所附代码提供了上述三种反卷积策略的实现。要开始，用户应该检查文件demo.m。反褶积子程序在以下文件中实现:deconvL2频率.m:在频域反褶积，假设高斯导数在前面。2.deconvL2.m:利用共轭梯度算法，假设在空间域上存在高斯导数先验的反褶积。3.deconvSps.m:利用迭代重加权最小二乘算法，在空间域上假设一个稀疏导数先验的反褶积。(The enclosed code provides the implementation of the above three deconvolution strategies. To start, users should check the file demo.m. The deconvolution subroutine is implemented in the following files: deconvL2 frequency. M: deconvolution in frequency domain, assuming that the Gauss derivative is ahead. 2. DeconvL2. M: uses the conjugate gradient algorithm to assume the existence of Gauss derivative prior deconvolution in the spatial domain. 3.deconvSps. M: Using iterative weighted least squares algorithm, a sparse derivative prior deconvolution is assumed in the spatial domain. Users should note that all of the above functions assume that the convolution filter has an odd number of pixels on both dimensions.)