minimize the noise transmitted through the matrix
主要用三个参数衡量 indicator
the condition number (CN)
the Equally Weighted Variance (EWV)
the error associated at every component of the Stokes vector
Whereas the CN quantifiles if the matrix A-1 is well-conditioned (i.e. far to singular), the EWV and the error associated at every component of the Stokes vector are related with the propagation of errors from the vector I to the solution S.
尽管 CN 合格 如果 矩阵条件比较好(远不是奇异),后两个参数与 I 到 S 的传播过程很有关
理解为: 通过CN 计算出的 A-1 越不是奇异矩阵越好,当满足这个条件 CN才是合格的。
CN
理论上CN的最小值为1,因为它代表一个幺正矩阵,且不会放大误差
由于 A 的每一行 row 相当于一个偏振片 ,可以看做是 Mueller 矩阵的第一行,对应着第一个光强参数
第一列的所有系数归一化后都为 1 ,所以A不会是一个幺正矩阵
?
所以 CN = 1 不可能得到
A 越接近幺正矩阵,CN 就越小
σ 代表不同于 0 的最大最小特征值 singular values
EWV
data redundancy 数据冗余
考虑到数据冗余
衡量了 I 到 S 的方差传递
The EWV indicator provides a useful estimation of the global error amplification in the solution vector S when some amount of noise is present in the intensity measurements vector I.
- Optimization polarimeter retarders 文献 criteriaoptimization polarimeter retarders文献 polarimeter retarders criteria categorize according leetcode criteria criteria pattern filter optimization optimization笔记convex optimization mixed-integer mathematical optimizer optimization perceptron machine support