Not all scanners are created equal. Comments and comparison test results

Below you will find the same objects scanned with different scanners. As you can see, not only dpi (resolution) is important. The quality of optical, electronic and mechanical components  has a great influence on what you can see in an image (and hence, the precision of the measurements you make from it). Scanners are made for different applications, the graphics industry and home use being the major ones. Requirements for scientific usage are different. Eye-pleasing images are fine, but it's better to accurately reproduce reality. Therefore, before selling scanners for scientific use, we test them carefully to make sure they have minimal qualities. You cannot rely on theoretical specifications alone.

A scan at 1200 dpi on a 600x1200 dpi scanner. A scan at 800 dpi on a 800x1600 dpi scanner. Actual width of target section is 5 mm

From the images above, you can see that more detail  is visible in a good quality scan at 800 dpi than in an average quality one at 1200 dpi. The 1200 dpi image will usually take more time to scan, consume more disk space and require more time to analyse but might not produce more precise measurements. It is also revealing to compare scanners at the limit of their range to evaluate their overall qualities.  The following scans of roots, tree-rings and targets illustrate this eloquently.

Roots scanned at 1600 dpi on on a HP6100C . Same sample scanned at 1600 dpi on our STD1600+ . Notice the root detail in the center of circle 'A' that is barely discernible in the scan to the left. True diameter of circle 'B' is 1.4 mm

Tree-rings scanned at 1600 dpi on a HP6100C. Same sample scanned at 1600 dpi on our STD1600+. True diameter of circle is 1.4 mm

Calibration target scanned at 1600 dpi on on a HP6100C. Same target at 1600 dpi on STD1600+. Same target at 1600 dpi on LC3200.

Note: Printouts of this page don't show as much detail as when viewed on screen.

Last Revision: January 23, 2008