Inconsistent damage determination
The present invention relates to the determination of damage to portions of a vehicle. More particularly, the present invention relates to determining whether determined damage to a vehicle is consistent with information provided as to the cause of the damage to the vehicle. Aspects and/or embodiments seek to provide a computer-implemented method for determining whether damage to a vehicle, which is determined using images of the damage to the vehicle, is consistent with information documenting the cause of the damage to the vehicle, for example insurance claim data or repair shop proposed repair data.
1. A method, comprising:
receiving a plurality of images of a vehicle, the vehicle comprising a plurality of parts;
determining, using a first set of one or more trained models, a damage state for at least one part of the plurality of parts based on the plurality of images;
predicting, using a second set of one or more trained models, one or more instances of damage to the vehicle based on accident data, wherein the accident data comprises data about an accident experienced by the vehicle;
combining the damage state for each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models;
generating a damage vector of the combined damage state of each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models; and
determining a measure of consistency between the damage state and the one or more instances of predicted damage to the vehicle based on the damage vector.
2. The method of claim 1 , wherein the accident data further comprises at least one or more of a type of damage or a severity of damage.
3. The method of claim 1 , wherein the accident data further comprises at least one or more of a vehicle motion state or a point of impact.
4. The method of claim 1 , wherein the accident data comprises data processed by one or more natural language models.
5. The method of claim 1 , wherein the second set of one or more trained models comprises at least a natural language processing model.
6. The method of claim 1 , wherein the second set of one or more trained models comprises at least one of a neural network, a recurrent neural network or a trained classifier.
7. A non-transitory computer readable storage medium comprising a computer program product comprising computer code configured to:
receive a plurality of images of a vehicle, the vehicle comprising a plurality of parts;
determine, using a first set of one or more trained models, a damage state for at least one part of the plurality of parts based on the plurality of images;
predict, using a second set of one or more trained models, one or more instances of damage to the vehicle based on accident data, wherein the accident data comprises data about an accident experienced by the vehicle;
combine the damage state for each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models;
generate a damage vector of the combined damage state of each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models; and
determine a measure of consistency between the damage state and the one or more instances of predicted damage to the vehicle based on the damage vector.
8. The non-transitory computer readable storage medium comprising a computer program product of claim 7 , wherein the accident data further comprises at least one or more of a type of damage or a severity of damage.
9. The non-transitory computer readable storage medium comprising a computer program product of claim 7 , wherein the accident data further comprises at least one or more of a vehicle motion state or a point of impact.
10. The non-transitory computer readable storage medium comprising a computer program product of claim 7 , wherein the accident data comprises data processed by one or more natural language models.
11. The non-transitory computer readable storage medium comprising a computer program product of claim 7 , wherein the second set of one or more trained models comprises at least a natural language processing model.
12. The non-transitory computer readable storage medium comprising a computer program product of claim 7 , wherein the second set of one or more trained models comprises at least one of a neural network, a recurrent neural network or a trained classifier.
13. A computing device, comprising:
a memory; and
one or more processors configured to:
receive a plurality of images of a vehicle, the vehicle comprising a plurality of parts;
determine, using a first set of one or more trained models, a damage state for at least one part of the plurality of parts based on the plurality of images;
predict, using a second set of one or more trained models, one or more instances of damage to the vehicle based on accident data, wherein the accident data comprises data about an accident experienced by the vehicle;
combine the damage state for each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models;
generate a damage vector of the combined damage state of each of the plurality of parts of the first set of one or more trained models and one or more instances of damage to the vehicle based on accident data of the second set of one or more trained models; and
determine a measure of consistency between the damage state and the one or more instances of predicted damage to the vehicle based on the damage vector.
14. The computing device of claim 13 , wherein the accident data further comprises at least one or more of a type of damage or a severity of damage.
15. The computing device of claim 13 , wherein the accident data further comprises at least one or more of a vehicle motion state or a point of impact.
16. The computing device of claim 13 , wherein the accident data comprises data processed by one or more natural language models.
17. The computing device of claim 13 , wherein the second set of one or more trained models comprises at least a natural language processing model.
18. The computing device of claim 13 , wherein the second set of one or more trained models comprises at least one of a neural network, a recurrent neural network or a trained classifier.
19. The damage vector of claim 1 , wherein the damage vector comprises the damage state for at least one part of the plurality of parts based on the plurality of images based on the first set of one or more trained models.
20. The damage vector of claim 1 , wherein the damage vector comprises the one or more instances of damage to the vehicle based on accident data, based on the second set of one or more trained models.