Method and apparatus for improving vehicle loss assessment image identification result, and server
Embodiments of this specification disclose a method and an apparatus for automotive part recognition based on a vehicle damage assessment image. The method includes: determining a damaged automotive part of a vehicle by recognizing a damage assessment image of the vehicle based on an image recognition algorithm; obtaining an automotive part list of the vehicle, wherein the automotive part list comprises a plurality of automotive parts corresponding to a plurality of automotive part identification serial numbers; matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part; and outputting an automotive part identification serial number of the determined automotive part.
1. A computer-implemented method for automotive part recognition based on a vehicle damage assessment image, comprising:
determining a damaged automotive part of a vehicle by recognizing a damage assessment image of the vehicle based on an image recognition algorithm;
obtaining an automotive part list of the vehicle, wherein the automotive part list comprises a plurality of automotive parts corresponding to a plurality of automotive part identification serial numbers;
matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part;
outputting an automotive part identification serial number of the determined automotive part; and
in response to that the automotive part list does not comprise an automotive part matching the damaged automotive part, querying the automotive part list for an automotive part having an automotive part inclusion relationship with the damaged automotive part, and determining the automotive part to be corresponding to the damaged automotive part.
2. The method according to claim 1 , wherein the obtaining an automotive part list of the vehicle comprises:
querying a database to obtain the automotive part list of the vehicle based on a vehicle identification code of the vehicle.
3. The method according to claim 1 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
querying whether the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part.
4. The method according to claim 3 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part further comprises:
in response to that the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part, determining the automotive part to be corresponding to the damaged automotive part.
5. The method according to claim 1 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
in response to that the automotive part list comprises at least two sub-level automotive part categories of the damaged automotive part, re-recognizing the damage assessment image of the vehicle until a unique automotive part corresponding to the damaged automotive part is determined in one of the at least two sub-level automotive part categories, or an upper limit of times for recognizing the damage assessment image is reached.
6. The method according to claim 1 , wherein the automotive part having the automotive part inclusion relationship with the damaged automotive part includes the damaged automotive part or is included in the damaged automotive part.
7. An apparatus for automotive part recognition based on a vehicle damage assessment image, comprising one or more processors and one or more non-transitory computer-readable memories coupled to the one or more processors and configured with instructions executable by the one or more processors to cause the apparatus to perform operations comprising:
determining a damaged automotive part of a vehicle by recognizing a damage assessment image of the vehicle based on an image recognition algorithm;
obtaining an automotive part list of the vehicle, wherein the automotive part list comprises a plurality of automotive parts corresponding to a plurality of automotive part identification serial numbers;
matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part;
outputting an automotive part identification serial number of the determined automotive part; and
in response to that the automotive part list does not comprise an automotive part matching the damaged automotive part, querying the automotive part list for an automotive part having an automotive part inclusion relationship with the damaged automotive part, and determining the automotive part to be corresponding to the damaged automotive part.
8. The apparatus according to claim 7 , wherein the obtaining an automotive part list of the vehicle comprises:
querying a database to obtain the automotive part list of the vehicle based on a vehicle identification code of the vehicle.
9. The apparatus according to claim 7 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
querying whether the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part.
10. The apparatus according to claim 9 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part further comprises:
in response to that the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part, determining the automotive part to be corresponding to the damaged automotive part.
11. The apparatus according to claim 7 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
in response to that the automotive part list comprises at least two sub-level automotive part categories of the damaged automotive part, re-recognizing the damage assessment image of the vehicle until a unique automotive part corresponding to the damaged automotive part is determined in one of the at least two sub-level automotive part categories, or an upper limit of times for recognizing the damage assessment image is reached.
12. The apparatus according to claim 7 , wherein the automotive part having the automotive part inclusion relationship with the damaged automotive part includes the damaged automotive part or is included in the damaged automotive part.
13. A non-transitory computer-readable storage medium configured with instructions executable by one or more processors to cause the one or more processors to perform operations comprising:
determining a damaged automotive part of a vehicle by recognizing a damage assessment image of the vehicle based on an image recognition algorithm;
obtaining an automotive part list of the vehicle, wherein the automotive part list comprises a plurality of automotive parts corresponding to a plurality of automotive part identification serial numbers;
matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part;
outputting an automotive part identification serial number of the determined automotive part; and
in response to that the automotive part list does not comprise an automotive part matching the damaged automotive part, querying the automotive part list for an automotive part having an automotive part inclusion relationship with the damaged automotive part, and determining the automotive part to be corresponding to the damaged automotive part.
14. The non-transitory computer-readable storage medium according to claim 13 , wherein the obtaining an automotive part list of the vehicle comprises:
querying a database to obtain the automotive part list of the vehicle based on a vehicle identification code of the vehicle.
15. The non-transitory computer-readable storage medium according to claim 13 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
querying whether the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part.
16. The non-transitory computer-readable storage medium according to claim 15 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part further comprises:
in response to that the automotive part list comprises an automotive part matching a feature attribute of the damaged automotive part, determining the automotive part to be corresponding to the damaged automotive part.
17. The non-transitory computer-readable storage medium according to claim 13 , wherein the matching the damaged automotive part with the plurality of automotive parts in the automotive part list to determine an automotive part corresponding to the damaged automotive part comprises:
in response to that the automotive part list comprises at least two sub-level automotive part categories of the damaged automotive part, re-recognizing the damage assessment image of the vehicle until a unique automotive part corresponding to the damaged automotive part is determined in one of the at least two sub-level automotive part categories, or an upper limit of times for recognizing the damage assessment image is reached.