Proposal of a Methodology to Identify Unidentified Decedent through Artificial Intelligence Techniques

Abstract

The use of artificial intelligence techniques for the identification and recognition of people is increasing. This article proposes a novel methodology based on artificial neural networks which is based on the classification of tattoos for their analysis and subsequent identification. The techniques used were carefully validated and tested to seek the highest precision in the result. The dataset used was compiled with sample tattoos for training. The model was loaded into a web model that is currently being tested based on Django and python. A dataset and python scripts were built which allowed its implementation.