A carregar...
Publicação
Threat Detection with Computer Vision
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 7.37 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
This document describes the work conducted during an internship experience at the AI Innovation
Department of Everis UK (now NTT Data). It reports what was done, learned, and developed with the
sole objective of having a commercial product solution for the company's clients.
The primary goal was to implement a solution in retail stores, to help assist the security team with
threat detection. To do so, the solution consists in deploying trained deep learning models into
hardware connected to the CCTV security cameras and detecting in that live feed any potential threats.
By the time I started working on this project, was at an advanced stage so I had to study all the work
previously done to understand what was needed and properly integrate the team fully. My
contribution was focused on the model training process, where I had to create and structure a dataset
and train a model capable of detecting the targeted classes quickly and accurately.
Descrição
Internship Report presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Business Analytics
Palavras-chave
computer vision deep learning inference security