Alert icon
We're changing our privacy policy. This stuff matters.  Learn more  Dismiss

Embedded Image Processing - Color Segmentation using Fuzzy Inference System

Loading...

Sign in or sign up now!
Alert icon
Upgrade to the latest Flash Player for improved playback performance. Upgrade now or more info.
1,453
Loading...
Alert icon
Sign in or sign up now!
Alert icon

Uploaded by on Nov 26, 2009

OpenCV library built for an embedded processor (i.MX51 - ARM Cortex A8)
This program performs color segmentation on frames grabbed from an USB Camera (webcam) in real time. It uses Ubuntu Linux as Operating System and the board used was the EVK board from Freescale Semiconductors (Freescale's i.MX51 processor):

http://www.freescale.com/imx51evk

the process used to perform this operation was:

0 - Re-scaling (the input image was downsized)
1 - Color System Conversion RGB to HSV
2 - Color filter (thresholds on both HUE and SAT planes for the first result and Fuzzy for the second)
3 - Color System Conversion HSV to RGB
4 - Color Conversion RGB to Grayscale
5 - Smooth with Gaussian filter 5x5

in order to get a complete operation, mathematical morphology (erode and dilate) and also binarization (after smoothing) are needed in this case.

the System's clock speed is 800Mhz with 512MB of RAM memory, and with all those steps above the detection time is only 1,3ms per frame.

Fuzzy Inference System information:

Discretization Points: 20
Inference Rules: 3 for each color
Implication Method: Mandani
Defuzzyfication Method: Largest of Maximum (LOM)

Link to this comment:

Share to:
see all

All Comments (0)

Sign In or Sign Up now to post a comment!
Loading...

Alert icon
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more