基于相位相干成像的TOFD检测缺陷图像增强处理研究

Enhancement of Defect Image in Ultrasonic TOFD Detection Based on Phase Coherence Imaging

利用超声衍射时差法(TOFD)对钢制安全壳(CV)焊缝进行缺陷检测时,存在信噪比低和分辨率差等现象。为解决上述问题,发展了一种基于相位相干成像(PCI)的TOFD的图像增强处理算法,该算法适用于平行扫查(B扫查)与非平行扫查(D扫查),针对B扫查和D扫查图像从增强算法原理及其试验效果进行了说明,为后续缺陷定位、定量研究奠定理论基础。首先,利用合成孔径聚焦技术(SAFT)对厚度48 mm的CV试块焊缝中底面开口槽B扫查和D扫查图像进行处理,得到SAFT图像;然后,利用PCI算法对SAFT图像进行相位相干处理。结果表明,PCI算法能够通过减小合成孔径声束宽度有效提高横向分辨率。此外,该算法还能够通过放大相位分布对幅值的作用有效抑制背景噪声,使缺陷信噪比得到提升。并在上述研究基础之上,比较了CCF和SCF权重矩阵增强效果,为实际图像处理中权重矩阵的选择提供参考。

Using time of flight diffraction（TOFD） to detect flaws in the weld of containment vessel（ CV）, the TOFD image has poor lateral resolution and low signal-to-noise ratio （SNR） , so the phase coherence imaging（PCI） algorithm was applied to the enhancement of image. Firstly, the synthetic aperture focusing technique （SAFT） were used to the enhance the B-scan and D-scan image of bottom surface notch which in the 48 mm thick CV block and SAFT image was obtained. Then the SAFT image was processed by PCI algorithm with weighting matrix. The results showed that the PCI algorithm made a significant improvement of lateral resolution by means of narrowing the beam width of synthetic aperture. Besides, the background noise was suppressed with the help of PCI algorithm by amplifying the effect of phase distribution to signal amplitude and the SNR was improved. Based on the above research, the comparison of CCF and SCF matrix enhancement was made, and it can provide reference for the choice of weighting matrix in practical image processing.