minAreaRect(contours) 的 output 是什么

Question

I have recently started working with openCV and and python. I have a project where I am finding contours using findContours . I get roughly around 6-8 contours on which I am looping to get the bounding box that fits the contour.

For that I have used minAreaRect(contours) which gives me rotated rectangle that should fit the contour. Now the output of this command is a list of tuples.

Each tuple looks like this ((81.0, 288.0), (22.0, 10.0), -0.0) I couldnt get any description on what each of that number mean?

I think it might be ((x-coordinate, y-coordinate),(width, height), rotation) .

Answer 1

You are correct. Having a look at OpenCV's (C++) documentation on cv::minAreaRect , we see that a cv::RotatedRect is returned. The full construcor of cv::RotatedRect is:

cv::RotatedRect::RotatedRect(const cv::Point2f& center, const cv::Size2f& size, float angle)    

The description of the corresponding parameters is:

center    The rectangle mass center.
size      Width and height of the rectangle.
angle     The rotation angle in a clockwise direction. When the angle is 0, 90, 180, 270 etc., the rectangle becomes an up-right rectangle.

Obviously, center and size are treated as tuples in the Python API, and all three parameters are returned as a tuple also. So, all in all this fits quite well your assumption.

Hope that helps!

Answer 2

Based on my observations that function gives u back following: (center(x, y), (width, height), angle of rotation) = cv2.minAreaRect(points)