US 11,720,969 B2
Detecting vehicle identity and damage status using single video analysis
Hong Bing Zhang, Beijing (CN); Jing Wu, Xian (CN); Fan Li, Xianyang (CN); Dan Zhang, Beijing (CN); Yun Jing Zhao, Beijing (CN); Xue Ping Liu, Beijing (CN); and Xin Fang Hao, Beijing (CN)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Feb. 7, 2020, as Appl. No. 16/784,309.
Prior Publication US 2021/0248681 A1, Aug. 12, 2021
Int. Cl. G06Q 40/08 (2012.01); G06Q 30/02 (2023.01); G06T 7/00 (2017.01); G06Q 10/10 (2023.01); G06V 20/10 (2022.01)
CPC G06Q 40/08 (2013.01) [G06Q 10/10 (2013.01); G06Q 30/0278 (2013.01); G06T 7/0002 (2013.01); G06V 20/10 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/30168 (2013.01); G06T 2207/30252 (2013.01); G06V 2201/08 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
providing, by one or more processors, an instruction for taking a video of a vehicle for a damage evaluation;
receiving, by one or more processors, the video of the vehicle for the damage evaluation;
verifying, by one or more processors, the vehicle in the video being the same vehicle for the damage evaluation, wherein verifying the vehicle includes:
detecting blurriness and extent of the blurriness in a frame of the video,
detecting an existing abrupt frame change of the video by analyzing a change of corresponding pixels between previous frame and current frame in the video with a video optical flow consistency analysis method, and
performing a cross validation, the cross validation including checking appearance time of the vehicle in public video records by checking timestamp of the video and verifying the time of the vehicle in the video;
evaluating, by one or more processors, a damage status of the vehicle based on an analysis of the video; and
outputting, by one or more processors, a damage report based on the damage status of the vehicle.