查找图像的轮廓Python OpenCV

Question

# import the necessary packages
import decimal
import imutils
import cv2
import numpy as np
import matplotlib.pyplot as plt

# load the image, convert it to grayscale, and blur it slightly
image = cv2.imread("hand.jpg",0)

# threshold the image, then perform a series of erosions +
# dilations to remove any small regions of noise
thresh = cv2.threshold(image, 45, 255, cv2.THRESH_BINARY)[1]
thresh = cv2.erode(thresh, None, iterations=2)
thresh = cv2.dilate(thresh, None, iterations=2)

# find contours in thresholded image, then grab the largest one
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
    cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1]
c = max(cnts, key=cv2.contourArea)
size = len(c);

refer_point = (207,130)

data = np.genfromtxt("data.txt", delimiter=',')

X = data[:,][:,0]

Y = data[:,][:,1]

for i in range(0,size):
    dist1= (((abs(207-X))**2)+((abs(130-Y))**2))**(1.0/2.0)

dist3 = round(dist1,2)
print dist3

plt.plot([dist3])

plt.show()

I am working on the above code. The code was executed perfectly but the contour points of the image were completely wrong. I observed this error after plotting the graph. Help me on this issue.

Answer 1

1) If there is no kernel specified for erosion and dilation, the image does not change. Try them with a kernel.

kernel = np.ones((3,3),np.uint8)
thresh = cv2.erode(thresh, kernel, iterations=1)
thresh = cv2.dilate(thresh, kernel, iterations=1)

2) cnts contains the points ((x,y) tuples) on the contour found. size is only the number of points on that contour. You acquire the size of it and do not process the points, instead you read a data file and plot something totally different. To properly see the contours, try the following after findContours:

# Load the image in color
image = cv2.imread("hand.jpg",cv2.IMREAD_COLOR)
# Draw the contours
cv2.drawContours(image, cnts, -1, (0,255,0), 3)
# Show the image with the contours
cv2.imshow("contours",image)

Then try to relate your data file with the contour points.