Violence Video Classification Performance Using Deep Neural Networks

Abstract

Violence is autonomous, the contents that one would not let children to see in movies or web videos. This is a challenging problem due to strong content variations among the positive instances. To solve this problem, implementation of deep neural network to classify the violence content in videos is proposed. Currently, deep neural network has shown its efficiency in natural language processing, fraud detection, social media, text classification, image classification. Regardless of the conventional methods applied to overcome this issue, but these techniques seem insufficiently accurate and does not adopt well to certain webs or user needs. Therefore, the purpose of this study is to assess the classification performances on violence video using Deep Neural Network (DNN). Hence, in this paper different architectures of hidden layers and hidden nodes in DNN have been implemented using the try-error method and equation based method, to examine the effect of the number of hidden layers and hidden nodes to the classification performance. From the results, it indicates 53% accuracy rate for try and error approach, meanwhile for equation based approach it indicates 51% accuracy rate.