The design, construction and evaluation of a resolution target for microfilming application

By designing and constructing a new resolution target for microfilming applications, it has been the goal of this thesis to overcome several problems suspected to be associated with existing methods, namely, 1) the problem of pre-knowledge of the test pattern and target array, 2) the ambiguity in determining the smallest distinguishable pattern, 3) the possibility of spurious resolution due to periodicity in the test pattern and 4) the lack of correlation with alpha-numeric typefaces. The new Psi target system was constructed through a series of artwork reductions, paste-ups and further reductions, all on lithographic material. The Psi target, a system based on the orientation of the test patterns, was tested side-by-side in a micro filming application with the German DIN microfilming target 19051, a similar orientation system. The goal of the experimentation was to compare the merits of the two systems and to establish whether one could justifiably replace the other. The variance of information capacity associated with seven viewers\u27 evaluation of the Psi target was compared with the variance in the same viewers\u27 evaluation of the DIN target images. This comparison was in the form of a statistical hypothesis test. In the case of the four different sized pairs of images tested, there was no significant difference (in all four tests) between the variances of the two target systems at the 90% confidence level. The average information capacity of each image was calculated from the seven viewers\u27 evaluations of the four pairs of images. The information capacity of one target was then plotted against the other to determine the relationship between the two targets. A linear regression analysis was performed on these four data points and the equation of the best fit line was obtained. This equation, with a correlation coefficient of .97 is, I(DIN) = .31 I(Psi) + 12.47, where I is information capacity. The standard error of the .31 slope was calculated as .052. The final conclusion of the thesis is that, since both targets had their own faults and merits, and since both systems produce different results (as apparent from the above equation)} it is not suggested that one system replace the other, but rather that each stand as a distinct microfilm resolution system