Comparisons of a shoemark with a shoesole (and standards) sometimes lead to
associations based on air bubbles (among other manufacturing or acquired c
haracteristics). Today, the assessment of the evidential value of air bubbl
es coincidences relies largely upon the examiner's experience: and/or follo
ws sometimes a verification based on the examination of a small number of a
nalogous pairs collected for the case at hand. Statistical data related to
the occurrence and characteristics of air bubbles on shoesoles in an attemp
t to model the potential variability have been gathered. Seventy-one pairs
of shoes with the same design, brand, model and size were obtained. Right a
nd left soles were photographed. An image-processing algorithm was develope
d to allow the systematic acquisition of data such as: (1) the number of ai
r bubbles on the sole and around given structural elements; (2) the measure
of air bubbles characteristics such as their surface and position. These d
ata allow a discussion of the assessment of the probability of finding on s
hoesoles (same design, brand, model and size) a certain number of air bubbl
es on a surface with the same positions and morphology. (C) 2000 Elsevier S
cience Ireland Ltd. All rights reserved.