Python Computer Vision -- Finding Duplicate Images With Simple Hashing

In this video, we will discuss how to find duplicate images
using simple mda hashing. The subsequent video will
focus on finding similar images using more complicated hashing.
Let me know what you thought.

Code will be updated to my github :

[sultantech]

*For people who are trying to follow this video in 2020:*
instead of using (from scipy.misc import imread) you need to use (from imageio import imread)

imread was removed in SciPy 1.2.0
Good Luck

[coreyms]

Very cool. I was actually just writing a similar program the other day because I have a lot of duplicate images taking up space on my Google Drive and was trying to find which ones were duplicates that could be deleted to save on space. It's cool to see your approach to this. Very similar to my own.

[yasserabdelsatar]

What is the name of the Simple algorithm or reference research paper of the presented Hashing algorithm ?

[RandomDudeB]

This is a super useful video. Thanks! Honestly, I'm already all-too familiar with regex, so it's nice to see you move onto the next topic. Though those were also good.

[koen]

Thanks! Not used for images in my case but this helped a lot :D

[merchantscape]

can't wait to see the next video!

[georgepatronus]

Can this be used or modified to detect duplicate pages in pdf (non-searchable image) files?

[dennisramadhan]

Hello, this is great. I am wondering how to show the original file name of the images in the result? because I don't want to use 0-1 duplicate, 2- 3 duplicate, etc

[RKUKAR]

Hi. I see you are doing this with magic the gathering cards. Just curious if you know how to get the master images and their attributes?

[solques]

I followed your tutorial however when i ran the duplicates list its always empty. What could be the problem?

[hadishrara]

great one dear but in the first one the imread in SciPy is not supported so u can use import imageio to continue using this code

i use this code without implementing the plotting method and it run successfully
so anyone have a problem in this code just replace part 1 with this code

import hashlib
import imageio
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
# %matplotlib inline
import time
import numpy as np

use the plotting u can put at the
import hashlib
from imageio import

[anushasundaresh]

Sir for imageio aa it's showing no module named..wt Should be done to correct it

[dudelebowski]

Hello I have a work to do and i’m not expert with machine learning so I have 2 Databases each one have images of persons but in a different emotion like the first database has angry faces and the second has happy faces if the same peuple in the first
My task is to find the corespondent image of a person (the input is a person in the first database the output is the same person in the second)
Can you please help me or give some advice

[makhus]

hit error while plotting

IndexError Traceback (most recent call last)
in <module>()
4 plt.title(file_indexes[1]), plt.xticks([]), plt.yticks([])
5
----> 6 plt.subplot(122),
7 + ' duplicate'), plt.xticks([]), plt.yticks([])
8 plt.show()

IndexError: list index out of range

[sbrugby]

how bout that github link, been almost 2 years