Get access

Exsanguinated Blood Volume Estimation Using Fractal Analysis of Digital Images


  • Presented at the 62nd Annual Meeting of the American Academy of Forensic Sciences, February 22–27, 2010, in Seattle, WA.

Additional information and reprint requests:
Scott I. Fairgrieve, Ph.D.
Department of Forensic Science
Laurentian University
935 Ramsey Lake Road
ON P3E 2C6


Abstract:  The estimation of bloodstain volume using fractal analysis of digital images of passive blood stains is presented. Binary digital photos of bloodstains of known volumes (ranging from 1 to 7 mL), dispersed in a defined area, were subjected to image analysis using FracLac V. 2.0 for ImageJ. The box-counting method was used to generate a fractal dimension for each trial. A positive correlation between the generated fractal number and the volume of blood was found (R2 = 0.99). Regression equations were produced to estimate the volume of blood in blind trials. An error rate ranging from 78% for 1 mL to 7% for 6 mL demonstrated that as the volume increases so does the accuracy of the volume estimation. This method used in the preliminary study proved that bloodstain patterns may be deconstructed into mathematical parameters, thus removing the subjective element inherent in other methods of volume estimation.