声全息成象超分辨率处理
SUPERRESOLUTION PROCESSING FOR ACOUSTIC HOLOGRAPHIC IMAGING
-
摘要: 本文提出一种声全息成象超分辨率处理方法,它利用物分布有限及由此导出的全息矩阵元素间联系,由已知的全息数据延拓得到扩大的全息数据,从而提高系统成象的分辨率。文中导出了超分辨率处理的计算公式,并讨论了实际计算中得到物分布先验知识的方法。还指出这种超分辨率处理方法可以迭代使用,从而在无噪声的情况下,从理论上能无限提高系统分辨率。作了计算机模拟计算,说明方法是有效的。最后讨论了处理方法对噪声的敏感问题,给出了一些情况下输入信噪比对输出信噪比的影响曲线,说明这方法适宜于物分布角范围较小的情况。Abstract: Superresolution processing for acoustic holographic imaging, which is based on the finite region of the object distribution and utilizes the relation between elements of holographic matrix from it, is proposed to increase the resolution of the imaging system. The calculating formulas are derived and knowledge of obtaing the object distribution in actual calculation is discussed. The iteration of the method can be applied to increase indefinitely the resolution of the imaging system when noise is absent. Simulation with a computer is made, which proves the validity of the method. Finally, sensitivity of the method to noise is discussed, and the curves of output noise to signal ratio versus input noise to signal radio are given for some examples, which show that the method is particulaly suitable to small object distribution.