Vehicle damage detection system and method
A vehicle damage detection system operating on a computer system for evaluating specific damage to a vehicle. The vehicle damage detection system capturing a 3D point cloud of a damaged vehicle with a 3D scanner and identifies one or more sites of vehicle damage based on comparison of the captured 3D point cloud of the damaged vehicle to a baseline 3D file by identifying points within the 3D point cloud of the damaged vehicle that deviate from an original equipment manufacturer (OEM) standard vehicle of the same type. The damage detection system can be used to determine and identify replacement parts and components and generate an estimate and parts list based on previously damaged vehicles having a similar damage pattern, allowing an improved method for vehicle damage assessment and repair.
1. A computer-implemented method for assessing damage to a vehicle, the method comprising:
capturing a 3D point cloud of a damaged vehicle and its damaged components with a 3D scanner;
comparing the 3D point cloud of the damaged vehicle to a baseline 3D file of an original equipment manufacturer (OEM) standard vehicle of the same type;
identifying a site of vehicle damage on the damaged vehicle based on the comparison of the captured 3D point cloud of the damaged vehicle to the baseline 3D file by identifying a set of deviating points within the 3D point cloud of the damaged vehicle that deviate from the baseline 3D file of the OEM standard vehicle;
quantifying a deviation of the identified set of deviating points on a x,y,z, plot based on location of points in the 3D point cloud of the damaged vehicle at the site of vehicle damage compared to the OEM standard;
displaying, on a user interface, an image overlay of the captured 3D point cloud of the damaged vehicle and the baseline 3D OEM standard vehicle and identifying points within the scanned point cloud of the damage vehicle that are beyond a predetermined threshold of deviation from the OEM standard vehicle; and
selecting, using a graphical user interface, specific points on the 3D point cloud of the damaged vehicle and the baseline 3D file to isolate a measured and angular difference of the specific points.
2. The method of claim 1 , wherein the baseline 3D file of the OEM vehicle of the same type is a point cloud file.
3. The method of claim 1 , wherein the 3D scanner uses LASER triangulation 3D scanning technology, structured light 3D scanning technology, modulated light 3D scanning technology, contact-based 3D scanning technology, LASER pulse-based 3D scanning technology, time-of-flight 3D laser scanning technology, or laser-based 3D scanning technology.
4. The method of claim 1 wherein the 3D point cloud of the damaged vehicle is accurate to less than 2 mm.
5. The method of claim 1 , further comprising matching the identified site of vehicle damage to a previously damaged vehicle in a damaged vehicle database and importing at least one of a repair estimate and a repair procedure for the previously damaged vehicle.
6. The method of claim 5 , wherein the repair estimate comprises a listing of one or more parts for replacement, part repair labor time, part replacement labor time, paint cost, coating cost, refinishing labor time, paint or coating dry time, disposal cost, and administrative cost.
7. The method of claim 5 , wherein the repair procedure provides one or more angular realignment, applied force, or applied force profile, for applying force at a tool anchoring location.
8. The method of claim 1 , wherein comparing the captured 3D composite file of the damaged vehicle and the baseline 3D composite file is executed by one or more convoluted neural networks connected to a processing system.
9. The method of claim 1 , wherein comparing the captured vehicle 3D point cloud to a baseline 3D file of an original equipment manufacturer (OEM) standard vehicle of the same type uses a Nearest Neighbor (NN) algorithm or a Nearest Neighbor Distance Ratio (NNDR) algorithm.
10. The method of claim 1 , wherein comparing the 3D point cloud of the damaged vehicle and the baseline 3D file capture at least one of curvature, smoothness, and geometrical edge similarities.
11. The method of claim 1 , wherein comparing the 3D point cloud of the damaged vehicle to a baseline 3D file comprises matching the damaged vehicle and baseline vehicle by one or more of vehicle model, vehicle year, vehicle make, and vehicle features.
12. The method of claim 1 , wherein the 3D point cloud of the damaged vehicle is displayed in a first color and the baseline 3D file of OEM standard vehicle in a second color.
13. The method of claim 1 , wherein displaying the image overlay comprises displaying the set of differentiating points and deviation distance on the x,y,z, plot.
14. The method of claim 1 , further comprising displaying the image overlay as a heatmap, where each color represents a varying degree of deviation of the 3D point cloud of the damaged vehicle to the baseline 3D file of the OEM standard vehicle.
15. A system for isolating vehicle damage on a damaged vehicle comprising:
a 3D scanner capable of capturing a point cloud of a damaged vehicle and its components to provide a 3D point cloud image of the damaged vehicle;
an original equipment manufacturer (OEM) vehicle database comprising a plurality of baseline 3D files of undamaged vehicles identified by vehicle model, vehicle year, and vehicle make;
a processor for storing and processing information relating to the 3D point cloud of the damaged vehicle and baseline 3D file to:
compare the captured vehicle 3D point cloud of the damaged vehicle to a baseline 3D file of an OEM standard vehicle of the same vehicle model, vehicle year, and vehicle make from the OEM vehicle database;
identify, based on the compared 3D point cloud of the damaged vehicle to the baseline 3D file of the OEM standard vehicle, a plurality of differentiating points between the 3D point cloud of the damaged vehicle and the baseline 3D file of the OEM standard vehicle;
quantify a deviation of the identified set of deviating points on a x,y,z, plot based on location of points in the 3D point cloud of the damaged vehicle at the site of vehicle damage compared to the OEM standard; and
overlay the captured 3D point cloud of the damaged vehicle and the baseline 3D file to create an image overlay; and
a graphical user interface for displaying the image overlay of the 3D point cloud of the damaged vehicle and the baseline 3D file and deviating points within the scanned point cloud of the damage vehicle that are beyond a predetermined threshold of deviation from the OEM standard and selecting specific points on the 3D point cloud of the damaged vehicle and the baseline 3D file to isolate a measured and angular difference of the specific points.
16. The system of claim 15 , wherein the processor allows the image overlay on the user interface to be manipulated in 360 degrees and zoomed in and out.
17. The system of claim 15 , further comprising a parts database comprising a parts list mapped to particular areas of damage on the damaged vehicle, and wherein the processor generates a parts list for repairing the damaged vehicle.
18. The system of claim 15 , further comprising a damaged vehicle database, wherein the processor matches the 3D point cloud of the damaged vehicle to a previously damaged vehicle of the same type, the damaged vehicle database comprising at least one of a repair estimate and a repair procedure for the previously damaged vehicle.
19. The system of claim 15 , wherein the 3D scanner is one or more of a LASER triangulation 3D scanner, structured light 3D scanner, time-of-flight 3D laser scanner, modulated light 3D scanner, contact-based 3D scanner, LASER pulse-based 3D scanner, and other laser-based 3D scanner.
20. The system of claim 15 , wherein the predetermined threshold of difference is less than 2 mm.