Finding red color in image using Python & OpenCV

Question

I am trying to extract red color from an image. I have code that applies threshold to leave only values from specified range:

img=cv2.imread('img.bmp')
img_hsv=cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
lower_red = np.array([0,50,50]) #example value
upper_red = np.array([10,255,255]) #example value
mask = cv2.inRange(img_hsv, lower_red, upper_red)
img_result = cv2.bitwise_and(img, img, mask=mask)

But, as i checked, red can have Hue value in range, let's say from 0 to 10, as well as in range from 170 to 180. Therefore, i would like to leave values from any of those two ranges. I tried setting threshold from 10 to 170 and using cv2.bitwise_not() function, but then i get all the white color as well. I think the best option would be to create a mask for each range and use them both, so I somehow have to join them together before proceeding.

Is there a way I could join two masks using OpenCV? Or is there some other way I could achieve my goal?

Edit. I came with not much elegant, but working solution:

image_result = np.zeros((image_height,image_width,3),np.uint8)

for i in range(image_height):  #those are set elsewhere
    for j in range(image_width): #those are set elsewhere
        if img_hsv[i][j][1]>=50 \
            and img_hsv[i][j][2]>=50 \
            and (img_hsv[i][j][0] <= 10 or img_hsv[i][j][0]>=170):
            image_result[i][j]=img_hsv[i][j]

It pretty much satisfies my needs, and OpenCV's functions probably do pretty much the same, but if there's a better way to do that(using some dedicated function and writing less code) please share it with me. :)

Red needs either two thresholds to convert zero and above and 180 and below or swap red and blue and then threshold the blue range. — fmw42
– fmw42, Commented Oct 14, 2024 at 17:24

derricw · Accepted Answer · 2015-05-19 18:28:10Z

42

I would just add the masks together, and use np.where to mask the original image.

img=cv2.imread("img.bmp")
img_hsv=cv2.cvtColor(img, cv2.COLOR_BGR2HSV)

# lower mask (0-10)
lower_red = np.array([0,50,50])
upper_red = np.array([10,255,255])
mask0 = cv2.inRange(img_hsv, lower_red, upper_red)

# upper mask (170-180)
lower_red = np.array([170,50,50])
upper_red = np.array([180,255,255])
mask1 = cv2.inRange(img_hsv, lower_red, upper_red)

# join my masks
mask = mask0+mask1

# set my output img to zero everywhere except my mask
output_img = img.copy()
output_img[np.where(mask==0)] = 0

# or your HSV image, which I *believe* is what you want
output_hsv = img_hsv.copy()
output_hsv[np.where(mask==0)] = 0

This should be much faster and much more readable than looping through each pixel of your image.

edited May 19, 2015 at 18:28

answered May 19, 2015 at 18:17

derricw

7,0643 gold badges33 silver badges36 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Yuriy Chernyshov Over a year ago

In case anyone interested. I am working with embedded devices, like Raspberry Pi. Next operation id very heavy for such devices: output_img[np.where(mask==0)] = 0. It can be replaced with much faster one: output_img = cv2.bitwise_and(output_img, output_img, mask= mask)

nathancy · Accepted Answer · 2019-10-02 03:37:28Z

To detect red, you can use a HSV color thresholder script to determine the lower/upper thresholds then cv2.bitwise_and() to obtain the mask. Using this input image,

We get this result and mask

Code

import numpy as np
import cv2

image = cv2.imread('1.jpg')
result = image.copy()
image = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
lower = np.array([155,25,0])
upper = np.array([179,255,255])
mask = cv2.inRange(image, lower, upper)
result = cv2.bitwise_and(result, result, mask=mask)

cv2.imshow('mask', mask)
cv2.imshow('result', result)
cv2.waitKey()

HSV color thresholder script with sliders, remember to change the image file path

import cv2
import sys
import numpy as np

def nothing(x):
    pass

# Load in image
image = cv2.imread('1.jpg')

# Create a window
cv2.namedWindow('image')

# create trackbars for color change
cv2.createTrackbar('HMin','image',0,179,nothing) # Hue is from 0-179 for Opencv
cv2.createTrackbar('SMin','image',0,255,nothing)
cv2.createTrackbar('VMin','image',0,255,nothing)
cv2.createTrackbar('HMax','image',0,179,nothing)
cv2.createTrackbar('SMax','image',0,255,nothing)
cv2.createTrackbar('VMax','image',0,255,nothing)

# Set default value for MAX HSV trackbars.
cv2.setTrackbarPos('HMax', 'image', 179)
cv2.setTrackbarPos('SMax', 'image', 255)
cv2.setTrackbarPos('VMax', 'image', 255)

