Imaging of distant objects in a terrestrial environment involves propagation of light through significant turbulence. Conventional methods for imaging for these applications are radar or focal plane imaging. Both of these methods have limitations, such as object rotation rate and post-processing. A different class of imaging approaches involving coherent illumination has several advantages (a) reduced sensitivity to illumination conditions, (b) reduced sensitivity to object rotation, (c) the use of arrayed receivers that are lighter and lower in cost, and (d) snapshot reconstruction of aberrated images with just one or a few frames. These advantages must be balanced by the challenges of speckle noise in the image reconstructions, and more difficult algorithms. Coherent pupil-plane and focal-plane techniques are investigated for image formation. The associated algorithms include root-reconstruction techniques, phase smoothing methods, polynomial fit approaches, blind iterative deconvolution, and multi-frame blind deconvolution (MFBD). These techniques and algorithms are surveyed for speed and quality of image formation.
|