Method and apparatus for obtaining photogrammetric data to estimate impact severity
View Patent ↗In one embodiment, the present invention includes a computer-implemented method to collect information to determine damage to a vehicle involved in a collision using photogrammetric techniques. When determined, this vehicle damage information, which may be in the form of crush measurement information such as a crush damage profile, can be displayed in a computer-generated view of the subject vehicle with a crush damage profile and used to estimate the impact severity. In some embodiments, based on the photogrammetric information derived, a direction of any shifting of the vehicle's components may be obtained and used along with other information to estimate a principal direction of force (PDOF) for one or more vehicles involved in the collision. Still further embodiments may be used to generate and/or audit repair estimates based at least in part on the photogrammetric information.
1. A computer-implemented method comprising:
receiving a plurality of images of a vehicle involved in an accident and undergoing a computer-implemented impact severity analysis in a computer;
evaluating, in the computer, the plurality of images for pose estimate information and storing a pose estimate for each of the plurality of images;
refining the stored pose estimate for each of the plurality of images via a bundle adjustment;
determining, in the computer, the location of damage points for the vehicle via triangulation using the plurality of images;
generating, in the computer, a crush profile for the vehicle based on a difference between the location of damage points for the vehicle and corresponding points for a benchmark vehicle, the benchmark vehicle non-damaged;
using the crush profile and a crush profile for a second vehicle involved in the accident, in the computer, to optimize an overlap of damage patterns on the first and second vehicles by comparison and alignment of the damage patterns on the first and second vehicles;
determining, in the computer, estimates of a principal direction of force (PDOF) for the vehicle and the second vehicle based on the optimized overlap; and
determining a final PDOF for the vehicle and the second vehicle based on an iterative adjustment of the PDOF estimates if the PDOF estimates are inconsistent, and storing the final PDOFs in the computer for later use.
2. The method of claim 1 , further comprising displaying the crush profile in a graphical user interface including a reference vehicle.
3. The method of claim 1 , further comprising receiving an input to select a reference pose to correspond to at least one of the plurality of images.
4. The method of claim 3 , further comprising selecting control points and crush points for at least one of the plurality of images.
5. The method of claim 1 , further comprising accessing the crush profile and selecting a plurality of stored crush profiles having corresponding crush profiles within a predetermined threshold of the crush profile.
6. The method of claim 5 , further comprising auditing a claim associated with the accident based on a comparison between the crush profile and the plurality of stored crush profiles.
7. The method of claim 1 , further comprising determining, in the computer, if a camera profile received with the plurality of images corresponds to a camera profile present in a profile database accessed by the computer.
8. The method of claim 7 , further comprising calibrating the plurality of images based on the camera profile if present, otherwise requesting additional information from a user of the camera.
9. An article comprising a machine-accessible storage medium including instructions that when executed cause a system to:
receive a plurality of images of a vehicle involved in an accident and undergoing a computer-implemented impact severity analysis in the system;
evaluate the plurality of images for pose estimate information and store a pose estimate for each of the plurality of images;
refine the stored pose estimate for each of the plurality of images via a bundle adjustment;
determine the location of damage points for the vehicle via triangulation using the plurality of images;
generate a crush profile for the vehicle based on a difference between the location of damage points for the vehicle and corresponding points for a reference vehicle, the reference vehicle non-damaged and substantially similar to the vehicle;
use the crush profile and a crush profile for a second vehicle involved in the accident to optimize an overlap of damage patterns on the first and second vehicles by comparison and alignment of the damage patterns on the first and second vehicles;
determine estimates of a principal direction of force (PDOF) for the vehicle and the second vehicle based on the optimized overlap; and
determine a final PDOF for the vehicle and the second vehicle based on an iterative adjustment of the PDOF estimates if the PDOF estimates are inconsistent, and store the final PDOFs in the system for later use.
10. The article of claim 9 , wherein the instructions further enable the system to display the crush profile in a graphical user interface including the reference vehicle.