# Initialize to check if HSV min/max value changes
hMin = sMin = vMin = hMax = sMax = vMax = 0
phMin = psMin = pvMin = phMax = psMax = pvMax = 0

output = image
wait_time = 33

while(1):

    # get current positions of all trackbars
    hMin = cv2.getTrackbarPos('HMin','image')
    sMin = cv2.getTrackbarPos('SMin','image')
    vMin = cv2.getTrackbarPos('VMin','image')

    hMax = cv2.getTrackbarPos('HMax','image')
    sMax = cv2.getTrackbarPos('SMax','image')
    vMax = cv2.getTrackbarPos('VMax','image')

    # Set minimum and max HSV values to display
    lower = np.array([hMin, sMin, vMin])
    upper = np.array([hMax, sMax, vMax])

    # Create HSV Image and threshold into a range.
    hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
    mask = cv2.inRange(hsv, lower, upper)
    output = cv2.bitwise_and(image,image, mask= mask)

    # Print if there is a change in HSV value
    if( (phMin != hMin) | (psMin != sMin) | (pvMin != vMin) | (phMax != hMax) | (psMax != sMax) | (pvMax != vMax) ):
        print("(hMin = %d , sMin = %d, vMin = %d), (hMax = %d , sMax = %d, vMax = %d)" % (hMin , sMin , vMin, hMax, sMax , vMax))
        phMin = hMin
        psMin = sMin
        pvMin = vMin
        phMax = hMax
        psMax = sMax
        pvMax = vMax

    # Display output image
    cv2.imshow('image',output)

    # Wait longer to prevent freeze for videos.
    if cv2.waitKey(wait_time) & 0xFF == ord('q'):
        break

cv2.destroyAllWindows()

Your color choosere is the best.. Now it is so simple to find approprite values...
Your code is not thorough enough in general for red. It needs two threshold ranges for red: 0 and some above 0 and also 180 and some below 180. Your one image is an insufficient test for red in general as it has values only at 180 and below. Some red images can have red colors on the yellow side and also elsewhere in the same image feature have some red on the magenta side.

Michael Dodd · Accepted Answer · 2019-03-19 08:56:06Z

Play with this.

#blurring and smoothin
img1=cv2.imread('marathon.png',1)

hsv = cv2.cvtColor(img1,cv2.COLOR_BGR2HSV)

#lower red
lower_red = np.array([0,50,50])
upper_red = np.array([10,255,255])


#upper red
lower_red2 = np.array([170,50,50])
upper_red2 = np.array([180,255,255])

mask = cv2.inRange(hsv, lower_red, upper_red)
res = cv2.bitwise_and(img1,img1, mask= mask)


mask2 = cv2.inRange(hsv, lower_red2, upper_red2)
res2 = cv2.bitwise_and(img1,img1, mask= mask2)

img3 = res+res2
img4 = cv2.add(res,res2)
img5 = cv2.addWeighted(res,0.5,res2,0.5,0)


kernel = np.ones((15,15),np.float32)/225
smoothed = cv2.filter2D(res,-1,kernel)
smoothed2 = cv2.filter2D(img3,-1,kernel)





cv2.imshow('Original',img1)
cv2.imshow('Averaging',smoothed)
cv2.imshow('mask',mask)
cv2.imshow('res',res)
cv2.imshow('mask2',mask2)
cv2.imshow('res2',res2)
cv2.imshow('res3',img3)
cv2.imshow('res4',img4)
cv2.imshow('res5',img5)
cv2.imshow('smooth2',smoothed2)




cv2.waitKey(0)
cv2.destroyAllWindows()

Rich S · Accepted Answer · 2024-10-11 11:50:59Z

On the HSV colour wheel, the 'H' value for red is 0°, (and 360°).

Your questions states that you're trying to get an 'H' range of -20 to +20, a way to achieve this is to swap the Red/Blue channels when converting to HSV. This way, Red becomes 240° (originally Blue's degrees), and we can simply choose values 220-260 (actually 110-130 in cv2, as the H values are halved)

To swap the Red/Blue channels, just use the cv2 constant COLOR_RGB2HSV rather than COLOR_BGR2HSV

Example code

img=cv2.imread('img.bmp')
img_hsv=cv2.cvtColor(img, cv2.COLOR_RGB2HSV) #### <-- note the change from BGR to RGB
lower_red = np.array([110,50,50]) #example value
upper_red = np.array([130,255,255]) #example value
mask = cv2.inRange(img_hsv, lower_red, upper_red)
img_result = cv2.bitwise_and(img, img, mask=mask)

This requires just one call to the inRange method, and doesn't require any additional logic.

Collectives™ on Stack Overflow

Finding red color in image using Python & OpenCV

4 Answers 4

1 Comment

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

1 Comment

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest