Recently, a method has been developed that uses computer vision to determine the deformations of subsets of an object. Although the method has been used successfully in a variety of applications, to date there has been no critical assessment of the key parameters in the system and their effect on the accuracy of the measured deformations. The present work presents the results of initial studies of this system. The system components are modeled, and a representative intensity pattern is chosen and deformed by known amounts. Then, the effects of varying the various parameters in the model are analyzed numerically. The most significant parameters are found to be (1) the number of quantization levels in the digitization process (i.e., the number of bits in the A/D converter), (2) the ratio of the frequency of the signal to the frequency of the sampling, and (3) the form of the intensity interpolation function.