V +: Video Analytics Solutions


The focus of this project is the development of video analysis solutions for the internet of things (IoT) platform from Maxtrack, the largest tracking and telemetry company in Latin America. Several problems in the surveillance and biometric domains are addressed in this project, including vehicle recognition and person detection, tracking, and identification.

One of the solutions that Sense researchers will develop is a method for detection of improper cellphone usage by drivers and also aggressive traffic behavior, technologies that can improve safety in cargo transportation and traffic in general. Also focusing on traffic surveillance, the project includes methods for vehicle license plate recognition and tracking through the use of surveillance cameras. Another solution that will be developed is the detection of spoofing: the attempt to circumvent face recognition systems by presenting static images (photographs or prints) of an authorized subject.

The project was divided into two stages. In the first, the researchers created methods for monitoring closed environments (galleries, shops, etc.). In the second stage, in progress, the Sense team created methods to monitor truck cabins and analyze highways through cameras installed in the vehicles. For this purpose, raspberries, which are small computers with minimal computational power, were used.