【摘要】 针对传统相位解包裹算法在生物细胞相位重构上遇到的困难,本文提出了一种改进的质量图导向解包算法。该方法以质量图导向法为基础,通过对质量图设置阈值来识别解包后可能出现异常跳变的像素点,然后对初始解包后的有效相位值分别进行横向和纵向一维多项式拟合,最后将异常像素点的相位值用拟合值的平均值进行代换即可重构细胞的相位。模拟结果表明该方法具有较高的解包精度,实验结果验证了该方法在实际测量中的可行性。更多还原

【Abstract】 An improved quality-guided phase unwrapping method is presented in order to reconstruct the phases of biological cells,which is not be unwrapped by traditional methods. The improved method is based on quality-guided phase unwrapping method. Firstly,some abnormal pixels can be identified by setting threshold for the quality map. Then one-dimensional polynomial fitting algorithm is used to fit the effective pixels horizontally and vertically. At last,the unwrapped phases of abnormal pixels can be replaced by the average of the fitted values. The simulation result shows that the method has higher precision. Also,the feasibility is verified by experiment results.更多还原