CODEWITHSUNDEEP
pythonopencv
mbnld
mbnld

Reputation: 61

How to calculate white area with Python and Opencv?

I am triyng calculate the area of the white pixels in this image:

My code :

import cv2
import matplotlib.pyplot as plt
import numpy as np

img = cv2.imread("tepsi.jpg")
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
mask = cv2.inRange(hsv,(21, 10, 15), (30, 255, 255) )
cv2.imshow("orange", mask)
cv2.waitKey()
cv2.destroyAllWindows()

How can I solve this problem? Actually my purpose is to find empty areas of food tray.

Thank you for your answers.

Upvotes: 2

Views: 4406

Answers (2)

mbnld
mbnld

Reputation: 61

enter image description hereThanks for your advices, i found a solution for my problem for above picture. But i will try your advices...

Note: My tray is white area and it is constant for now.

My solution :

import numpy as np
import cv2

image = cv2.imread("petibor_biskuvi.png")
dimensions = image.shape

height= image.shape[0]
width = image.shape[1]
size = height*width

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (7, 7), 0)


_,thresh = cv2.threshold(gray,150,255,cv2.THRESH_BINARY_INV)

cnts, hier = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
size_elements = 0
for cnt in cnts:
    cv2.drawContours(image,cnts, -1, (0, 0, 255), 3)
    size_elements += cv2.contourArea(cnt)
    print(cv2.contourArea(cnt))



cv2.imshow("Image", image)
print("size elements total : ", size_elements)
print("size of pic : ", size)
print("rate of fullness : % ", (size_elements/size)*100)
cv2.waitKey(0)

Upvotes: 4

iGian
iGian

Reputation: 11183

This is one option. Since white in BGR is (255, 255, 255), id suggest you convert the image to a boolean true where each (independent) channel equals to 255:

b, g, r = cv2.split(img)
wb = b == 255
wg = g == 255
wr = r == 255

The value of all of the channels must be 255 (True) for the same pixel, so, use np.bitwise_and:

white_pixels_if_true = np.bitwise_and(wb, np.bitwise_and(wg, wr))

Finally, get the count of the true value and the size of the image and find the percentage of white pixels:

img_size = r.size
white_pixels_count = np.sum(white_pixels_if_true)
white_area_ratio = white_pixels_count / img_size

Given the area of the image you can multiply the area by the white_area_ratio to get the white area.

Upvotes: 1

Related Questions