Background live video auditing method

Abstract

The invention discloses a background live video auditing method, which comprises the following steps that: a user initiates microphone connection, and sends a video, an audio and a video screenshot to a cloud end; machine auditing: performing detection through a picture recognition technology, and actively processing violation; manual auditing: binding a video room with an auditing worker, and then performing auditing by the auditing worker; group leader auditing: enabling the group leader to check the working condition of the auditing worker; and checking the auditing result of the auditing worker in a specified time period through the system, and checking whether misjudgment and misauditing exist or not. The method has the beneficial effects that machine auditing and manual auditing are combined, on one hand, the auditing efficiency is improved through machine auditing, the situation that manual auditing is not timely is avoided, and the labor intensity of manual auditing is reduced; the manual auditing solves the problem that content is difficult to audit through machine auditing; the auditing effect on the auditing content is improved; group leader auditing is mainly used for auditing an auditing result of the manual auditing, so that user actions are misjudged and the user is forced to be offline caused by subjective errors of manual auditing is prevented.

Description

technical field [0001] The invention relates to the technical field of software management, in particular to a background live video review method. Background technique [0002] With the rapid development of Internet technology, webcasting is more and more widely used in people's daily life. Due to the diversity of live broadcast content, it may involve topics such as terror, violence, pornography or political sensitivity. Therefore, users During the live broadcast, the background also needs to judge whether the live broadcast content violates regulations, so as to prevent the illegal content from spreading rapidly on the Internet. [0003] In view of the above problems, the existing technology usually adopts machine review or manual review; as follows [0004] 1. A Chinese patent discloses a method for identifying illegal pictures and related devices (application number: CN201811157555.1), which is used to improve the efficiency of pornography services. The method in the ...

AI Technical Summary

Problems solved by technology

[0007] 1. In the existing technical review methods, only one of the review methods is adopted, manual review or machine review. If it is only manual review, it will increase the labor intensity of the staff and reduce the review efficiency; if it is only machine review, because many content machines are Unable to audit, this will affect the audit effect, resulting in inaccurate identification results, which may misjudge user actions and force offline users;

[0008] 2. The existing technology does not have a supervision mechanism for manual review, because manual review inevitably has certain subjective factors, so there will also be cases of misjudgment and misjudgment

Images