node-js-server

JS-ARUCO2 sample on NodeJS

This sample demonstrate how to use js-aruco2 in a NodeJS application to detect markers on a video stream generated with FFMPEG.

Requirements

• FFMPEG installed or ffmpeg executable available in this folder
• NodeJS + NPM

Configuration

• Configure your camera input editing the config.camera, config.cameraWidth and config.cameraHeight parameters in the package.json (instructions for setting the camera on different OS are available in the package.json).

• Optionally configure the server port (default 8081) editing the config.port attribute.

• Optionally change the endpoint secret editing the config.secret attribute, in order to avoid stream hijacking if deployed on centralized server.

• Optionally change the ArUco markers dictionary used editing the config.dictionaryName attribute (default ARUCO_MIP_36h12). Available values are ARUCO_MIP_36h12 and ARUCO.

• Optionally change the default hamming distance of the selected dictionary editing the config.maxHammingDistance attribute.

Getting started

1. Initialize the project:

   npm run init
   

2. Start the server:

   npm run server
   

3. Start the FFMPEG stream of your camera:
   • For Window based OS:

     npm run stream_win
     

   • For Linux based OS:

     npm run stream_unix
     

   • For MacOS based OS:

     npm run stream_osx
     

4. The recognized markers are visualized in the console. Additionally (mainly for testing purposes) the last processed image will be continously saved as JPEG in this folder with name camera_out.jpg

Notes

• In order to avoid frames dropping and not overload the system, ffmpeg will send 1 frame per second to the server.