Video recognition method, system and device based on pattern mutation frame and feature calculation

Abstract

The invention discloses a video identification method, system and device based on pattern mutation frame and feature calculation. The method comprises the following steps of: calculating the pattern data of each picture frame in the video; according to the pattern data of the picture, calculating the pattern similarity between two adjacent pictures; comparing the similarities of image patterns between two adjacent frames to obtain abrupt frames; obtaining a characteristic code to be identified according to the abrupt change frame; according to the pre-stored video characteristic code, searching and matching the characteristic code to be identified, and obtaining the video recognition result; Wherein the to-be-identified signature and the video signature are both in a text format. The invention can avoid missed judgment and misjudgment, improve the identification accuracy, greatly reduce the difficulty of sample library retrieval, has low complexity, and can be widely applied in the field of image processing technology.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a video recognition method, system and device based on sudden pattern frame and feature calculation. Background technique [0002] Glossary: [0003] Pattern is the abbreviation of picture fingerprint. Image fingerprint, also known as image hash (Image Hash), specifically maps the high-dimensional content features of the image into the binary space to generate a low-dimensional hash sequence that can represent the image. [0004] The abrupt frame refers to a frame in which the content of the picture is quite different from that of the previous frame in the continuous multiple frames of the video stream, and the difference can be measured by the pattern difference. [0005] The abrupt frame interval refers to the frame number difference between two consecutive abrupt frames in the video stream. [0006] In the field of TV advertisement broadcast monitoring, it is neces...

AI Technical Summary

Problems solved by technology

[0008] The key to the above method lies in the calculation method of the feature value of the sample video. There are two problems in this calculation method: 1) The video to be checked has few own features, which is prone to missed and misjudged, and the recognition accuracy is low; 2) , In order to improve the accuracy of video recognition, the video feature value obtained by the existing calculation method has a large amount of information, which increases the difficulty of sample library retrieval and high complexity

Images