Procesamiento de imágenes de twitter o urls asociados, clasificados como sospechosos de estar relacionados con trata de personas

Abstract

The use of the internet is an essential part of today's life. However, it has also become a door for extortion, kidnapping, fraud, human trafficking, among other crimes. In this context, this Master's thesis intends to process and classify images published by Twitter users to determine whether or not there is possible human trafficking, for which faces and torsos are detected in said photos. The images were obtained from a list of Twitter messages that were already labeled as suspects of the crime of human trafficking, within the work carried out in the research project PIS-17 10.

The aforementioned seed project works to detect scenarios related to the crime of human trafficking using natural language processing and focuses exclusively on the group of minors to identify people who are being abused and who are promoted by the media. digital. In such a way that in this thesis it is proposed to carry out the processing of images that allow classifying people under 14 years of age through their photographs published on Twitter that serve to attract “clients”. It should be noted that, in most countries, the law defines that any person under the age of 14 who is being used for sexual exploitation is directly considered a victim of human trafficking.