The following work has been computer simulated. A cumulative gaussian step response and the step response of a photographic emulsion 1 were taken through the derivative-transform edge gradient analysis technique and Tatian's method of analysis. Random noise levels, truncation intervals and sampling intervals on the analytic edges were varied to determine their influence on each technique. The variance and means of the calculated m.T.F.s were then statistically tested for no difference of the two techniques. The exact noise free M.T.F. was also calculated and compared to the M.T.F.s calculated by the two techniques. The results show a statistical difference in the two techniques at low frequencies. This difference was deemed not significant from a practical standpoint because the magnitude of the differences between the M.T.F.s was small compared to the actual magnitude of the M.T.F.s at low frequencies. Also the two techniques produced M.T.F.s of high deviations from the exact noise free transform at high frequencies.
Library of Congress Subject Headings
Department, Program, or Center
School of Photographic Arts and Sciences (CIAS)
LaFlesh, Robert, "Comparison of the derivative-transform edge gradient analysis technique to tatian's method of edge gradient analysis, varying random noise, truncation interval and sampling internal on analytical edges" (1980). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus