本文采用拓展重叠关联迭代引擎算法,系统地模拟研究了x射线光斑尺寸、扫描点数及波前形貌对扫描相干衍射成像重建图像质量的影响。结果表明,在实际入射光斑尺寸不易确定时,算法中初始猜测光斑尺寸可以由模拟入射光斑最大光强值的1%对应边界确定。对物体进行扫描时,入射光斑个数大于7×7且重叠度优于70%可以重建出较好的图像。入射光斑的振幅和位相分布对重建图像质量也有着显著影响。
Based on the extended ptychographical iterative engine (ePIE) algorithm, the effects of the incident probe size, scanning position number, and wavefront morphology on the reconstructed image were systematically studied with numerical simulation. The results show that the guessed initial probe size in ePIE can be determined by the corresponding boundary of 1% of maximum intensity of the simulated incident probe when the real incident beam size is difficult to determine exactly. When imaging a sample by PCDI, a better image can be reconstructed by choosing the number of scanning points as more than 7×7 and a scanning overlap ratio as larger than 70%. The amplitude distributions and incident spot phase also have significant effects on the reconstructed image quality