CODEWITHSUNDEEP
image-processingembedded-linuxface-recognitionbeagleboard
Akshay
Akshay

Reputation: 31

Real time image processing

I am working on a project that requires me to recognize faces using a camera connected to an embedded board. This board must use linux platform and perform real time image processing in opencv. I will be using one camera only. The board should be affordable as well as the performance should be good. I have read about raspberry pi, beagleboard and the pandaboard. . Which is the best possible board for this-???

Upvotes: 0

Views: 971

Answers (1)

TomP89
TomP89

Reputation: 1460

I can't specifically advise you on which board to choose, but in more general terms, the specification of the board you choose should depend largely on what performance you are expecting from the facial recognition.

If you are just performing one recognition every few seconds or so then an average spec'd board would likely be fine, however if you want to perform many recognitions per second then you you will want a much higher performance board, especially where image processing is concerned.

I would advise you build your program (if possible) on a standard PC and get a feel for its performance on there, and profile the most expensive parts of the operation then you can make a more informed choice.

Upvotes: 2

Related Questions