基于CT序列图像的旋转体积测量方法

针对工程应用中常见的轴向变化剧烈工件，研究一种Zernike矩亚像素层间拟合轮廓旋转积分的高精度体积测量算法。首先用Zernike矩方法提取亚像素级轮廓，其次采用多项式拟合切片轮廓，并进行等相角间隔采样；然后对等相位的轮廓点采用多项式拟合并通过外插法自动预测工件顶端，将各拟合轮廓旋转一个相角，对旋转块积分并累加得到待测工件的体积。分别对体积相等的3个试块进行CT扫描测量实验，结果表明该算法的测量误差低于0.7%；对实际含内腔工件进行测量实验，结果表明该算法测量精度是Canny台体法的1.5倍。

Rotation volume measurement method based on CT sequence images

According to the common engineering workpiece with dramatic axial contour changes, a high-accuracy inter slice contour rotating integral volume measuring algorithm of Zernike moment sub-pixel edge detection is presented. Firstly, a sub-pixel level contour is detected by Zernike moment method. Secondly, using the polynomial fitting method the contour is fitted, and then the fitted contour is sampled by the same angle interval. Each edge point at the same phase angle of each slice is used to fit the contour, and then the apex of work-piece is calculated through extrapolation method. Finally, each fitted contour is rotated through one phase angle interval to calculate the integral as a rotation volume, the volume of each rotation is accumulated to be the total volume of the workpiece. Volumetric measurement experiments are carried out towards three volume-equal testing workpieces. The results indicate that the measurement errors are less than 0.7% using the proposed method. An actual work-piece with internal cavity is measured. The result indicates that the measurement accuracy of the proposed method is 1.5 times than the one of the Canny-frustum method.