Abstract
The analysis of single images is inherently limited to resolving less than two orders of magnitude of fragment linear size. However, field implementation of image analysis for the characterization of blast fragments requires resolving three orders of magnitude. The combined processing of “zoomed” images at different magnifications can overcome this limitation. The simplest algorithm to combine images works at the level of particle counts at each scale. Each histogram can be band-pass filtered to avoid the effects of disintegration and fusing. It is shown that the merging of images at different scales reduces the need for calibration corrections embedded in the unfolding function.