Camera Output Processing Platform Suggestions

[arico3]

Hello Everyone!!

We are working on an application that requires the continuous monitoring of camera output.
The Device will be a standalone device, thus the requirement of an embedded platform.

I am attaching a link of a sample video which we need to process.

Please suggest what kind of platform will be best suited for it.

 

[kaiserschmarren87]

Do you want to track the motion in real time ? Or process them ? Or Extract specific data from the video or frame ? There are several platforms available and it depends on the requirement and its complexity.

Below are some of the techniques available:
1. Interfacing the Analog Camera with FPGA Board for Real-time Video Acquisition.
http://www.mecs-press.org/ijigsp/ijigsp-v6-n4/IJIGSP-V6-N4-4.pdf

2. Embedded Real Time Video Monitoring System using Arm.
http://www.iosrjen.org/Papers/vol3_issue7%20%28part-6%29/C03761418.pdf

3. Beaglebone: Video Capture and Image Processing on Embedded Linux using OpenCV.
http://derekmolloy.ie/beaglebone/beaglebone-video-capture-and-image-processing-on-embedded-linux-using-opencv/

 

[arico3]

Thanks for your reply and suggestions.

The video shows a powder kept in a tube that liquefies due to its internal reactions over a period of time.

We wanted to analyze this change in state of the powder.

The processor should be able to capture and analyze the images (video stream), during the reaction period, for the change in state taking place.

 

[kaiserschmarren87]

I got the logic behind it. I referred to this particular document to get some basics:
**broken link removed**

I hope this is what you mean in general. So you want to track the changes in the video frames to analyze the change in the state of the powder.

 

[arico3]

Thanks a ton for your kind reply. This is similar to what we wanted to work on.

However I require your suggestions on how to implement the same on a stand-alone embedded system, bypassing the use of a computer

Thanks

 

[kaiserschmarren87]

I am glad that it helped you. To bypass the computer and implement it on FPGA alone, then you will need to have a data flow or flow chart made by yourself on how to implement it. It means, you have to apply several algorithms for image processing in FPGA using VHDL or Verilog. Also you need to know what kind of Camera will be interfaced to FPGA, using USB or CameraLink interface.

So if you make a plan for this then it will be easy for you to analyse and I am not an expert but a beginner with less experience in this domain.

 

[arico3]

But we are planning to implement it on an embedded linux based platform.
Any suggestions regarding it?