Users of desktop scanning technology are conscious about the quality the image detail reproduction. Repeatability and accuracy are based on certain standards. Established scanning procedures for charged couple device (CCD) desktop scan ners are primarily influenced by the Nyquist Criterion which states that an image must be sampled at twice the frequency of the line screen ruling that will be used to print the final image. These "rules of thumb" appear in many techni cal and trade publications that are consulted by the printing industry in order to insure the quality of printed products. Quality is not the only factor influenced by scanning ratios. File size, processing time, and cost are also affected. The purpose of this study was to discover if it is possible to reduce sampling ratios below 2:1 while still preserving the desired level of image detail. The paired comparisons method of testing was employed to determine the threshold where a population could no longer discern the difference between images scanned at a 2:1 scanning ratio and images scanned at lower frequencies. The tests involved a high and low detail image scanned at ratios between 1:1 and 2:1. These were out put by an image-setter to films at 85 lpi and 133 lpi. The results of the statistical analysis show that it is possible to reduce the scanning frequency below the limit established by the Nyquist Criterion while still maintaining a consistent level of perceived detail in an image. Also, it was shown that as the level of detail in an image decreases and as the line screen frequency used increases the scanning ratio can be allowed to decrease. Therefore, it is possible to reduce the size of a file before it is created and consequentially to reduce the time required to per form image processing functions.
Library of Congress Subject Headings
Scanning systems; Image processing--Digital techniques; Electronic publishing--Equipment and supplies; Desktop publishing
Department, Program, or Center
School of Print Media (CIAS)
Beaulieu, Mike, "Scanning ratios for desktop imaging" (1993). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus