Purpose: Neural networks might be an appropriate solution for the categorization of child sexual abuse materials (CSAM) in forensic cases. The aim of this study was to present a neural network model that may be able to categorize objects and behaviors, which are visible in CSAM, using pictures visually similar to CSAM (AB/DL), involving persons who have paraphilic preferences for watching adult women or men dressed like children or involved in activities typical for children, such as playing.
Methods: The dataset consisted of 2251 photos divided into five classes. 1914 photos were randomly used for the training of the neural network, while 337 photos were used for its later validation. The Fast.ai and PyTorch libraries were used for the training of the neural network using the ResNet152 model. We used five classes, two of which were imported from the sexACT dataset, and three of which that were collected for this study.
Results: The model was able to classify selected classes with a relatively high accuracy (95%); on the other hand, further improvement of the network is needed, considering the fact that the final validation loss was moderate (0.17).
Conclusions: The model presented might be effective in the classification of several objects and behaviors presented in a number of pornography categories ("paraphilic infantilism", "sexual activity", "nude women", "dressed women", "sexual activity - spanking"). As the results are promising, further research on real CSAM is justified.
Keywords: CSAM; neural networks; pornography.
Copyright © 2022 Institute of Psychiatry and Neurology.