Large depth of field Fourier ptychographic microscopy based on energy-oriented adaptive step-size strategy

Fourier ptychographic microscopy (FPM) enables imaging with both a large field of view (FOV) and high resolution (HR), which is crucial for life sciences, biomedical research, and pathological diagnosis. Traditional FPM models assume the sample is a two-dimensional thin object, where different illumination angles result in shifts of the sample's 2D spectrum. However, in practical pathological imaging, samples often have thickness, contradicting the 2D object model and degrading FPM's reconstruction quality for thick samples, thus hindering its use in pathological diagnosis. To address this issue, this paper proposes an energy-oriented adaptive step-size Fourier ptychographic microscopy (EA-FPM) model. Based on the multi-layer FPM model and specific depth-of-field (DOF) requirements for digital pathology, this approach reduces the number of layers and uses an energy-oriented coefficient matrix for each layer. The coefficient matrix is determined by the intensity contrast at the focal plane during iteration. At last, EA-FPM speeds up convergence, ultimately reconstructing an image with a large DOF. Compared to traditional FPM models, this approach significantly extends the DOF without increasing the number of captured images, while maintaining the imaging FOV and resolution.