Violence Detection in Real Life Videos using Deep Learning

Abstract

These days, it is essential to avoid or identify violence as soon as possible because it is spreading in an unpredictable way. Violence must be identified on real-time films taken by numerous surveillance cameras at all times and in all locations, which makes it difficult to perform. It might be time-consuming to continuously monitor CCTV surveillance cameras, thus it is essential to automatically spot any unusual activity. As soon as violent behaviours take place, it should be able to make a trustworthy real-time detection and notify the appropriate authorities. The dataset contains both violence and non-violence videos from real life situations. A methodology for detecting violence has been presented by us that uses a network similar to the U-NET with the encoder mobilenetv2 to extract spatial features before moving on to an LSTM block for the extraction of temporal features and binary classification. The results of the trial revealed that the precision is 95% and the accuracy is 94% utilising a dataset based on real life situations. The recommended model uses minimal computer resources while yet producing useful results.