Video stream processing and archival engine

In complex video processing of 24 x 7 hours of live feeds, there are instances we have different video sources which are are to be captured irrespective of Hardware. Our solution helps to capture the video feed from any source and converts that into image of interest in the desired format and moves to an archival store.

Background

  • Goal: To build an application which helps to capture the image at an instance from an video feed
  • Customer is a Aviation Sector Startup providing solution on security infrastructure
  • Interested to build Cross platform app to be deployed across multiple video resolution
  • Need is that custom configured during deployment to support different frames per second scenarios

Challenges

  • Integrating a video streaming solution to capture the video stream
  • Move the grabbed file to an storage infrastructure for future use.
  • Take care of memory leaks and optimize performance for deployed on the security infrastructure.
  • There should not be any loss of color and other meta data from the video stream.
  • Effortlessly convert any files into other desired file formats with no limitations and quality loss.

Solution Approach

 

We build a configurable application in Python with Qt which can capture the video stream integrated with the hardware. Python Qt was leveraged to have a cross platform application. Also adopted JSON based configurator including parameter validation at the front end. The solution was built in such way to manage different hardware configuration, frames per second and target storage medium.

Benefits

  • Developed and deployed the solution in shorter time frame
  • Co-Engineered the solution with different partners collaboratively
  • Built a configuration application to manage the customization with onsite support