Image PSNR

Problem

Compute the peak signal-to-noise ratio between two same-shaped images in decibels.

Return 10 * log10(max_value^2 / MSE) rounded to 4 decimal places, where MSE is the mean squared pixel error. Return 0.0 when the images are identical.

Test cases

case 1

in: [[[1,2],[3,4]],[[1,1],[3,6]],255]

expected: 47.1617

case 2

in: [[[0]],[[1]],7]

expected: 16.902

case 3

in: [[[1,2],[3,4]],[[1,2],[3,4]],255]

expected: 0

case 4

in: [[[10,20]],[[12,18]],255]

expected: 42.1102

Starter codePython

def image_psnr(a, b, max_value):
    # Your code here
    pass

4 test cases · run your code against them in the editor