US 12,340,577 B2
Prediction method
Shigeki Shinoda, Tokyo (JP); Daisuke Ikefuji, Tokyo (JP); Shin Tominaga, Tokyo (JP); Yuzo Senda, Tokyo (JP); and Hirofumi Inoue, Tokyo (JP)
Assigned to NEC CORPORATION, Tokyo (JP)
Appl. No. 17/781,841
Filed by NEC Corporation, Tokyo (JP)
PCT Filed Dec. 17, 2019, PCT No. PCT/JP2019/049340
§ 371(c)(1), (2) Date Jun. 2, 2022,
PCT Pub. No. WO2021/124436, PCT Pub. Date Jun. 24, 2021.
Prior Publication US 2023/0005258 A1, Jan. 5, 2023
Int. Cl. G06K 9/00 (2022.01); G06T 7/73 (2017.01); G06V 20/10 (2022.01)
CPC G06V 20/182 (2022.01) [G06T 7/73 (2017.01); G06T 2207/30181 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A prediction method performed by a computer and comprising:
on a basis of a river image of a river associated with capturing position information representing a position at which the river image of the river is captured, detecting river condition information representing a condition of the river at the position; and
on a basis of the capturing position information, the river condition information, and topography information representing topography of the river, when predicting a river condition representing the condition of the river at a given point of the river different from the position at which the river image was captured, as a river condition at the given point of another river that is determined to have topography similar to the topography of the river based on the topography information, weighting each of the river condition information detected from each of a plurality of river images and predicting a river condition on a basis of a plurality of the river condition information weighted.