Hand-wrist bone assesment for automatic age identification

Unknown people who have died in suspicious circumstances or in mass disaster often have no evidence to indicate chronological age. Using bone has been being mostly used by an experienced doctor to identify age since it provides the most accurate result compared to other diagnoses. However, due to manual identification, the doctor requires tremendous amount of time and the result is highly subjective to the doctor decisions and experience. In this paper, an automatic identification of person age is performed only from a hand-wrist bone structure. It provides a simpler way and reliable as compared to use full-hand bone structure. An active contour with balloons model is studied to extract hand-wrist bone parts namely epiphysis and metaphysis. From these parts, the width ratio is employed to identify person age. Experiments show encouraging results for the proposed method to automatically identify person age. We believe this can benefit for the future of forensic identification.