Development of a Multi-Object Detection and Human Tracking System from Cooperative Dual Cameras in an Unmanned Aerial Vehicle
Jay Robert B. del Rosario, Argel Alejandro Bandala, Janela Assumpta L. Angeles, Aldwin Jerome P. Cabebe, Jnssen Jeremy D. Co, Santamaria, S. Don Emmanuel · 2021 IEEE 13th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) · 2021
Object detection and human tracking are vital for pairing computer vision to a quadrotor. One application for this is monitoring traffic by either classifying vehicles or tracking people through the crowd. This was proven true by the developed system, garnering a detection rate of 77% for various vehicles in the Philippines, and showing the ability to track people on a dual camera system albeit as a proof-of-concept only. Though this can be a potentially huge field for digital image processing, there is still a challenge of creating a fast and responsive application through wireless communication. This study presented a quantitative data when approaching a quadcopter equipped with dual camera system streamed through WiFi connection.