Information auditing method based on real-time video

Abstract

The invention relates to an information auditing method based on a real-time video, and the method comprises the following steps that: S1, an auditor initiates a video invitation to a to-be-audited user in a background system, and the to-be-audited user answers the video invitation at a terminal of an app; s2, the app collects a video of the user in real time and transmits the video to auditors, the app performs facial analysis on the user to be audited and transmits analysis data to a background system, and the background system processes and analyzes the analysis data and gives out a corresponding data result; s3, the auditor judges a final auditing result according to the communication with the user in the background system comprehensive analysis video and the data result displayed by the background system; and S4, the auditor feeds back an auditing result to the user through the background system, and stores the video in the auditing process into the server 2. According to the invention, the real-time video is introduced into an information auditing scene, and technologies such as facial analysis are combined, so that the information auditing efficiency and quality are improved.

Description

technical field [0001] The invention relates to the technical field of information review, in particular to an information review method based on real-time video. Background technique [0002] Internet loans refer to loans provided by Internet companies to customers through the loan companies controlled by them through the Internet. Online loans should abide by the existing regulatory provisions of loan companies, give full play to the advantages of online loans, and strive to reduce financing costs for customers. [0003] Before making an online loan, information review is usually required to verify the identity information of the borrower. Information review usually means that the information or evidence provided needs to be proven reliable, which is widely used in credit risk judgment when applying for credit, public security and other fields. In scenarios that require information review such as loans, it has gone through stages such as counter review, telephone review, ...

AI Technical Summary

Problems solved by technology

[0004] The above-mentioned existing technical solutions have the following defects: 1. It is inconvenient to audit at the counter, and it is necessary to make an appointment to go to an audit location

2. Telephone review only has one dimension of voice, so the judgment is biased

3. Although robot voice review can save labor costs, it needs to convert the customer's voice recognition into text first, then match the text with the database maintained in advance, and return the voice reply according to the matching result

Because there are various conversion processes in the middle, the information is likely to be misidentified, and the whole process has a large time delay, it is difficult to guarantee the natural communication back and forth, so the audit results are often biased

Therefore, the above three technical solutions all reduce the accuracy of information review results to a certain extent

Images