Skip to content

Latest commit

 

History

History
37 lines (30 loc) · 1.34 KB

README.md

File metadata and controls

37 lines (30 loc) · 1.34 KB
Raw

Cloud Server

Overview

Cloud server receives the video streams and performs Traffic Density Measurement algorithm on them. If the encoding request is sent by the fog node, the cloud server computes the heuristic encoding parameters and sends them back to the fog node.

In the detection_logs folder, the results of the traffic density measurement algorithm can be found.

Prerequisites

  • OpenCV
  • sklean
  • pillow
  • numpy 1.15.0
  • torch 1.3.0
  • tensorflow-gpu 1.13.1
  • CUDA 10.0
pip3 install -r ./traffic_density_measurement_algorithm/requirements.txt
  • Download YoloV4 model weights and move the in this folder: Click Here

Usage

In order to initialize the server, we run the following from the current folder:

  • --encoding_time or -et is an optional argument used to pass the duration of video analysis to calculate the encoding parameters (in seconds) 
    python3 server.py [-et <encoding_time>]
  • Example 
    python3 server.py
    OR 
    python3 server.py -et 20