MODEL-BASED INVERSION OF SPECKLE INTERFEROMETER FRINGE PATTERNS

Micrometer-scale rigid-body translations are determined from electroni c speckle interferometric fringe patterns. An iterative minimum error procedure employs the relative hinge order of picked positions of frin ge maxima and minima within a single interferogram to calculate the di splacement field directly. The method does not calculate the displacem ent at a single point but relies on the assumption that the character, but not the magnitudes or directions, of the displacements over the v iewing area of the interferogram is known. That is, a model of the dis placements exists. On perfect,noise-free forward modeled fringe patter ns calculated for an 8.0-mu m displacement, the phase error is less th an 2 x 10(-6) fringe orders (1.3 x 10(-5) rad) and probably results on ly from numerical noise in the inversion. On real fringe patterns obta ined in electronic speckle interferometric experiments, mean phase err ors are generally less than 5 x 10(-5) fringe orders (3.2 x 10(-4) rad ), suggesting that;the technique is robust despite errors resulting fr om speckle noise, lack of accuracy in positioning of experimental comp onents, and image-distortion corrections. (C) 1998 Optical Society of America.