• Optics and Precision Engineering
  • Vol. 24, Issue 6, 1520 (2016)
WANG Jue1,2, CHENG Yan1, CAI Yu-fang2, and WANG Hui-Qian2
Author Affiliations
  • 1[in Chinese]
  • 2[in Chinese]
  • show less
    DOI: 10.3788/ope.20162406.1520 Cite this Article
    WANG Jue, CHENG Yan, CAI Yu-fang, WANG Hui-Qian. SART + TVM reconstruction algorithm based on improved Ray-Box Intersection weight coefficient matrix[J]. Optics and Precision Engineering, 2016, 24(6): 1520 Copy Citation Text show less

    Abstract

    Based on the problem of uniform sampling and sparse reconstruction, a research on calculation method for two weight coefficient matrix, including Ray-box Intersection and traditional bilinear interpolation, was conducted. Improvement on logic flaws of Ray-Box Intersection was made, and the improved algorithm was put forward by combining line generating method with the slope of the ray and pixel location. In this algorithm, the starting point and terminal point of the ray across the reconstruction region can be obtained through algebraic relation, and then the maximum integer no greater than its coordination is used to, in accordance with the ray slope along the X direction and Y direction, work out the lower left coordination collection of all pixel blocks passed through by the ray in the projection area using the line generating method. For each pixel block, the distance of intersection point between the ray and pixel block is their weighted value. It indicates that compared with the weight coefficient matrix algorithm of Ray-Box Intersection, speed of the improved one has increased by 96%, and the peak signal to noise ratio has increased by 19%. In comparison with the bilinear interpolation algorithm, rate of convergence of the improved one is faster, and the peak signal to noise ratio has increased by 5% under the same time period. The improved Ray-Box Intersection algorithm can achieve better reconstructed images and clearer details.
    WANG Jue, CHENG Yan, CAI Yu-fang, WANG Hui-Qian. SART + TVM reconstruction algorithm based on improved Ray-Box Intersection weight coefficient matrix[J]. Optics and Precision Engineering, 2016, 24(6): 1520
    Download Citation