Video counterfeiting detection method and system, storage medium and video monitoring terminal

Abstract

The invention belongs to the technical field of video monitoring information processing, and discloses a video counterfeiting detection method and system, a storage medium and a video monitoring terminal, and the method comprises the steps: dividing a video frame sequence into overlapped sub-frame groups, and carrying out the similarity check of each sub-frame group as a query segment; taking thecorrelation between adjacent frames in the sub-frame groups as sub-frame group features, and calculating the similarity between the sub-frame groups; if the similarity is greater than a threshold value, selecting a corresponding sub-frame group as a forged frame group candidate; and determining the counterfeiting type and position by detecting the time relationship between adjacent frames in the sub-frame group. According to the method, the sample video is copied and forged in two frames; when the method is used for detection, results of all sample videos are averaged to obtain a final result,and experimental results show that the method can effectively detect copy forgery in the videos, can effectively detect two types of copy forgery and determine the position forgery position, and is higher in precision ratio and recall ratio compared with a literature algorithm.

Description

technical field [0001] The invention belongs to the technical field of video surveillance information processing, and in particular relates to a video forgery detection method, system, storage medium, and video surveillance terminal. Background technique [0002] At present, video surveillance is not only an important means in the field of social security, but also an important way of judicial evidence collection. Due to the wide application of video surveillance systems, the number of video surveillance cameras continues to increase, and massive surveillance video data is facing the security challenge of being tampered with. Ensuring the integrity and authenticity of surveillance video data is critical. Video surveillance is a powerful means to prevent crimes and ensure social harmony and stability. In the field of judicial evidence collection, surveillance video is an important basis for courts and investigators to learn the truth. However, the rapid development of vide...

AI Technical Summary

Problems solved by technology

The previous detection methods based on the correlation of adjacent frames need to compare all possible frames in any length and position to search for duplicates, and the computational complexity is high, especially when the degree of correlation falsification is high and the number of data frames is large. The precision rate and duplication check rate of the two kinds of repeated forgery between frames are relatively low

[0004] Through the above analysis, the problems and defects of the existing technology are: the current detection method based on the correlation of adjacent frames needs to compare all possible frames in any length and position to search for duplicates, and the computational complexity is high. The precision rate and duplicate check rate of repeated forgery are relatively low

[0005] The difficulty in solving the above problems and defects is: compared with digital images, video detection in the time domain and space domain is very difficult, and the technology is not mature, while the traditional correlation-based detection method has high computational complexity. If the degree of correlation falsification is high and there are many data frames, the reliability is relatively low

[0006] The significance of solving the above problems and defects is: video surveillance is an important means of evidence collection. If the reliability of video forgery monitoring technology is not high, the credibility of video data collected through various camera equipment will be seriously affected.

When the video is forged but difficult to detect, or when the accuracy rate and repetition rate are low, the fairness and correctness of the judiciary will be seriously affected

Images