| Augmented reality(AR)technology is a technology that superimposes virtual information with the natural world to improve the interactive experience of the real environment and has been widely used in many fields.Dynamic projection mapping technology is a display method of augmented reality technology.It is a technology that projects virtual image information on non-stationary objects in the real world through a projector.This technology focuses on projecting high followability on high-speed,high-dynamic moving objects.High-realistic projected images,real-time tracking of the moving direction of high-speed,high-dynamic moving objects,and synchronous projected images are the key to the realization of this technology.In addition,real-time perception of the pose information of dynamic projection mapping objects can better promote technology development in high virtual-real fusion and interaction.Therefore,a high-speed active visual tracking platform was designed and built for dynamic projection objects.Based on this platform,the thesis studied the pose perception method of dynamic projection objects,and the real-time tracking and pose perception of high dynamic objects were realized.The projected images were accurately projected onto the tracked object.First,the high-speed active vision tracking scheme for dynamic projection mapping objects was studied in this thesis to solve the problem of poor followability of the projected image relative to the moving object when using a high-speed and high-dynamic moving object as the projection mapping object.The thesis included designing a high-speed active vision tracking platform coaxial with the camera’s shooting optical path and the projector’s projection optical path.This platform tracked the movement of the object by adjusting the direction of the optical axis and realized the function of real-time projection mapping of the projected image on the moving object.The feature points of infrared LEDs were designed on the dynamic projection mapping object to avoid interference brought by the projection image to the feature point extraction of the projected object,which improved the accuracy and stability of recognition.In addition,the thesis designed a visual servo control algorithm to control the direction of the optical axis of the high-speed active vision tracking platform in real-time and to achieve tracking and projection of high-speed,highly dynamic moving objects.Secondly,this thesis designed a dynamic and static coordinate system based on a high-speed active vision tracking platform to address the problem of real-time pose perception of dynamic projected objects.This thesis utilized the EPn P(Effective Perspective-n-Point)algorithm to solve the pose of the dynamic projection mapping object in the dynamic coordinate system.It investigated the conversion relationship between the dynamic coordinate system and the static coordinate system to obtain the pose of the dynamic projection mapping object in the static world coordinate system.Finally,the tracking pixel errors,attitude angle errors,and position coordinate errors were tested based on the high-speed active vision tracking platform to verify the reliability of the high-speed active vision tracking method and the accuracy of the pose perception method for dynamic projection mapping objects.The response time of the program was tested based on the platform and the maximum response time was at most 3.5 ms,which proved that the method in this thesis has good real-time performance. |
