A multi-label video event detection method based on lstm network

Abstract

A multi-label video event detection method based on LSTM network: generate a model based on LSTM network for all video image sequences in the Concurrent Event Dataset database, including: obtaining the probability distribution of each video image sequence corresponding to all label sets and passing the obtained The probability distribution updates the network to obtain a model based on the LSTM network; for the video to be detected, the model based on the LSTM network is used to obtain the probability distribution of the corresponding label set. The present invention generates multiple event reports of surveillance video through the method of the present invention, avoiding object monitoring and tracking process; for the processing of surveillance video, a brand-new network structure is designed based on the long-term and short-term memory network; greatly improving the efficiency and efficiency of surveillance video processing Robustness, which improves the poor recognition effect of traditional methods for multiple events occurring at the same time.

Description

technical field [0001] The present invention relates to a video event detection method. In particular, it relates to a multi-label video event detection method based on LSTM network. Background technique [0002] The purpose of surveillance footage is to monitor human behavior, activity, or other visual events that occur in the footage. Now, there are more and more applications in the fields of military, public safety, business and law. The development of this field has arisen with the increase of cheap computing power, the popularity of digital cameras, and the popularity of image sensors. In addition, the inefficiency of manual monitoring and monitoring systems (eg Ref. [1]) is also a factor. We all know that it is impossible for humans to process large amounts of data constantly. For this reason, errors usually occur. Furthermore, the resources to manually observe the output are very expensive. Therefore, how to know the content information in the video has been a p...

AI Technical Summary

Problems solved by technology

For the data itself, the main challenges are low resolution, large amount of data, complex event sets and scenarios, and blocked data sources

For the method, there are mainly the following limitations: 1) Many methods rely on foreground and background segmentation technology, but this technology will cause error accumulation

2) Many methods rely on detection and tracking, however for different videos and moving objects, the robustness of detection and tracking is low

These disadvantages reduce the efficiency of time analysis

3) When the amount of data increases, the amount of calculation will increase significantly

Therefore, the two recognition methods lose the time of simultaneous occurrence

Images