Convolutional neural network model-based violence and terrorism video detection method

Abstract

The invention discloses a convolutional neural network model-based violence and terrorism video detection method, and relates to computer vision and machine learning. The method comprises the following steps of 1) training a deep neural network model; and 2) detecting violence and terrorism videos online. By utilizing a low-level feature of a deep learning model combination, a more abstract high-level representation attribute or feature is formed to discover distributed feature representation of data. Video image feature descriptors with good description capabilities can be obtained through the model. The feature descriptors cover feature information, at all levels from low to high, of video images, so that the detection precision and recall rate of violence and terrorism videos are greatly improved and increased. A deep convolutional network is trained through a small amount of samples to obtain excellent detection performance. The detection precision of terrorism pictures reaches more than 99%, and the recall rate of the terrorism pictures reaches more than 98%. The detection precision of terrorism videos reaches 95%, and the recall rate of the terrorism videos reaches 99%. The training process is free from artificial participation, and massive data is generated automatically according to a small amount of the samples.

Description

technical field [0001] The invention relates to computer vision and machine learning, in particular to a method for detecting violent and terrorist videos based on a convolutional neural network model. Background technique [0002] The rampant terrorism has brought great disaster and pain to people all over the world. It has caused heavy casualties and property losses, hindered national security and social stability, and its harmfulness far exceeds that of ordinary criminal crimes. One of the main channels through which terrorist ideas are disseminated is via video images. At present, the detection of terrorist videos is mainly through manual review and labeling, which consumes a lot of financial and material resources. If the video or image file is re-edited (such as remake, transcription), manual review is required again, which is very inefficient. Therefore, in the face of the Internet with an increasing amount of data, a new type of technology is needed to automatical...

AI Technical Summary

Problems solved by technology

[0005] The existing various video image feature descriptors (such as sift, gist, mser, hessian, etc.) have limited description capabilities, and it is difficult to fully and accurately describe the content of video images, especially in violent and terrorist videos that need to be detected for specific targets , resulting in relatively low precision and recall of the detection work

so in most cases only manual review

Manual review is time-consuming and laborious, which is very limited in the current era of information explosion

Images