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 computer-implemented method, comprising:
receiving a plurality of images of the vehicle;
determining, per part, one or more classifications for each of the plurality of images using one or more trained models, wherein each classification comprises at least one indication of damage;
receiving accident data, wherein the accident data comprises data about the cause of the damage to the vehicle;
processing the accident data to predict damage to one or more parts of the vehicle, wherein processing of the accident data comprises the use of one or more language models; and
determining a measure of consistency between the one or more indications of damage based on a correlation between the determined at least one indication of damage and the predicted damage.
2. The computer-implemented method of claim 1 , wherein determining whether the at least one indication of damage to at least one part has a measure of consistency above a predetermined threshold that indicates the at least one indication of damage was caused in a single accident.
3. The computer-implemented method of claim 2 , wherein determination of the single accident comprises any one or any combination of: location of the one or more indications of damage on the vehicle; type of damage; severity of damage; one or more additional structured queries comprising point and/or of impact; vehicle motion state; manually generated text data.
4. The computer-implemented method of claim 1 , further comprising:
outputting one or more indications of damage or relevant images corresponding to one or more vehicle parts determined to have a measure of consistency above a predetermined threshold.
5. The computer-implemented method of claim 1 , wherein processing the accident data comprises the use of any or any combination of: Natural Language Processing; a neural network; a convolutional neural network; a recurrent neural network; and/one or more trained classifiers.
6. The computer-implemented method of claim 1 , further comprising determining a plurality of parts of the vehicle that are represented in the plurality of images of the vehicle comprising the use of a plurality of classifiers.
7. The computer-implemented method of claim 6 , wherein each one of the plurality of classifiers is operable to detect each of the parts of the vehicle.
8. The computer-implemented method of claim 1 , further comprising determining one or more sets of relevant images of each of the plurality of parts from the plurality of images of the vehicle.
9. The computer-implemented method of claim 8 , further comprising determining a damage state of each of the parts of the vehicle using the one or more relevant images of each of the plurality of parts of the vehicle, the one or more damage states being determined as one or more quantitative values.
10. The computer-implemented method of claim 1 , wherein determining the one or more classifications or determining the one or more damage states comprises using a multi-task learned approach.
11. The computer-implemented method of claim 1 , wherein determining the one or more classifications or determining the one or more damage states comprises using a graph neural network.
12. The computer-implemented method of 1 , wherein determining the one or more classifications for each of the plurality of images further comprises determining one or more feature representations.
13. The computer-implemented method of claim 1 , further comprising:
outputting the measure of consistency of each of the parts of the vehicle wherein the indication of damage to at least two of the parts is determined to be correlated.