基于特征重建的工业CT算法

Study on Industrial CT Algorithm Based on Characteristic Reconstruction

工业CT检测系统中,散射等噪声约束条件较多,传统的CT重建方法得到的重建图像噪声大、分辨率低,不适合对一些较小的疵病进行检测.文章依赖被检测对象内部结构特征,利用小波函数在RADON变换前后的特性,通过对投影数据预滤波,直接重建图像的特征部位,在不损失太多图像背景分辨率的情况下,凸显了被检测对象内部较小的疵病特征,提高了检测分辩率.

In the industrial CT detection systems, owing to the restrictions of noises sucn as scattering noise, the traditional methods for CT reconstruction result in the reconstructed image with high noise and low resolution so that they cannot detect small flaws. Relying on the inner structure character of detected object, according to the character of wavelet function before and after RADON transform, the characteristic part of the image is reconstructed by pre-filtering the projected data. The method can protrude the inner small flaw of detected object with the less sacrifice of the resolution of image backgruoud. The method can commendably increase the detect resolution.