基于成像矩阵的物数据求解方法

Solving method of object data based on imaging matrix

提出了一种利用成像矩阵求解物数据的方法。通过对物、像矩阵的一维化处理,构建了二维成像矩阵,将物矩阵卷积点扩散函数矩阵得到像矩阵的卷积运算关系转变为物向量乘以成像矩阵进而得到像向量的乘积运算关系。最后将像向量乘以成像逆矩阵便可求解物向量,实现了从光学系统像数据中对物数据的精确求解,且计算误差为10-12量级。

A method for solving object data based on imaging matrix is proposed. Through one-dimensional processing of the object and image matrixes, a two-dimensional imaging matrix is constructed. Thus,the operation rules of image matrix by convolution of object matrix and point spread function matrix were transformed into the image vector by one of the product of objectvector and imaging matrix. Finally,the image vector is obtained by multiplying theinverse matrix, precise solution of the object data from the image data of optical sysized and the calculation error can reach the order of 10^-12.