Project RVD : Realtime Violence Detection with Raspberry Pi

• Client: J.Oh
• Starts at: 2021/04/08
• Ends at: 2021/05/17
• Members: Sukwang Lee, Hyewon Rho, Jeonghun Park, Jimin Chae, Yeong-Min Choi

Introduction

• Violence detection by using MobileNet + LSTM (Binary classification : Violence / Non-Violence)
• Add captions of violence dection result on video screen (video file or realtime video streaming) >> Fast Link
• Using Raspberry Pi & camera module for realtime video streaming >> Fast Link

Environment

• Linux OS
• Python ≥ 3.8
• OpenCV ≥ 4
• Tensorflow ≥ 2.4.0
• Keras 2.4.0

Tools

• Jupyterlab ≥ 3.0
• Pycharm ≥ 2
• Raspberry Pi +3
• Camera Module

Result

Examples

Change of Accuracy & Loss of Model

• Result Log : View Table

Reference

Datasets

• RWF2000 : https://github.com/mchengny/RWF2000-Video-Database-for-Violence-Detection
• Hocky : http://visilab.etsii.uclm.es/personas/oscar/FightDetection/index.html
• raw.zip : https://github.com/niyazed/violence-video-classification
• cam1, cam2 : https://github.com/airtlab/A-Dataset-for-Automatic-Violence-Detection-in-Videos/tree/master/violence-detection-dataset/non-violent/cam1
• RVD : -

Examples

• UCF Anomaly Detection Datasets : https://webpages.uncc.edu/cchen62/dataset.html
  • Those video files were used only for making output videos & GIFs.

We're Inspired from them

• Pedro Frodenas's Github
• Souhaiel BenSalem's Github