Computational Comparison of CNN Based Methods for Violence Detection

Abstract

In this paper, we approach with four different CNN-based models i.e., VGG-19, VGG-16, InceptionV3 and MobileNetV3 with an improved version of the previous models for violence detection and recognition from videos. The proposed models use the pre-trained models as the base model for feature extraction and for classification after freezing the rest of the layer, the head model is prepared with averagepooling2D of (5, 5), and after flattening only one dense layer having 512 nodes with ‘ReLU’ activation function, dropout layer of 0.5 and last output layer with only 2 classes and ‘softmax’ activation function. This head model of fully connected layers was used in the proposed models. These models are trained and evaluated on the Hockey fight dataset and Real life violence situations detection datasets. The experimental results are far better in terms of accuracy and other performance metrics and the models have reduced parameters and less computational time than previous models.

Violent Video Classification with Transfer Learning Approach Using Inception-V3 and Support Vector Machine

“…F1-score measures test accuracy, a Harmonic Mean between precision and recall, with a score between 0 and 1. In present study will focus on calculating the F1-score (9) to check the model behavior using following formula (1).…”

“…Researchers are increasingly interested in violence detection and prevention due to its accessibility and accessibility. Violence detection and classification are critical in spotting anomalies in video situations, particularly ones with minimal or no violence (1) . Public protests and riots can escalate into violent incidents, necessitating the establishment of intelligent systems that are efficient and precise.…”