what could it be a good descriptor for simple greyscale symbols? [closed]

asked 2017-03-07 11:59:37 -0600

pv gravatar image

I need to find a good method to extract key points from drawing symbols like elements in CAD drawing (eg in 2 objects for the bathroom that can be placed on a floor plan).

I've already tried all type available in openCV (FAST, SURF, Harris, ..) but all of them failed to give me any point.

The images are blank and white (with some sort of grey scale) and I'd like to find the position of a limited set of symbols inside bigger images containing other unknown objects/shapes (that I'm not interested in anyway).

The basic idea is to get the descriptors for my symbols and check if the target image contains similar descriptors.

edit retag flag offensive reopen merge delete

Closed for the following reason the question is answered, right answer was accepted by sturkmen
close date 2020-11-11 09:08:50.028508

Comments

1

it sounds a bit , like you're the umpteenth person trying to abuse feature matching for object classification

(that plan will never work)

maybe have a look at BagOfWords

berak gravatar imageberak ( 2017-03-07 12:17:43 -0600 )edit

Also, Hu Moments can give you a good descriptor of a binary/greyscale symbol, but it's got to be one connected component or you have to know how to group them automatically.

Tetragramm gravatar imageTetragramm ( 2017-03-07 22:36:46 -0600 )edit

if your plans are neither distorted or rotated, you might try simple template matching

berak gravatar imageberak ( 2017-03-08 01:17:23 -0600 )edit

template matching is a good idea.

jsxyhelu gravatar imagejsxyhelu ( 2017-03-08 08:34:29 -0600 )edit
add a comment