how OpenCv indexing works?

Question

I am new to OpenCv and numpy and I have a problem figuring out how OpenCv indexing works

so first I am reading an image using this piece of code

import cv2
img = cv2.imread(im_address)
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
print("img size is {}".format(img.shape))

and the output is

(1280, 960)

where 1280 is the height and 960 is the width,but when I try to draw a line in the image using following line I get a vertical line except a horizontal line

cv2.line(img, (98, 0), (98, 1279), (0, 0, 125), 2)
cv2.namedWindow('image', cv2.WINDOW_NORMAL)
cv2.resizeWindow('image', 700, 500)
cv2.imshow("image", img)
cv2.waitKey(0)
cv2.destroyAllWindows()

does OpenCv treat the first index of any coordinate as width? or I am missing something?

From the docs it looks like Points are described as (x,y) NOT (height,width). — wwii, Aug 24 '19 at 18:43

score 1 · Accepted Answer · answered Aug 24 '19 at 18:48

1

Yes, OpenCV treats the first and second coordinates as horizontal and vertical respectively. Specifically, the first coordinate traverses the columns and the second traverses the rows.

answered Aug 24 '19 at 18:48

rayryeng

102,964
22
184
193

tnks, I was a little confused because img type is numpy.ndarray – Saro Aug 24 '19 at 19:09
@salmanrezaie Yeah, this is a result of keeping with their `cv::Mat` type. The nomenclature there is to do width, height. NumPy arrays do height, width. – rayryeng Aug 24 '19 at 19:24

how OpenCv indexing works?

1 Answers1