Method and system for detecting network pornography videos in real time

Abstract

The invention discloses a method and system for detecting network pornography videos in real time. The method comprises the following steps: according to the length of a network video, establishing a pre-extracted frame number queue KFN of key frames, wherein KFN={n1, n2,..., nN} with the condition of n1<n2<...<nN; according to the pre-extracted frame number queue of key frames, extracting a key frame in order and decoding the key frame; carrying out pornographic content detection on the decoded key frame and judging whether the key frame contains the pornographic content; according to the detection result of the pornographic content of the single key frame together with the detection results, which are obtained previously, of other key frames, carrying out decision fusion judgment; if judging the key frame contains the pornographic content, considering the video contains the pornographic content and completing detection; if judging the key frame does not contain the pornographic content, considering the video does not contain the pornographic content and completing detection; and if whether the key frame contains the pornographic content is uncertain, continuing detecting the single key frame.

Description

technical field [0001] The invention relates to the technical field of network pornographic video detection, in particular to a method and system for real-time detection of network pornographic video. Background technique [0002] With the rapid development of network bandwidth and video technology, the number of online videos and the scale of users have shown a huge increase. Subsequently, the problem of the proliferation of pornographic videos on the Internet has become increasingly serious, and has seriously interfered and polluted the existing network environment. In order to purify the network space and curb the wanton proliferation of pornographic videos on the Internet, a very important technical means is to apply pornographic content detection technology for online videos to effectively supervise online videos. [0003] Existing pornographic video detection methods are all detection methods for traditional video programs, and there is no detection method specificall...

AI Technical Summary

Problems solved by technology

However, online videos are usually short in length, monotonous in content, and contain relatively few shots

At this point, time-consuming shot segmentation processing is unnecessary

[0006] (2) The model-based detection method has a large amount of online calculation

This greatly affects the efficiency of the detection program running online

[0007] (3) Features carry a lot of noise

Due to the uneven visual quality of online videos, fine features extracted in traditional methods (such as torso detection, skin texture analysis, audio features) have large performance fluctuations on online videos, and these features will bring noise interference to the detection process

Images