Ask Your Question
0

Weak edge(Curve) detection in low contrast (Update)

asked 2017-11-13 19:51:32 -0600

Ziri gravatar image

updated 2017-11-13 22:46:44 -0600

Hi Everyone . After trying different methods including sobel , canny , shaar , shen , FFt filtering , haar wavelet edge detection , CLAHE for contrast enhancement ...

I ended up using Adaptive threshold to detect a curve in low contrast .

Can you please suggest a method to group blobs belonging to the curve ? ( I already tried erode/dilate with "line " structuring element ).

Any suggestion is welcome

Thank you!!

image description

Original Image : image description

edit retag flag offensive close merge delete

Comments

Would be better if you Include your Original Image, Instead of this thresholded one.

Balaji R gravatar imageBalaji R ( 2017-11-13 22:38:17 -0600 )edit

Hi , I added original image (curve shape and orientation are random ) . Thank you.

Ziri gravatar imageZiri ( 2017-11-13 22:47:59 -0600 )edit

1 answer

Sort by ยป oldest newest most voted
1

answered 2017-11-13 20:01:25 -0600

Tetragramm gravatar image

So, this is a bit out of the box, but have you tried a particle filter tracking algorithm?

What you have there is essentially a bunch of really noisy measurements of a target. Each column is one measurement. You have a probability of missing the true target, and a probability of detecting a false target, which you can probably make good guesses at from the data.

What you would be tracking is the vertical location in the data, with the horizontal being "time".

I don't have any links handy, but a search for particle filtering should turn up some tutorials. This problem is pretty simple, with one "target", so that's nice.

Of course, this assumes that the line is as you have it shown, going from one side to another, no doubling back. It gets a little harder if your problem is like that, but not impossible.

edit flag offensive delete link more

Comments

Thank you. I will try your suggestion.

Ziri gravatar imageZiri ( 2017-11-13 20:13:29 -0600 )edit
Login/Signup to Answer

Question Tools

1 follower

Stats

Asked: 2017-11-13 19:51:32 -0600

Seen: 50 times

Last updated: Nov 13