The study concludes that if the comparison of two intra-oral radiographs produces a higher correlation coefficient than the minimum threshold for autocorrelation of the absolute distances, relative distances, and triangle perimeters, then identification is positive. In border-line cases, positive identification could result in greater accuracy if the interdependencies among the variables are observed altogether. In addition, the study offers the possibility of extending the comparisons with subjects having crown restorations that do not include cement-enamel junctions. Image acquisition procedures are simple and fast, and digital technology ensures a standardized method of acquisition and comparison. Another advantage of this technique is that the operator only needs to locate two points (the cement-enamel junctions), because the others are construction points, thereby providing a lower probability of errors. Intra-oral radiographs of the inferior right first molar, without restorations, were acquired at 2 different times in 70 subjects, using a standardized technique, followed by a morphometric analysis. The program automatically supplied values of the absolute distances; relative distances; shape factors; moments; perimeter values; and the areas of the triangles, which were obtained by joining key reference points. The method was tested on dental records from 30 patients in order to show the specificity and sensitivity of the system. 1 table, 6 figures, and 10 references