一种基于超声C扫描成像原理的图像边缘检测方法

针对超声C扫描图像中缺陷边缘的模糊问题，通过建立超声C扫描成像的数学模型，分析了异种材料结合界面超声波反射能量的分布情况。结果表明，由于聚焦探头焦点直径的存在．使得结合界面中缺陷边缘两侧反射声波能量发生变化，从而引起了超声C扫描图像中缺陷边缘的模糊现象。基如此，提出了一种快速确定C扫描图像缺陷边缘的方法。对平底孔超声C扫描图像边缘检测的试验结果表明，该方法适用于超声C扫描图像的边缘检测．明显优于Sobel算法和Canny算法，能够快速准确检测出缺陷边缘。

Method to Detect Edges of the Defects in C-scan Image Based on Mechanism of Ultrasonic Imaging

The mathematical model of ultrasonic C-scan image was built and the energy of the ultrasonic waves reflected from the bonding interface between two different materials was analyzed. It shows that the physical limit of focused transducer with circular focus causes the energy variation of reflected wave near the defect boundary. The energy distribution results in the blur of edge in C-scan image. Based on the analysis, a method to detect the edge of defect in C-scan image was proposed. The edge of defect was detected according to the variation of the gray values near the edge, which is determined by the energy distribution. The method was performed on the C-scan image of flat-bottom hole. The experimental results show that the defect edge can be detected rapidly and accurately. It is more adaptive for the edge detection of ultrasonic C-scan image than Sobel algorithm and Canny algorithm.