کنفرانس بین المللی فناوری اطلاعات و دانش

صفحه اصلی / دوازدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

Violence detection using one-dimensional convolutional networks

نویسندگان :

Narges Honarjoo¹ Ali Abdari² Azadeh Mansouri³

1- دانشگاه خوارزمی 2- دانشگاه خوارزمی 3- دانشگاه خوارزمی

کلمات کلیدی :

violence detection، one-dimensional convolution، real-time application، temporal pooling

چکیده :

Violence detection in surveillance video processing is a useful capability helping discover abnormal events in different places. Utilizing methods considering the accuracy and complexity simultaneously can provide systems suitable for real-time applications. In this paper, by exploiting one-dimensional convolutional networks a new approach is proposed which extracts the temporal features across consecutive frames properly. This approach not only represents a series of frames with a robust feature vector, but it also is low-complexity and can be applied for real-time applications. The experimental results on Hockey, ViolentFlow reveal the efficiency of the proposed method