opencv cv2.drawcontours contour argument: contour vs [contour] [duplicate]

Question

my input :

my code :

# import the necessary packages
import numpy as np
import argparse
import imutils
import cv2


print("Your OpenCV version: {}".format(cv2.__version__))
print("Are you using OpenCV 2.X? {}".format(imutils.is_cv2()))
print("Are you using OpenCV 3.X? {}".format(imutils.is_cv3()))
print("Are you using OpenCV 4.X? {}".format(imutils.is_cv4()))


# load the image
image_orig = cv2.imread('box1.jpg')
height_orig, width_orig = image_orig.shape[:2]

# cv2.imshow('image_orig' , image_orig)
# cv2.waitKey(0)
cv2.imwrite('image_orig.png', image_orig)


# copy of original image
image_to_process = image_orig.copy()
    
image_gray = cv2.cvtColor(image_to_process, cv2.COLOR_BGR2GRAY)
image_gray = cv2.GaussianBlur(image_gray, (5, 5), 0)

    
# cv2.imshow('image_gray' , image_gray)
# cv2.waitKey(0)
cv2.imwrite('image_gray.png', image_gray)
    
# perform edge detection, then perform a dilation + erosion to close gaps in between object edges
image_edged = cv2.Canny(image_gray, 50, 100)
image_edged = cv2.dilate(image_edged, None, iterations=1)
image_edged = cv2.erode(image_edged, None, iterations=1)
    
# cv2.imshow('image_edged' , image_edged)
# cv2.waitKey(0)
cv2.imwrite('image_edged.png', image_edged)
    
 
# find contours in the edge map
cnts = cv2.findContours(image_edged, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)

cnts = cnts[1] if imutils.is_cv2() else cnts[0] #### #### using CV4
    
print(len(cnts))


 
# loop over the contours individually

# prints contours in red color  thickness = 1
image_contours = image_to_process.copy()
for c in cnts:
    
    cv2.drawContours(image_contours, c , -1 ,(0,0,255),1)

cv2.imshow('1', image_contours)
cv2.waitKey(0)



# prints contours in red color  thickness = 10
image_contours = image_to_process.copy()
for c in cnts:
    
    cv2.drawContours(image_contours, c , -1 ,(0,0,255),10)

cv2.imshow('10', image_contours)
cv2.waitKey(0)

# prints contours in red color  thickness = cv2.FILLED
image_contours = image_to_process.copy()
for c in cnts:
    
    cv2.drawContours(image_contours, c , -1 ,(0,0,255), cv2.FILLED)

cv2.imshow('filled', image_contours)
cv2.waitKey(0)
    
# prints contours in red color  thickness = cv2.FILLED
image_contours = image_to_process.copy()
for c in cnts:
    
    cv2.drawContours(image_contours, [c] , -1 ,(0,0,255), cv2.FILLED)

cv2.imshow('filled_[]', image_contours)
cv2.waitKey(0) 

output:

Any particular reason:

cv2.drawContours(image_contours, [c] , -1 ,(0,0,255), cv2.FILLED)

works while

cv2.drawContours(image_contours, c , -1 ,(0,0,255), cv2.FILLED)

doesn't , and why

cv2.drawContours(image_contours, c , -1 ,(0,0,255), 10) do work ?

Which one is the right way to code it ? Am I missing anything ?

Answer 1

cv.drawContours() always expects a list of contours.

cv.findContours() always returns a list of contours.

They mirror each other.

There is no findContour and no drawContour.

Always pass a list of contours, even if it's a list of one contour.

If you pass it just a contour, which is an array of points, it'll interpret that as a list of contours (of course), where each contour is a list of one point. So it'll draw the individual points of your single contour.

Try giving it a simplified contour, so that the points of it aren't so densely spaced. Or give it a decimated contour, something like contour[::10]. You'll see that it draws those points.

im = cv.imread("19nHf6z3.jpg")
gray = cv.cvtColor(im, cv.COLOR_BGR2GRAY)

(_, binarized) = cv.threshold(gray, 0, 255, cv.THRESH_BINARY_INV | cv.THRESH_OTSU)

(contours, _) = cv.findContours(binarized, cv.RETR_EXTERNAL, cv.CHAIN_APPROX_NONE)

canvas = (im >> 1) # the original but dimmed
for contour in contours:
    contour_decimated = contour[::10]
    cv.drawContours(canvas, contour_decimated, -1, color=(0, 0, 255), thickness=5)
    # expecting a list, it'll intepret each point as a contour