Ask Your Question

Detecting the status of pedestrian traffic signal.

asked 2014-08-05 08:08:06 -0500

prasanna aarthi gravatar image

updated 2014-08-05 08:11:55 -0500

I am pretty new to open CV.My project involves identifying traffic signal status in pedestrian crossing,Its either a walk symbol in green,red or yellow.In some places its also text like "WALK" and "DONT WALK". I need to identify this using open CV .My initial thoughts are

Extracting the highest RGB value pixels. Using clustering algorithm to group the pixels to a cluster.
Using template matching to filter out other clusters.
Finally find the color the cluster that has passed filtering.

Is this an optimal way of doing? Or is there any other algorithm or approach I need to take in order to make the whole process fast and accurate.Please help

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2014-08-05 09:31:57 -0500

Haris gravatar image

In other way, instead of using RGB better to use HSV in color based segmentation, by which you can easily crate threshold image based on the lower and higher color value.

If you know the color to be segmented, you could do

Do these steps for every predefined color range and decide which color is most dominant in the frame.

Also see the answer here might be helpful to select proper hsv range.

edit flag offensive delete link more

Question Tools

1 follower


Asked: 2014-08-05 08:08:06 -0500

Seen: 229 times

Last updated: Aug 05 '14