What is cv2.inRange used for? What is its syntax suppose to be?

inRange#python

asked 2017-11-21 06:52:05 -0600

Santhosh1
318 ●11 ●15 ●37

updated 2017-11-21 06:54:01 -0600

I'm trying to replicate the Magic Wand tool of the Photo Shop

I want to sent a tolerance to my selected ROI in an image, as mentioned in this post.

I found this link, thought it might be useful.

I came across this cv2.inRange but this is the Documentation, I found on inRange.

I have no clue as to what kind of values have to be actually passed into inRange.

If you have any better idea of how to replicate Magic Wand, do share. I posted my idea here

edit retag flag offensive close merge delete

Comments

1

please do NOT try to abuse a computer-vision library to reproduce photoshop.

berak ( 2017-11-21 07:05:01 -0600 )edit

1

@berak I am, do and always will respect to the Field of Computer Vision. I need its property which has been around for a decade. I don't know how to code it in OpenCV

Santhosh1 ( 2017-11-21 07:14:54 -0600 )edit

yea, i'm just saying ..

berak ( 2017-11-21 07:24:11 -0600 )edit

add a comment

0

answered 2017-11-22 04:59:14 -0600

StevenPuttemans

20029 ●16 ●82 ●207 http://stevenputtemans...

updated 2017-11-22 04:59:41 -0600

Replicating magic wand, is by implementing a technique called GrabCut. You need to define a color range of pixels on which you will force your wand, but the general idea can be found in this paper:

http://citeseerx.ist.psu.edu/viewdoc/...

edit flag offensive delete link

add a comment

0

answered 2017-11-22 08:53:33 -0600

kbarni

4134 ●1 ●13 ●49

The magic wand tool in Photoshop uses the active contours (or snakes) algorithm to follow the outline of an object. It was present in old OpenCV implementations, but it was removed as it wasn't very accurate. You can find the source code here, you can reuse it in your project.

Anyway, probably the GrabCut will give better results - even if technically it isn't the same algorithm.