How to detect drawn shape

Question

What I am trying to do is let the player draw with his mouse for example and if he tries to draw a circle for example I can figure it out and replace the inaccurate shape of the player with the sprite of a perfect circle for example. I thought about using larp() and calculate the distance of each point to a point supposed center of the circle. If the distances are relatively close then it will be a circle. But I'm sure there is a better way to do this ? Thank you for your help :)

Comments

You can approach things two ways I think. The first thing is just to smooth out the line drawn by the user. I think there are multiple ways to do that, but I found this interesting article on the subject: http://merowing.info/2012/04/drawing-smooth-lines-with-cocos2d-ios-inspired-by-paper/

The other way, is to provide drawing tools, like circle, rectangle, etc.. The user will draw what he thinks is a circle, and you can match it to the closest perfect circle.

I think these needs to be separate functions otherwise you could come into a situation where the user is drawing something, and your software thinks its close to a circle for example, so it matches it to a perfect circle and it's not what the user wants.

---

I've experimented with the same idea and managed few working examples

My thinking was...

CPU version:
1) prepared set of patterns:
- each pattern has same size (200x200px for ex.)
- pattern is collection of circles

2) while player draws you sample his strokes and scaled it to the same size as pattern images

3) compare given samples with predefined pattern -> futher from the center of a point, the less accurate (you can set for example 5px radius)

4) ... do a lot of optimalizations and thinking :)

GPU version:
1) prepared set of patterns are images with thick lines that fades from center to sides
- stronger color = more acurate to the pattern

2) player draws to texture as well

3) render player's texture with fragment shader and pattern as mask
- for example if pixel is on the mask, save it's color to x channel (higher number = closer to perfect shape)
- if not od the mask, use y channel

4) check the resulting image -> sum up X and Y channels to see how accurate it was