US 11,718,151 B2
Determination device, vehicle control device, determination method and determination program
Shin Sakurada, Toyota (JP)
Assigned to TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed by TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed on Oct. 2, 2019, as Appl. No. 16/590,597.
Claims priority of application No. 2018-217423 (JP), filed on Nov. 20, 2018.
Prior Publication US 2020/0156437 A1, May 21, 2020
Int. Cl. B60H 1/00 (2006.01); G06V 40/16 (2022.01)
CPC B60H 1/00742 (2013.01) [B60H 1/00971 (2013.01); G06V 40/172 (2022.01); B60H 2001/00733 (2019.05)] 10 Claims
OG exemplary drawing
 
1. A determination device, comprising:
circuitry configured to:
first, extract a high-visibility area from a plurality of areas, the high-visibility area having a visibility state estimated to be higher than another area of the plurality of areas;
second, determine whether a vehicle exists in the high-visibility area;
third, only when the vehicle is determined to exist in the high-visibility area, determine that a predetermined attached matter is on a transparent member of the vehicle when an image recognition rate of first image information about a periphery of the vehicle is equal to or lower than a predetermined threshold; and
send control information to the vehicle, the control information causing the vehicle to take action in response to determining that the predetermined attached matter is on the transparent member of the vehicle, wherein
the first image information is picked up through the transparent member by a first pickup device that is mounted on the vehicle,
the predetermined threshold hold is determined based upon second image information about a periphery of another vehicle that exists in the high-visibility area picked up by a second pickup device that is mounted on the other vehicle,
vehicles that are in a predetermined vehicle state adapted for being an object of a statistical process are selected,
the vehicles that are in the predetermined vehicle state adapted for being the object of the statistical process are vehicles corresponding to image information in which an image recognition rate is within a predetermined normal range, and
the visibility state in the high-visibility area is estimated from the second image information corresponding to each of the selected vehicles.