Comparison of deep learning classification models for facial image age estimation in digital forensic investigations

Monika Roopak, Saad Khan, Simon Parkinson, Rachel Armitage

Abstract: Highlights•A new children image datasets containing over 20K images is constructed.•Deep Learning techniques for the detection of IIoC images are investigated.•Pre-trained deep learning models are exploited in conducting the experiments.•Extensive evaluation of the datasets is performed in three different image identification strategies.•Results are compared to find best suitable image identification strategy for the digital forensic investigation.