Visual Object Detection and Tracking System Design based on MobileNet-SSD
Chen Gao, You Zhai, Xiwei Guo · 2021 7th International Conference on Computer and Communications (ICCC) · 2021
In order to realize real-time object detection and collaborative control of multi pan-tilts for unmanned platform, a scheme of visual object detection and tracking system based on MobileNet-SSD algorithm is proposed. This scheme can realize real-time visual object detection of people and vehicles in mobile platform without GPUs. First, objects are detected based on MobileNet-SSD algorithm. Second, the result of object detection is used for calculating attitude variation matrix for controlling multi pan-tilts. Finally, the multi pan-tilts can equip with different equipment, namely, three two-axis pan-tilts tracking objects to complete a various of tasks together. The experimental results show that the real-time performance of object detection is feasible, and the two-axis pan-tilts can track objects effectively.