A Fast Pose Estimation Method Based on New QR Code for Location of Indoor Mobile Robot

2019 
Based on traditional QR Code, we propose a new improved QR Code. On the foundation of this QR Code, we propose a pose estimation method based on two stages. In the first stage, a decision whether or not QR Code exists is made efficiently based on the proposed clustering method. In the second stage, based on the estimated image range containing QR Code, we first reckon all the positions of pixel blocks, then, a strategy in which every reckoned pixel block is reprocessed is employed. In this situation, we can discard the invalid image based on the first stage efficiently. Meanwhile, the strategy used in the second stage guarantees the efficiency and accuracy for the accurate extraction of QR Code. It is demonstrated that the proposed method can extract QR Code successfully in complex background. Based on the extracted QR Code, the pose can be estimated easily. The achievements of this article can be used in the location of indoor mobile robot directly, further, applications such as grasping and so on are also the potential users.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    10
    References
    0
    Citations
    NaN
    KQI
    []