I am working on text localization in scanned documents. I am using morphological operations to arrive at the text regions. But the documents also contain QR and bar codes which are also getting detected. How can I create a filter which can identify QR and bar code regions? I have tried pyzbar library, it could only detect when the code is clear. My dataset is quite noisy. I have also tried the approach given in https://www.pyimagesearch.com/2014/11/24/detecting-barcodes-images-python-opencv
This gives me many false positives and also this approach works only if the image has only one code(QR/bar) in it.