关键词: STM32, 智能小车, April Tag算法, OpenMV

Abstract:

In order to improve the accuracy and real-time performance of the intelligent vehicle following the target， an automatic following system based on the improved April Tag algorithm is designed. Aiming at the problems of low real-time performance of current target tracking algorithm and poor recognition performance when the target is small， an improved scheme based on April Tag algorithm is proposed. The real-time performance is improved by reducing the image quality when the label is large while the recognition accuracy is not affected. The recognition accuracy is improved by improving the image quality and reducing the detection range when the label is too small. The experimental results show that the improved April Tag algorithm has better accuracy and recognition speed， and improves the real-time performance and accuracy of target tracking effectively， which meets the requirements of target recognition and tracking for intelligent following robot.

Key words: STM32, mart car, April Tag algorithm, OpenMV