0

I'm trying to create a mask for a color range. I know that I need to convert the image to HSV before passing that into inRange. The next two arguments for inRange are the lower value and upper value for the color we're creating a mask for.

What color space are the lower and upper in?

I've seen HSV and BGR according to answers on here. The documentation does not call out the specific color space to use. For example, this answer (How to define a threshold value to detect only green colour objects in an image :Opencv) says the values should be in HSV but passes in values that seem like BGR (e.g. 255). PS - any tips on tools/tricks to pick a specific range? I've been using Google's color picker

eSquirel
  • 41
  • 11
  • for each channel you choose an upper and a lower bound. You can present a Mat with any color space you like to. By default, images are loaded as BGR. – Micka Nov 27 '20 at 23:26
  • The image colorspace for the image you put in the cv2.inRange() determines the colorspace for the values you use for lower and upper colors. – fmw42 Nov 28 '20 at 01:09

2 Answers2

3

The cv2.inRange() function only checks if array elements lie between the elements of two other arrays.

So if your src array is shaped as BGR or RGB or HSV your Upper and Lower boundaries should be in the same colorspace. Usually for color segmentation the HSV colorspace is more used, but you can try other colorspaces too.

Rômulo Férrer Filho
  • 129
  • 1
  • 5
2

use the following code to find a specific object HSV color value. And set this HSV value in cv2.inRange lower and upper value. It is a opencv program using your device camera and track-bar you can find HSV upper and lower value:

import cv2
import numpy as np


def nothing(x):
    pass


cap = cv2.VideoCapture(0)
cv2.namedWindow("Trackbars")

cv2.createTrackbar("L - H", "Trackbars", 0, 255, nothing)
cv2.createTrackbar("L - S", "Trackbars", 0, 255, nothing)
cv2.createTrackbar("L - V", "Trackbars", 0, 255, nothing)
cv2.createTrackbar("U - H", "Trackbars", 255, 255, nothing)
cv2.createTrackbar("U - S", "Trackbars", 255, 255, nothing)
cv2.createTrackbar("U - V", "Trackbars", 255, 255, nothing)

while True:
    _, frame = cap.read()
    hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)

    l_h = cv2.getTrackbarPos("L - H", "Trackbars")
    l_s = cv2.getTrackbarPos("L - S", "Trackbars")
    l_v = cv2.getTrackbarPos("L - V", "Trackbars")
    u_h = cv2.getTrackbarPos("U - H", "Trackbars")
    u_s = cv2.getTrackbarPos("U - S", "Trackbars")
    u_v = cv2.getTrackbarPos("U - V", "Trackbars")

    lower_blue = np.array([l_h, l_s, l_v])
    upper_blue = np.array([u_h, u_s, u_v])
    mask = cv2.inRange(hsv, lower_blue, upper_blue)
    mask2 = cv2.cvtColor(mask, cv2.COLOR_GRAY2BGR)
    result = cv2.bitwise_and(frame, frame, mask=mask)

    cv2.imshow("frame", frame)
    cv2.imshow("mask", mask)
    cv2.imshow("result", result)

    key = cv2.waitKey(1)
    if key == 27:
        break

cap.release()
cv2.destroyAllWindows()

Output: enter image description here for my object red HSV upper value-[193,188,186], lower value-[135,127,95]

sourab maity
  • 1,025
  • 2
  • 8
  • 16