CODEWITHSUNDEEP
pythonopencv
Bahy Mohamed
Bahy Mohamed

Reputation: 303

how to change the color in grabcut mask based on a class

here's my image:

enter image description here

i have 2 classes dog and truck i also have their bbox coordinates so i want to create a mask for this image so i applied grapcut algorithm here's my code

img=cv2.imread('dog.jpg')
mask=np.zeros(img.shape[:2],np.uint8)

bgModel=np.zeros((1,65),np.float64)
fgModel=np.zeros((1,65),np.float64)
tmpimage=image
masks=[]
for i in recs:
    cv2.grabCut(img,mask,i,bgModel,fgModel,10,cv2.GC_INIT_WITH_RECT)
    mask2=np.where((mask==2)|(mask==0),0,1).astype('uint8')
    masks.append(mask2)
    #img=image*mask2[:,:,np.newaxis]
finalmask=np.zeros(img.shape[:2],np.uint8)
for i in range(len(masks)):
   finalmask=finalmask+masks[i]
plt.imshow(finalmask)
# plt.colorbar()
#plt.xticks([]),plt.yticks([])
plt.show()

here's the final mask:

enter image description here

what i want to do is i want the area of the dog to have different color than area of the truck in this image both are having the same yellow color

Upvotes: 0

Views: 471

Answers (1)

Jeru Luke
Jeru Luke

Reputation: 21203

In your case finalmask is a 2-channel image. The output you get from plt.imshow(finalmask) is a binary image where yellow -> 255 and purple -> 0.

I have created a mask with 3-channels and assigned random colors for each mask.

Code:

# creating 3-channel mask:
finalmask = np.zeros(img.shape, np.uint8)

for i in range(len(masks)):
    # generating random color 
    # plucked from: https://stackoverflow.com/questions/28999287/generate-random-colors-rgb
    color = tuple(np.random.choice(range(256), size=3))

    # create another mask to place color in white regions
    im = np.zeros(img.shape, np.uint8)
    im[masks[i] == 255] = color

    # add with final mask
    finalmask = finalmask + im

Results:

Here sample masks I created:

enter image description here & enter image description here

End result:

enter image description here

Note: If you want fixed color for each mask label, you can create a dictionary and refer from it.

Upvotes: 1

Related Questions