I work on a project for my thesis which involves optimizing the QR-code recognition rate in an robot application where zbar is currently used. I am familiar with the concepts, developed also my own algorithm in matlab to easily play around with preprocessing the images and try to check general possibilities to improve the recognition rate.
My problem is that I really have difficulties to see how the qr-detection algorithm in zbar works in detail. Although it is fairly well documented I miss the big-picture of this algorithm. I don’t know how to browse code well outside of the visual studio world and get I get a little lost in all these point/line comparisons, calculations and grouping of lines and result if I just look at the files top-down.
So I would be glad if somebody could help me with the following things:
Could you point me to some information where the algorithm you are using is described, a short documentation, some little help about the file structure/contents/starting functions, or maybe even some papers which cover the algorithm? (Am i right that there is nothing in the wiki?)
Are there any simple settings existing to favor recognition rate over speed (like scanning every line or changing thresholds) that I could easily try before trying to change the code?
If I manage to compile the code, are there some simple, known tweaks to improve the recognition rate, if it means that it will be slower?
I would appreciate if somebody could give me a few hints.
Of course I will also share my findings here!
Thanks!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I found the paper about QR Code Recognition. in that paper, the detail description of QR Code scanning procedure exists but i'm not sure if zbar's algorithm is same as that. it's been a long time that you post, but if you'd like to view the paper, i will help you. just give me a e-mail.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I work on a project for my thesis which involves optimizing the QR-code recognition rate in an robot application where zbar is currently used. I am familiar with the concepts, developed also my own algorithm in matlab to easily play around with preprocessing the images and try to check general possibilities to improve the recognition rate.
My problem is that I really have difficulties to see how the qr-detection algorithm in zbar works in detail. Although it is fairly well documented I miss the big-picture of this algorithm. I don’t know how to browse code well outside of the visual studio world and get I get a little lost in all these point/line comparisons, calculations and grouping of lines and result if I just look at the files top-down.
So I would be glad if somebody could help me with the following things:
Could you point me to some information where the algorithm you are using is described, a short documentation, some little help about the file structure/contents/starting functions, or maybe even some papers which cover the algorithm? (Am i right that there is nothing in the wiki?)
Are there any simple settings existing to favor recognition rate over speed (like scanning every line or changing thresholds) that I could easily try before trying to change the code?
If I manage to compile the code, are there some simple, known tweaks to improve the recognition rate, if it means that it will be slower?
I would appreciate if somebody could give me a few hints.
Of course I will also share my findings here!
Thanks!
I found the paper about QR Code Recognition. in that paper, the detail description of QR Code scanning procedure exists but i'm not sure if zbar's algorithm is same as that. it's been a long time that you post, but if you'd like to view the paper, i will help you. just give me a e-mail.