Title:
NEURAL NETWORK-BASED DATA VERIFICATION METHOD AND APPARATUS, DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/047211
Kind Code:
A1
Abstract:
A neural network-based data verification method, being used for vehicle maintenance. The method comprises the steps: obtaining vehicle maintenance data to be verified that is transmitted by a user, and extracting a vehicle damage picture and a repair scheme in the vehicle maintenance data; matching a corresponding standard repair scheme and a corresponding historical damage picture thereof from a repair scheme database on the basis of the repair scheme; separately performing damage degree analysis on the damage picture and the historical damage picture by means of a part damage degree analysis model, calculating a difference value between the degree of damage reflected by the damage picture and that reflected by the historical damage picture, and according to the difference value, verifying the vehicle maintenance data; in the case that the corresponding standard repair scheme does not exist in the repair scheme database, invoking a target repair scheme by means of a maintenance data verification model according to the degree of damage, and verifying the vehicle maintenance data according to the target repair scheme; and if a verification result is not correct, replacing the repair scheme with the standard repair scheme or the target repair scheme, and providing reliable vehicle maintenance data.
Inventors:
LI YUE (CN)
Application Number:
PCT/CN2020/093590
Publication Date:
March 18, 2021
Filing Date:
May 30, 2020
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06F16/9535
Foreign References:
| CN110781381A | 2020-02-11 | |||
| CN108985466A | 2018-12-11 | |||
| CN108563214A | 2018-09-21 | |||
| CN105955243A | 2016-09-21 | |||
| CN104850114A | 2015-08-19 | |||
| US5099436A | 1992-03-24 |
Attorney, Agent or Firm:
SCIHEAD IP LAW FIRM (CN)
Download PDF: