As a non-destructive testing technology, γ-photon imaging technology has a good application prospect in the industrial field. However, due to the γ-photon scattering problem caused by high-density metal and the rapid industrial detection requirement, γ-photon imaging technology has not been effectively applied in industrial detection. Thus, this paper proposed a compensatory single scatter simulation (C-SSS) and combined it with the 3D ordered subset expectation maximization (3D-OSEM) image reconstruction algorithm for optimization to solve the above problems. To verify the effectiveness of the C-SSS algorithm, this paper compared C-SSS with the single scatter simulation by two types of phantoms and the inner cavity detection of industrial components. The comparison results proved that C-SSS can effectively improve the measurement accuracy and reduce the scanning time. In addition, after the algorithm code was optimized and accelerated by the graphic processing unit, the execution time of C-SSS was ∼4 min and a single iteration of 3D-OSEM took ∼3 min, basically satisfying the requirements of rapid industrial detection. Simulation and experiment results show the potential of the proposed algorithm in industrial detection.

中文翻译：

---

工业γ光子3D成像方法的补偿散射校正研究

γ光子成像技术作为一种无损检测技术，在工业领域具有良好的应用前景。然而，由于高密度金属引起的γ-光子散射问题和快速的工业检测要求，γ-光子成像技术尚未有效地应用于工业检测。因此，本文提出了一种补偿性单散射仿真（C-SSS），并将其与3D有序子集期望最大化（3D-OSEM）图像重建算法结合起来，以解决上述问题。为了验证C-SSS算法的有效性，本文将C-SSS与通过两种体模进行的单散射仿真和工业组件的内腔检测进行了比较。比较结果表明，C-SSS可以有效提高测量精度，减少扫描时间。另外，通过图形处理单元对算法代码进行优化和加速后，C-SSS的执行时间约为4分钟，而3D-OSEM的单次迭代约为3分钟，基本满足了快速工业检测的要求。仿真和实验结果表明了该算法在工业检测中的潜力。