摘In the ultrasonic nondestructive evaluation of the quality of solid state welded joints, such as friction bonding and diffusion bonding, the main difficulty is the identification of micro defects which are most likely to emerge in the welding process. The ultrasonic echo on the screen of a commercial ultrasonic detector due to a micro defect is so weak that it is completely masked by noise, and impossible to be pointed out. In the present paper, wavelet analysis (WA) is utilized to process A scan ultrasonic signals from weak bonding defects in friction bonding joints and porosity in diffusion bonding joints. First, perception of WA for engineers is given, which demonstrates the physical mechanism of WA when applied to signal processing. From this point of view, WA can be understood easily and more thoroughly. Then the signals from welding joints are decomposed into a time scale plane by means of WA. We notice that noise and the signal echo attributed to the micro defect occupy different scales, which make it possible to enhance the signal to noise ratio of the signals by proper selection and threshold processing of the time scale components of the signals, followed by reconstruction of the processed components.
