US 12,361,723 B2
Object recognition device
Masanari Takaki, Kariya (JP)
Assigned to DENSO CORPORATION, Aichi-pref. (JP)
Filed by DENSO CORPORATION, Kariya (JP)
Filed on Sep. 2, 2022, as Appl. No. 17/929,477.
Application 17/929,477 is a continuation of application No. PCT/JP2021/008176, filed on Mar. 3, 2021.
Claims priority of application No. 2020-039161 (JP), filed on Mar. 6, 2020.
Prior Publication US 2022/0415058 A1, Dec. 29, 2022
Int. Cl. G06V 20/58 (2022.01); G01S 17/931 (2020.01); G06V 10/20 (2022.01); G06V 10/25 (2022.01)
CPC G06V 20/58 (2022.01) [G06V 10/25 (2022.01); G06V 10/255 (2022.01)] 10 Claims
OG exemplary drawing
 
1. An object recognition device configured to be mounted to a vehicle, the object recognition device comprising:
a processor;
a non-transitory computer-readable storage medium; and
a set of computer-executable instructions stored on the non-transitory computer-readable storage medium that cause the processor to:
form a cluster point group by executing clustering for a plurality of reflecting points specified by radiating transmission waves in a plurality of irradiation directions and receiving reflected waves of the transmission waves; and
recognize the cluster point group as an object,
wherein the set of computer-executable instructions further cause the processor to:
extract a plurality of reflecting points of interest from a plurality of outer edge reflecting points located at an outer edge on the vehicle side viewed from the top in the cluster point group at an extraction ratio based on a distance between the vehicle and the cluster point group;
determine whether an outer shape of an upper surface that is an outer shape of the recognized object viewed from a top has a concave section when viewed from the vehicle side, using the plurality of reflecting points of interest, and
in response to determining that the outer shape of the upper surface has the concave section, divide the cluster point group into a plurality of quasi-cluster point groups based on a position of the concave section and recognize each of the plurality of quasi-cluster point groups as an object, and
the extraction ratio is smaller as the distance between the vehicle and the cluster point group is shorter.