US 12,217,458 B1
Camera extrinsic parameter calibration method, image stitching method and apparatus thereof
Haoran Yang, Chengdu (CN)
Assigned to TP-LINK CORPORATION PTE. LTD., Singapore (SG)
Filed by TP-LINK CORPORATION PTE. LTD., Singapore (SG)
Filed on Jan. 11, 2024, as Appl. No. 18/410,683.
Application 18/410,683 is a continuation of application No. 18/241,377, filed on Sep. 1, 2023, abandoned.
Int. Cl. G06T 7/80 (2017.01); H04N 5/262 (2006.01)
CPC G06T 7/80 (2017.01) [H04N 5/2624 (2013.01); G06T 2207/20212 (2013.01); G06T 2207/30244 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for calibrating extrinsic parameters of a camera, comprising:
acquiring a plurality of images captured by the camera based on a plurality of sets of extrinsic parameters to be calibrated, wherein each of the plurality of sets of extrinsic parameters to be calibrated include a horizontal rotation angle and a vertical rotation angle of Euler angles instead of all Euler angles;
obtaining at least one matched image pair based on feature point matching among the plurality of images, wherein each matched image pair comprises matched feature point pairs with a quantity satisfying a threshold condition; and
calibrating the plurality of sets of extrinsic parameters to be calibrated according to a first difference representation between coordinate representations of each matched feature point pair of each matched image pair in a reference coordinate system,
wherein, the camera comprises a single camera and the camera is configured to rotate in at least one of a horizontal direction or a vertical direction when capturing images,
wherein obtaining at least one matched image pair based on feature point matching among the plurality of images comprises:
performing feature point matching on each image pair in the plurality of images, and determining at least one image pair whose number of matched feature point pairs meets the threshold condition as the at least one matched image pair; or
determining pairwise adjacent images in the plurality of images to obtain candidate adjacent image pairs, and performing feature point matching on each adjacent image pair in the candidate adjacent image pairs, and determining at least one candidate adjacent image pair whose number of matched feature point pairs meets the threshold condition as the at least one matched image pair.