Search found 6 matches

by guggy
2017-03-14T06:52:37-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Re: Compare Images

That true, thanks again!
by guggy
2017-03-14T05:38:19-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Re: Compare Images

Sorry my bad,

this is the command :

"convert ${image1} ${image2} +repage -resize "256x256^!" -metric RMSE -format %[distortion] -compare info:"

Its what we talks about before.

the result I get is : 0.1826

version: ImageMagick 6.8.9-9
platform: Ubuntu 16.04
by guggy
2017-03-14T01:23:36-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Re: Compare Images

O.k Thanks ,

I tested more than 1000 images and its works really good.
I have 3 pairs that are not the same but they got score of 1.5 like this pair :

https://goo.gl/5EQezx
https://goo.gl/zJxphY

Do you know maybe why IM not found more differences between them?
by guggy
2017-03-09T07:00:26-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Re: Compare Images

Thanks allot!

This is really helpful,
I just have a small question, why if I ran this command with "x:" at the end instead of "info:" ,

I get a result display that cover almost all the image in red, if the response number say they almost equal?
by guggy
2017-03-09T02:00:46-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Re: Compare Images

O.K thanks, I'll be more clear, What I'm trying to do is to test 10,000 images and check for duplicates, the Images are not in the same size. I tried the 'rmse' command but this one also gave me allot of differences between two images. What is the recommended way to do that? Its also have to be in a ...
by guggy
2017-03-08T10:31:20-07:00
Forum: Developers
Topic: Compare Images
Replies: 10
Views: 19716

Compare Images

1. encoding: compare -channel Gray -dissimilarity-threshold 2.0 -metric phash $image1 $image2 x: for this pair: https://goo.gl/KaOhGJ https://goo.gl/b4hvlZ And get this response : 1.8198 (after trim) Then I run this pair: https://goo.gl/tscR66 https://goo.gl/vByunQ response : 1.95927 (after trim ...