Juvenile age estimation from facial images.

Age determination from images can be of vital importance, particularly in cases involving suspected child sexual abuse (CSA). It is imperative to determine if an individual depicted in such an image is indeed a child, with a more concise age often sought, as this may affect the severity of offender sentencing. The aims of this study were to establish the accuracy of visual age estimation of the juvenile face in children aged between 0 and 16years and to determine if varying levels of exposure to children affected an individual's ability to assess age from the face. An online questionnaire consisting of 30 juvenile face images was created using SurveyMonkey®. The overall results suggested poor accuracy for visual age estimation of juvenile faces. The age, sex, occupation and number of children of the participants did not affect the ability to estimate age from facial images. Similarly, the sex and age of the juvenile faces did not appear to affect the accuracy of age estimation. When specific age groups are considered, sex may have an influence on age estimation, with female faces being aged more accurately in the younger age groups and male faces more accurate after the age of 11years, however this is based on a small sample. This study suggests that the accuracy of juvenile age estimation from the face alone is poor using simple visual assessment of images. Further research is required to determine exactly how age is assessed from a facial image, if there are indicators, or features in particular that lead to over- or under-estimation of juvenile age.

[1]  Pasquale Poppa,et al.  Can facial proportions taken from images be of use for ageing in cases of suspected child pornography? A pilot study , 2011, International Journal of Legal Medicine.

[2]  G Geserick,et al.  Age estimation of unaccompanied minors. Part I. General considerations. , 2006, Forensic Science International.

[3]  S Brédart,et al.  An ‘Other-Race’ Effect in Age Estimation from Faces , 2001, Perception.

[4]  G Geserick,et al.  Age estimation of unaccompanied minors. Part II. Dental aspects. , 2006, Forensic science international.

[5]  Zhi-Hua Zhou,et al.  Automatic Age Estimation Based on Facial Aging Patterns , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  A. Cigada,et al.  Towards a method for determining age ranges from faces of juveniles on photographs. , 2014, Forensic science international.

[7]  Arlan L. Rosenbloom Age estimation based on pictures and videos presumably showing child or youth pornography , 2014, International Journal of Legal Medicine.

[8]  M. Guerzoni,et al.  Pitfalls at the root of facial assessment on photographs: a quantitative study of accuracy in positioning facial landmarks , 2013, International Journal of Legal Medicine.

[9]  D. Canning,et al.  Difficulties with age estimation of internet images of south‐east Asian girls , 2003 .

[10]  M. J. Collins,et al.  Age estimation: The state of the art in relation to the specific demands of forensic practise , 2000, International Journal of Legal Medicine.

[11]  M. Ratnayake,et al.  The juvenile face as a suitable age indicator in child pornography cases: a pilot study on the reliability of automated and visual estimation approaches , 2014, International Journal of Legal Medicine.

[12]  Marco Grandi,et al.  The difficult issue of age assessment on pedo-pornographic material. , 2009, Forensic science international.