Video Stabilization with a Dual System Based on an IMU Sensor for the Mobile Robot

Abstract

Most of the applications in mobile robots use the image data from the monitoring system to compile algorithms. The undesired movement in the video is an obstacle to compute commands, and it requires to reduce this motion for stabilizing the videos. Thus, we presented a dual system to remove the undesired movement from global and local motion by the mechanical system and digital system, respectively. In the mechanical system, we used a PID controller to control the rotation in the roll and the pitch axes in the servo motors to stabilize video in the line of sight, which held the camera’ support base by feedbacking the sensed data from an IMU sensor. Additionally, we applied the KLT feature tracking by integrating an IMU data to estimate the motion for removing the local motion. The results of the dual system were demonstrated as this algorithm reduced the undesired movement when compared with a single combined mechanical and digital system.