Compare Jpg Image Files Using Histogram

• Tweet
• 

        // use progressively more costly method to see if two jpg files are different

#!/usr/bin/python

from PIL import Image
import glob
import os
import sys
import binascii
import md5

def histogram_md5(im):
    m = md5.new()
    h = im.histogram()
    m.update(str(h))
    return m.digest()

if len(sys.argv) < 1:
    print "No file names provided."
    sys.exit()

if len(sys.argv) is not 3:
    print "two files only!"
    sys.exit()

file1 = sys.argv[1]
file2 = sys.argv[2]

print "open ", file1
im1 = Image.open(file1)

print "open ", file2
im2 = Image.open(file2)

print "sizes: ", im1.size, " " , im2.size

if im1.size != im2.size:
    print file1, " and ", file2 , " are different"
    sys.exit()


print "info: " ,im1.info , " " , im2.info

if (im1.info != im2.info):
    print file1, " and ", file2 , " are different"
    sys.exit()


print "mode: " ,im1.mode , " ", im2.mode

if im1.mode != im2.mode:
    print file1, " and ", file2 , " are different"
    sys.exit()

file1_hm5 = histogram_md5(im1)
file2_hm5 = histogram_md5(im2)

print "histogram md5: ", binascii.b2a_hex(file1_hm5), " ", binascii.b2a_hex(file2_hm5)

if file1_hm5 != file2_hm5:
    print file1, " and ", file2 , " are different"
    sys.exit()

f1 = os.path.getsize(file1)
f2 = os.path.getsize(file2)

if f1 != f2:
    print file1, " and ", file2 , " have different sizes."
    print "possibly because they have different meta data."

print "looking at every bit of each file. this can take a while..."

imc1 = list(im1.getdata())
l1 = len(imc1)
imc2 = list(im2.getdata())
l2 = len(imc2)

if l1 != l2:
    print file1, " and ", file2 , " are different"
    sys.exit()

if imc1 != imc2:
    print file1, " and ", file2 , " are different"
    sys.exit()

print file1, " and ", file2 , " contain the same image data."

Tags:
• digest
• histogram
• image
• jpg
• Python