Coherent speckle noise is modeled as a multiplicative noise process that has a negative exponential probability density function. Using a homomorphic transfor mation, this speckle noise is converted to a signal-independent, additive process. The speckled images are randomly jittered from frame-to-frame against a uniform background to simulate image motion and/or platform jitter. Multiple images are logarithmically transformed and ensemble averaged in the bispectral domain. The bispectrum ignores this image motion so no blurring results from the ensemble averaging. Object Fourier magnitude and phase information are also retained in the bispectrum so that the resultant image can be uniquely reconstructed. This value is then exponentiated to complete the image reconstruc tion process. Since speckle masks the resolution of details in the noisy image and effectively destroys the object structure within the image, it is seen that image reconstruction using bispectrum estimation results in images that regain their object structure. Both one-dimensional and two-dimensional images were tested using separate bispectral signal reconstruction algorithms for each.
Library of Congress Subject Headings
Image processing; Speckle; Optical data processing
Department, Program, or Center
School of Photographic Arts and Sciences (CIAS)
Wear, Steven M., "Shift-invariant image reconstruction of speckle-degraded using bispectrum estimation" (1990). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus