question about mask of Kvasir dataset

[Qinger27]

Thanks for your great work first! It is a great research!
When I check the Kvasir data, I find each mask has more than two different pixel values. I randomly choose a mask and print different pixel values, and it is a list of [0 1 2 3 4 5 6 7 248 249 250 251 253 254 255]. And then I visualize the different pixels values. The 0 represents bachground and 255 represents polyps. The left is around the polyps, i.e. in the boundary of polyps.
Could you tell me how you process this situation?
Can I change these mask into binary image by sample threshholding?
Really look forward to your reply!
Thanks again!

[GewelsJI]

hi @Qinger27
you need to binarise them using Image.open(your_file_name).convert('1')

Thanks