Picture-based vehicle loss assessment
Methods, systems, and computer-readable storage media for generation of a vehicle repair plan. Implementations include actions of receiving vehicle damage data including an image of a damaged vehicle. The vehicle damage data is processed to determine a component region. The component region is processed to determine a damaged area and a damage type of a portion of the damaged vehicle. A maintenance plan is generated for the damaged vehicle based on the damaged area and the damage type. The maintenance plan is initiated for the damaged vehicle.
1. A computer-implemented method comprising:
receiving, at a cloud server and from a mobile device, an image of a damaged vehicle;
analyzing, by the cloud server, the image of the damaged vehicle using a component identification model to identify a region of the image that includes a vehicle component, wherein the vehicle component comprises multiple parts;
in parallel with analyzing the image of the damaged vehicle using the component identification model, analyzing, by the cloud server, the image of the damaged vehicle using a damage identification model to identify a region of the image that includes a damaged part, and a damage type indicated in the image;
determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model;
in response to determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model, generating, by the cloud server, a maintenance plan for the damaged vehicle based on the identified damage type; and
transmitting, by the cloud server, the maintenance plan to the mobile device.
2. The method of claim 1 , wherein the image of the damaged vehicle is analyzed by the component identification model only after determining that the image satisfies predetermined quality criteria.
3. The method of claim 1 , wherein the component identification model further identifies a position of the identified region in the image.
4. The method of claim 1 , wherein the component identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
5. The method of claim 1 , wherein the damage identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
6. The method of claim 1 , wherein the maintenance plan is generated for a the damaged part that has a most severe amount of damage.
7. The method of claim 1 , wherein the maintenance plan includes estimated modification costs for maintenance of the vehicle component.
8. A non-transitory computer readable storage medium storing instructions executable by a data processing apparatus and upon such execution cause the data processing apparatus to perform operations comprising:
receiving, at a cloud server and from a mobile device, an image;
analyzing, by the cloud server, the image of a damaged vehicle using a component identification model to identify a region of the image that includes a vehicle component, wherein the vehicle component comprises multiple parts;
in parallel with analyzing the image of the damaged vehicle using the component identification model, analyzing, by the cloud server, the image of the damaged vehicle using a damage identification model to identify a region of the image that includes a damaged part, and a damage type indicated in the image;
determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model;
in response to determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model, generating, by the cloud server, a maintenance plan for the damaged vehicle based on the identified damage type; and
transmitting, by the cloud server, the maintenance plan to the mobile device.
9. The medium of claim 8 , wherein the image of the damaged vehicle is analyzed by the component identification model only after determining that the image satisfies predetermined quality criteria.
10. The medium of claim 8 , wherein the component identification model further identifies a position of the identified region in the image.
11. The medium of claim 8 , wherein the component identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
12. The medium of claim 8 , wherein the damage identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
13. The medium of claim 8 , wherein the maintenance plan is generated for a the damaged part that has a most severe amount of damage.
14. The medium of claim 8 , wherein the maintenance plan includes estimated modification costs for maintenance of the vehicle component.
15. A system comprising:
one or more processing devices; and
one or more storage devices storing instructions that are executable by the one or more processing devices to perform operations comprising:
receiving, at a cloud server and from a mobile device, an image;
analyzing, by the cloud server, the image of a damaged vehicle using a component identification model to identify a region of the image that includes a vehicle component, wherein the vehicle component comprises multiple parts;
in parallel with analyzing the image of the damaged vehicle using the component identification model, analyzing, by the cloud server, the image of the damaged vehicle using a damage identification model to identify a region of the image that includes a damaged part, and a damage type indicated in the image;
determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model;
in response to determining that the region of the image that was identified by the damage identification model matches the region of the image that was identified by the component identification model, generating, by the cloud server, a maintenance plan for the damaged vehicle based on the identified damage type; and
transmitting, by the cloud server, the maintenance plan to the mobile device.
16. The system of claim 15 , wherein the image of the damaged vehicle is analyzed by the component identification model only after determining that the image satisfies predetermined quality criteria.
17. The system of claim 15 , wherein the component identification model further identifies a position of the identified region in the image.
18. The system of claim 15 , wherein the component identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
19. The system of claim 15 , wherein the damage identification model is generated based on a convolutional neural network (CNN) and a region proposal network (RPN).
20. The system of claim 15 , wherein the maintenance plan is generated for a the damaged part that has a most severe amount of damage.