|
|
|
This article [Opt. Eng. 57(4), 043110 (2018)] was originally published on 25 April 2018. The authors wish to correct a reference that was accidentally omitted. A new Ref. 31 has been added to the last paragraph of Sec. 2.2, and the text has been updated as: To accelerate the computational efficiency, FAST13 detector combining with binary descriptor is still a better choice. SPHORB30 introduces a geodesic grid used in climate modeling to detect FAST corner and construct robust BRIEF5 descriptor on the hexagonal sphere grid. Inspired by the great performance and efficiency of ORB6 in perspective image, we adjust BRIEF used in ORB to adapt the distortion of omnidirectional image. The method most similar to ours is mdBRIEF,31 a distorted and masked version of the BRIEF descriptor. In this method, the binary tests are distorted, thus the descriptor is adapted to different image regions. Besides of adjusting BRIEF descriptor, we further propose new methods to distinguish true and false matches. The new reference has been inserted in the reference list, and the last three references have been renumbered: ReferencesS. Urban, M. Weinmann and S. Hinz,
“mdBRIEF-a fast online-adaptable, distorted binary descriptor for real-time applications using calibrated wide-angle or fisheye cameras,”
Computer Vision and Image Understanding, 162 71
–86
(2017). https://doi.org/10.1016/j.cviu.2017.08.011 Google Scholar
D. Scaramuzza, A. Martinelli and R. Siegwart,
“A toolbox for easily calibrating omnidirectional cameras,”
in IEEE/RSJ Int. Conf. on Intelligent Robots and Systems,
(2006). Google Scholar
J. W. Bian et al.,
“GMS: Grid-based motion statistics for fast, ultrarobust feature correspondence,”
in IEEE Conf. on Computer Vision and Pattern Recognition (CVPR),
(2017). Google Scholar
|
