2d Boolean Array To Image

Question

I have an 2d array consists of boolean values. For mode of fromarray both mode='1' and mode='L' not working properly. Both of them returns an black image while it also should have white values for where indexes point to True. How can I get the proper image?

    import numpy as np
    from PIL import Image

    maskArr = np.array(arr) * 255

    print(maskArr.shape) # Returns (600, 800)
    print(np.where(maskArr == True)) # Returns a tuple of index of True values.
    maskImg1 = Image.fromarray(maskArr, mode='1')
    maskImg2 = Image.fromarray(maskArr, mode='L')
    maskImg1.save("./maskImg.jpg")
    maskImg2.save("./maskImg2.jpg")

maskImg1

maskImg2

Answer 1

You want to ensure that maskArr has dtype uint8 , and that you pass mode='L' to PIL:

maskArr = np.array(arr, dtype=np.uint8) * 255

maskImg =  Image.fromarray(maskArr, mode='L')
maskImg.save("./maskImg.jpg")

Currently, your maskArr has dtype int64 , so PIL doesn't know to treat 255 as white and 0 as black.

When I run this code with arr = [[True, False]*100,[False, True]*100]*100 , I get this output: