FastQR Files

This set of files contains a trained cascade for FIP detection both in its original and modified forms. The modified form consists of the original cascade withou the last stages to minimize false negatives in the first stage of QR code detection.

The paremeters and the algorithm are described in: L. F. F. Belussi and N. S. T. Hirata, Fast component-based QR code detection in arbitrarily acquired images, Journal of Mathematical Imaging and Vision, 2012, DOI: 10.1007/s10851-012-0355-x.


The following script runs the detector against a set of test images:

run_QR_code_detector.sh (OpenCV required)

*******************************************************************************
*** Please always cite the following work when using this code:             ***
***                                                                         ***
*** Fast component-based QR code detection in arbitrarily acquired images,  ***
*** L. F. F. Belussi and N. S. T. Hirata,                                   ***
*** Journal of Mathematical Imaging and Vision, 2012,                       ***
*** DOI: 10.1007/s10851-012-0355-x.                                         ***
*******************************************************************************

It is very important to tune the parameters correctly for a new application. They are well described in the paper cited but, if you need assistance in determining the best values for your application, contact me (L. F. F. Belussi) sending me some sample images and I will do my best to help you.


Feel free to contact us, we will be glad to help.

L. F. F. Belussi ( belussi@ime.usp.br / felipe.belussi@gmail.com ) 
N. S. T. Hirata  ( nina@ime.usp.br )