ajgallego / document-image-binarization Goto Github PK

python document-image-binarization/binarize.py -imgpath x.jpg -modelpath MODELS/model_weights_dibco_6_256x256_s96_aug_m205_f64_k5_s2_se3_e200_b32_esp.h5 -w 256 -s 96 -f 64 -k 5 -stride 2 -th 0.5 --demo
Using TensorFlow backend.
Traceback (most recent call last):
File "document-image-binarization/binarize.py", line 12, in
util.init()
File "/home/x/x/x/document-image-binarization/util.py", line 14, in init
np.set_printoptions(threshold=np.nan)
File "/home/x/.local/lib/python2.7/site-packages/numpy/core/arrayprint.py", line 246, in set_printoptions
floatmode, legacy)
File "/home/x/.local/lib/python2.7/site-packages/numpy/core/arrayprint.py", line 93, in _make_options_dict
raise ValueError("threshold must be numeric and non-NAN, try "
ValueError: threshold must be numeric and non-NAN, try sys.maxsize for untruncated representation

I have been unable to run this application.

Hi! Great work!

Could you tell me what the "run_th_test" method is about? I couldn't find it.

I suppose this is the method which makes the reconstruction of the image patches after forwarding them through the network and thresholding the output activation map and maybe compute the f-measure/ pseudo f-measure score using the whole image.

Thanks in advance,
Dayvid

Hi,Aigallego,
I have test DIBCO datasets successfully,so I want to train my own dataset,I create a folder according to your code "train.py",then ran the code ,but it implied no aug_folder,so create one respectively.then I trained, the error is
"nonetype has the attribute shape",
the number of train data is 105,and test data is 256,but I do not have so much test data ,maybe I should sugumentaion before,but I do not know what I should do .could you help me? thanks very much！

Thank you for open-sourcing this!

When I run the script with --test, the best threshold is not set and uses default value -1. How to know the threshold with a pre-trained model?

This looks interesting. Could you please provide information on installation / requirements, or a proper setup.py for release on PyPI?

Hi, I don't see corresponding ground-truth binarized images for Ensiedeln and Salzinnes datasets. How can these datasets be trained?

Hi,

Congrats on the work! I read the paper since I am involved in a mass digitization process right now and I have found that binarization has a huge impact on how well OCR performs. I was trying to put this project to work although it's a bit dated. I set up a virtualenv using Python 3.7.5 and installed the dependencies. Tensorflow 1.x is now deprecated so I had to install it using the 1.15.5 wheel in PyPI. I also tried several versions of Keras but none worked. Is there a way to make it work so I can use the binarize CLI to test my use case?

Thanks!

Can you tell me about the data set path arrangement? (such as the PHI dataset)
Thanks!

you have made a great job on binarization task. I would like to know more regarding the data augmentation technique you told in your paper. Because I can not find it in your code. Thanks a lot

Source code not available yet.

Thank you for the great work in document image binarization~

Is "model_weights_dibco_6_256x256_s96_aug_m205_f64_k5_s2_se3_e200_b32_esp.h5" corresponds to the model trained on the dibco series database except h-dibco2016？

When I run the binatize.py with dataset of h-dibco2016(http://vc.ee.duth.gr/h-dibco2016/benchmark/), the evaluation results are not so good as the conclusions of your publication. The average FM value of ten images is 86.39965. How to reach the FM value of SAE method: 91.65？

After using the pre-trained model, do I need some other operations to achieve the publication results？

By the way，the following is the evaluation results of h-dibco2016 in other pre-trained models:
DIBCO2016_Dataset.csv

What puzzles me is that some evaluation results are better than published ones, and some are worse. But I did not carry out additional training. I just specified the pre-trained model in the binatize.py. As for the evaluation, I use os.popen() to call the command line and execute: BinEvalWeights.exe and DIBCO_metrics.exe. It's a very simple process. Did I neglect some operations when using the pre-trained model?

Thank you again and I am looking forward to your help sincerely~