Measure image similarity from a set











up vote
0
down vote

favorite












I'm trying to OCR some info from a document.
The input document can be one of some layouts, but all layouts are known.
I'll use CNN to detect what layout fits to the inputed document.
Once I know what's the document type I will align it to a mask, using image registration, and as I know the needed info position in the mask i'll use some OCR algorith in those cordinates in the inputed image to get my info out already labeled.



The problem is once I know the document type would be great to measure the most similar image from a set of images of the same type to be sure the image registration goes well. I've seen some topics trying to find a way to measure this distance but as all the documents are the same type, and so very similar, im afraid that hashing the image wont be precise, and using the number os points matched in image registration return me false positives.



Do you guys can advice me about the best way to find the most similar image from a set of similar images?
Is there a smarter way to aproach my problem?
I'm currently using python, opencv and tesseract.










share|improve this question






















  • "similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
    – Micka
    Nov 22 at 6:03















up vote
0
down vote

favorite












I'm trying to OCR some info from a document.
The input document can be one of some layouts, but all layouts are known.
I'll use CNN to detect what layout fits to the inputed document.
Once I know what's the document type I will align it to a mask, using image registration, and as I know the needed info position in the mask i'll use some OCR algorith in those cordinates in the inputed image to get my info out already labeled.



The problem is once I know the document type would be great to measure the most similar image from a set of images of the same type to be sure the image registration goes well. I've seen some topics trying to find a way to measure this distance but as all the documents are the same type, and so very similar, im afraid that hashing the image wont be precise, and using the number os points matched in image registration return me false positives.



Do you guys can advice me about the best way to find the most similar image from a set of similar images?
Is there a smarter way to aproach my problem?
I'm currently using python, opencv and tesseract.










share|improve this question






















  • "similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
    – Micka
    Nov 22 at 6:03













up vote
0
down vote

favorite









up vote
0
down vote

favorite











I'm trying to OCR some info from a document.
The input document can be one of some layouts, but all layouts are known.
I'll use CNN to detect what layout fits to the inputed document.
Once I know what's the document type I will align it to a mask, using image registration, and as I know the needed info position in the mask i'll use some OCR algorith in those cordinates in the inputed image to get my info out already labeled.



The problem is once I know the document type would be great to measure the most similar image from a set of images of the same type to be sure the image registration goes well. I've seen some topics trying to find a way to measure this distance but as all the documents are the same type, and so very similar, im afraid that hashing the image wont be precise, and using the number os points matched in image registration return me false positives.



Do you guys can advice me about the best way to find the most similar image from a set of similar images?
Is there a smarter way to aproach my problem?
I'm currently using python, opencv and tesseract.










share|improve this question













I'm trying to OCR some info from a document.
The input document can be one of some layouts, but all layouts are known.
I'll use CNN to detect what layout fits to the inputed document.
Once I know what's the document type I will align it to a mask, using image registration, and as I know the needed info position in the mask i'll use some OCR algorith in those cordinates in the inputed image to get my info out already labeled.



The problem is once I know the document type would be great to measure the most similar image from a set of images of the same type to be sure the image registration goes well. I've seen some topics trying to find a way to measure this distance but as all the documents are the same type, and so very similar, im afraid that hashing the image wont be precise, and using the number os points matched in image registration return me false positives.



Do you guys can advice me about the best way to find the most similar image from a set of similar images?
Is there a smarter way to aproach my problem?
I'm currently using python, opencv and tesseract.







python image opencv similarity measure






share|improve this question













share|improve this question











share|improve this question




share|improve this question










asked Nov 21 at 15:37









Rodrigo Faria

1




1












  • "similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
    – Micka
    Nov 22 at 6:03


















  • "similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
    – Micka
    Nov 22 at 6:03
















"similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
– Micka
Nov 22 at 6:03




"similarity" is not a well defined concept. There are a lot of possibilities to compute some kind of similarity, like SSIM, PSNR, Sift Feature Matching, chamfer matching, bag of words and so on. But none of them will cover everything what a human (with or without donain knowledge) would call "similar".
– Micka
Nov 22 at 6:03

















active

oldest

votes











Your Answer






StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53415549%2fmeasure-image-similarity-from-a-set%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown






























active

oldest

votes













active

oldest

votes









active

oldest

votes






active

oldest

votes
















draft saved

draft discarded




















































Thanks for contributing an answer to Stack Overflow!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


To learn more, see our tips on writing great answers.





Some of your past answers have not been well-received, and you're in danger of being blocked from answering.


Please pay close attention to the following guidance:


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53415549%2fmeasure-image-similarity-from-a-set%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Berounka

Sphinx de Gizeh

Different font size/position of beamer's navigation symbols template's content depending on regular/plain...