US 12,220,172 B2
Systems and methods for combining fundus images of an eye
Mariko Hirokawa, Yokohama (JP); and Yasushi Tanabe, Fujisawa (JP)
Assigned to NIKON CORPORATION, Tokyo (JP)
Appl. No. 17/615,046
Filed by NIKON CORPORATION, Tokyo (JP)
PCT Filed May 31, 2019, PCT No. PCT/JP2019/021868
§ 371(c)(1), (2) Date Nov. 29, 2021,
PCT Pub. No. WO2020/240867, PCT Pub. Date Dec. 3, 2020.
Prior Publication US 2022/0230307 A1, Jul. 21, 2022
Int. Cl. A61B 3/12 (2006.01); A61B 3/00 (2006.01); G06T 7/00 (2017.01)
CPC A61B 3/12 (2013.01) [A61B 3/0025 (2013.01); A61B 3/0058 (2013.01); G06T 7/0012 (2013.01); G06T 2207/10101 (2013.01); G06T 2207/20212 (2013.01); G06T 2207/30041 (2013.01); G06T 2207/30101 (2013.01)] 9 Claims
OG exemplary drawing
 
1. An image processing method comprising:
acquiring a first direction fundus image imaged in a state in which a gaze of an examined eye is directed in a first direction, and a second direction fundus image imaged in a state in which the examined eye is directed in a second direction different from the first direction;
generating a transformed image by transforming one fundus image of the first direction fundus image and the second direction fundus image;
generating a combined image by combining the transformed image and the other fundus image of the first direction fundus image and the second direction fundus image; and
outputting the combined image,
wherein generating the transformed image comprises:
respectively extracting at least four feature points on the fundus of the examined eye, from the first direction fundus image and the second direction fundus image; and
generating the transformed image, in which the at least four feature points of the other fundus image are matched to the at least four feature points of the one fundus image, by projectively transforming the one fundus image, and
wherein generating the combined image comprises:
performing a first weighting processing in the transformed image, with respect to a first region where the transformed image and the other fundus image overlap;
performing a second weighting processing in the other fundus image, with respect to a second region where the transformed image and the other fundus image overlap; and
combining the transformed image and the other fundus image based on performing the first weighting processing and performing the second weighting processing.