US 11,915,278 B2
System and method for automatically detecting damages in vehicles
Eliron Ekstein, Even Yehuda (IL); Roman Sandler, Haifa (IL); Alexander Kenis, Kiryat Motzkin (IL); and Alexander Serebryiani, Ashkelon (IL)
Assigned to RAVIN AI LTD., Haifa (IL)
Appl. No. 17/437,714
Filed by RAVIN AI LTD., Haifa (IL)
PCT Filed Mar. 9, 2020, PCT No. PCT/IL2020/050274
§ 371(c)(1), (2) Date Sep. 9, 2021,
PCT Pub. No. WO2020/183459, PCT Pub. Date Sep. 17, 2020.
Claims priority of application No. 19162604 (EP), filed on Mar. 13, 2019.
Prior Publication US 2022/0148287 A1, May 12, 2022
Int. Cl. G06K 9/00 (2022.01); G06Q 30/0283 (2023.01); G06V 10/75 (2022.01); G06Q 30/0645 (2023.01); G06Q 40/08 (2012.01); G07C 5/00 (2006.01)
CPC G06Q 30/0283 (2013.01) [G06Q 30/0645 (2013.01); G06Q 40/08 (2013.01); G06V 10/751 (2022.01); G06V 2201/06 (2022.01); G07C 5/006 (2013.01)] 21 Claims
OG exemplary drawing
 
1. An automatic system for detecting a damage to a vehicle, comprising:
a. at least one camera for capturing “handover” and “return” sets of images of a vehicle, wherein each of the images within said sets is captured irrespective of the vehicle's movement or of the vehicle's orientation with respect to the capturing camera;
b. a database for pre-storing images of basic-parts of at least one vehicle, wherein said basic-parts are selected from the vehicle door-handles, wipers, the right and left mirrors, the four wheels, the license plates, lights, and manufacturer's symbols;
c. a first determination unit which is configured to, based on said pre-stored basic-parts images, determine the location of one or more basic-parts within images of said “handover” and “return” sets of captured images;
d. a second determination unit which is configured to, based on said determined locations of the one or more vehicle's basic parts within the captured images, respectively, determine the locations of “other parts” of the vehicle within said “handover” and “return” sets of images, thereby to form “handover” and “return” part-images, respectively;
e. a transformation unit which is configured to separately transform each pair of “handover” and “return” part-images, respectively, to a same plane, wherein said transformation further includes performing of one or more of: displacement, rotation and zoom of a first image relative to a second image in the pair;
f. a comparison unit configured to separately compare pixel-by-pixel each of said transformed pairs of “handover” and “return” part-images, to thereby detect a difference above a predefined threshold, and to alert when such a difference above the threshold is detected.