Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Board for image processing

Status
Not open for further replies.
Z

zakbenavi

Newbie level 2
Joined
Feb 18, 2018
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
21
Hi everyone,
I'm working on a wearable object detection project using image processing.
I'm writing the code in MATLAB and then translating the code via the MATLAB Coder to C.

Which board is preferred to use in such a project?
I also need that the board will support a camera module.
It needs to be small (wearable).
It needs to have enough pins to connect speakers.

Thanks,
Zak
 

Hi,

Image processing is such a wide area, thus it is impossible to answer..

Pixel count, bits per pixel, calculation algorithms, data output rate, interfaces, programming language...
...are at least necessary to give some kind of good recommendation.

Klaus
 

Hi KlausST,
As I mentioned before, I'm using an object detection project.
The algorithm is basically edge detection and filters.
 

Hi,

By far not enough information.

Maybe this helps to understand the problem:
"I want to visit my mother for her birthday. What transport medium should I use?"
Even the new information that the birthday is in two days ... doesn't help much to answer the above question.
--> There are still too many unknown details.

Klaus
 

Raspberry Pi with Linux supports camera and broad range of C/C++/Python compilers. But the performance might be the issue.
 

As I mentioned before, I'm using an object detection project.
The algorithm is basically edge detection and filters.
Click to expand...

Your specifications are still broad; the edge enhancement process is just the initial stage to highligh the contourn feature. There are algorithms that require more or less core processing, particularly if the image varies in size and orientation within region of interest, and of course depends on the absolute size in pictures of the scaled picture. There are powerful ARM based designs which could meet the need. I would also consider using the OpenCV library, with which you could easily perform these initial filtering steps. Any Linux distro is able to deal with that,
 

Some specifications (not an exhaustive list) that need to be considered:
  1. Frame rate
  2. Image resolution - i.e. number of horizontal pixels, number of vertical lines
  3. Colour, RGB, B&W
  4. Colour encoding and camera output, PAL, NTSC, USB.
  5. Number of bits per pixel
  6. Power supply voltage and current availability
  7. Size of system

Processing delay - do you want minimal delay in processing images so that processed images appear 'live', i.e. one frame or so delay behind real time
For 'live' image processing an FPGA ( depending on video specifications) might provide sufficient processing power. An example board: https://www.crowdsupply.com/jie-zou/rggber-dev-kit

For 'static' and lower resolution 'live' image processing where images are processed at a more leisurely pace a fast SBC may be sufficient e.g. Raspberry Pi
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top