Multi-focus microscopic image fusion algorithm based on sparse representation and pulse coupled neural network

Microscopic image with limited depth of field in the microscopic system is distinct only in finite region. To break through the limitation of field depth and obtain amplified distinct microscopic images with ultra-depth of field, multi-focus microscopic image fusion algorithm based on sparse representation and pulse coupled neural network is put forward in this essay. First, non-subsampling shear wave transformation is adopted to decompose the microscopic image to obtain high and low frequency components. The low-frequency components are fused with modified sparse representation, while the high-frequency components are fused with modified pulse coupled neural network. In the end, NSST is conducted to obtain the fused microscopic image. The simulation results show that whether it is assessed subjectively or objectively, compared with the other three fusion algorithms, the proposed algorithm can retain better the edge contour, details and texture information of the multi-focus source microscopic image, with no missing details, which has improved the contrast and clarity of microscopic images.