CODEWITHSUNDEEP
pythonopencv
Leopoldo
Leopoldo

Reputation: 845

How to properly use cv2.pointPolyginTest() and cv2.findcontours()?

I would like to use cv2.pointPolygonTest to check if point is placed inside or outside contours, but I cannot figure out why it does not work.

This is the way I get contours:

img_name = a295121c-f893-43f5-8d00-6bfddbc19658.jpg"
im = cv2.imread(img_name)
im_gray = cv2.cvtColor(im, cv2.COLOR_RGB2GRAY)

ret, thresh = cv2.threshold(im_gray, 10, 255, cv2.THRESH_BINARY)
contours, hierarchy = cv2.findContours(thresh, cv2.RETR_LIST, cv2.CHAIN_APPROX_TC89_L1)

And this is how I check if point is inside or outside detected object.

    x1 = cv2.pointPolygonTest(contours[0], (x, y), False)

On this tresholded image it works fine:

test1

x1 is properly calculated when (x,y) is inside object

np.shape(contours) is equal to (1, 241, 1, 2)

However on this image:

enter image description here

all points are calculated as outside

np.shape(contours) is equal to (11,)

I suppose I don't use contours in a proper way, but I cannot figure out which dimension of contours should I pass to cv2.pointPolygonTest()

Upvotes: 3

Views: 7232

Answers (1)

kcc__
kcc__

Reputation: 1648

I wonder if your code to test the point is only this: 

 x1 = cv2.pointPolygonTest(contours[0], (x, y), False)

In this code, the function just test the first contour. Maybe in the second image there are multiple contours for please try to use a loop.

Upvotes: 3

Related Questions