Interlayer Interpolation Method of Two-Dimensional Slice Image Based on Dynamic Programming and Contour Factor

In this paper, the problem of image edge blurring caused by insufficient interlayer information in thick CT images in solid rocket motor grain defect detection is discussed. The existing methods usually adopt static interpolation method, and the interpolation effect is not ideal in the nonlinear changing region, and the edge is unnatural. In this method, the similarity weight coefficient is obtained by calculating the contour area of defects, and the interpolation steps are dynamically adjusted by the similarity weight coefficient, and the dynamic interpolation calculation is carried out to optimize the interpolation quality. The experimental results show that the proposed method can effectively alleviate the transition stiffness problem of traditional linear interpolation methods in areas with significant contour changes, and the standard deviation of curvature distribution of the model is reduced by 85%, and the consistency of normal vectors is improved to 0.991. This method significantly improves the image quality and provides a more accurate basis for subsequent image registration and defect detection.