RESOLUTION ISSUES IN SCIENTIFIC PHOTOGRAPHY

RESOLUTION TESTS FILM, NIKON CP5000, SLIDE and XGA PROJECTORS

Prof.  John Hart

Program in Atmospheric and Oceanic Sciences

University of Colorado

Boulder, CO 80302

January 2002

hart@tack.colorado.edu

nimbus.colorado.edu/hart/science.htm

www.crystalcanyons.net

In our laboratory at the University of Colorado we have conducted numerous resolution tests.  The goal is to quantify the best way to display and the requirements for scientific imaging in various applications.

Below are magnifications of the central portion of a resolution test pattern using different cameras.

Figure 1.   Replica (via a high quality microscope) of a test pattern recorded on 35mm Fuji Provia film using prime lens optics at f11.  The highest resolved grid is indicated by the arrow (at column 1, row 3).  It has 75 line pairs per mm, or about 2700 line pairs across the 35mm film.  The contrast is about 10% (vs. 100% for pure black to pure white).  In the row 0 column 2 pattern the contrast is about 50% or 5 times greater.  

Figure 2.  Central element of the resolution chart when shot with a Nikon Coolpix 5000 5.2Megapixel digital camera.  In figure 1, only the central square is shown (the one with 0's and 1's on the edges).   In both figure 1 and 2 the whole chart (containing many of these patterns) spans the medium width (film or CCD).

Figure 3.  Magnification of the central part of the CP5000 resolution image.  Item col 0, row 3 (i.e. 0/3) is resolved (IMHO, in the same sense as the film doing 1/3).  However, note the LACK OF GRAIN NOISE in the digital picture (compare with figure 1).  Clean grain-free digital images are an advantage that helps make prints than would normally be associated with film at a given resolution.

Figure 4.  The original Fuji-Provia test pattern slide (raw film, no glass mount) after being projected onto a matte screen by an Ektagraphic Projecter with a high quality projection lens.  Now, column 1, row 1 or 2 (upper right blocks) are barely resolved (compare with figure 1).  The optics of the projection system causes a small loss in detail.  Note:  This was made by photographing the screen with the tele-macro mode of the digital camera.

Figure 5.  The same test pattern on-screen when projected by a Sanyo 1024 x 768 pixel LCD digital projector.  The pixels are clearly evident.  The central test pattern seen in figures 1 - 3 is gone.  Resolution is OK at col/row = -1/3 (i.e. the contrast is still pretty good), but taps out by -1/5.  This corresponds to about 650 and 800 lines (across the whole image) respectively.

Figure 6.  A 512 line pair test pattern as projected out of the Sanyo.  The contrast is essentially 100%.  Compare with the 0/1 block (lower right) of the film image (fig. 1)

Table 1:  Measured Resolution of Film and Digital

Home