ocrd_anybaseocr
ocrd_anybaseocr copied to clipboard
OCR-D
DFKI Layout Detection for OCR-D
On a workspace with >500 pages, running the cropper yields a ``` OSError: [Errno 12] Cannot allocate memory ``` This happens after VSZ (virtual memory) exceeds 32 GB. In contrast,...
In https://github.com/kba/ocrd_anybaseocr/blob/c65f67e3afc740d70acca18dc3d2c2b778d54d18/ocrd_anybaseocr/cli/ocrd_anybaseocr_deskew.py#L159, the rotation is applied without also enlarging the image respectively. This not only looses information (in the corners), but also violates our consistency principle. Subsequent processors will inevitably...
A DFG requirement when scanning is to show a part of the opposite page. On some pages this tends to be a problem, since `anybaseocr-crop` does not crop the text...
Hi Martin, we met shortly in Bonn. You have explained the de-warping which was very interesting for me. I have tried out a bit - and after some environment issues,...
I'll get `ValueError: tile cannot extend outside image` Images (850 MB): https://digi.ub.uni-heidelberg.de/diglitData/v/testset-5-zeitschr-ca-1870.zip ``` File "/dwork/ocrd-schroot-ubuntu-eoan/usr/local/ocrd_all/venv/lib/python3.6/site-packages/click/core.py", line 610, in invoke return callback(*args, **kwargs) File "/dwork/ocrd-schroot-ubuntu-eoan/usr/local/ocrd_all/venv/lib/python3.6/site-packages/ocrd/cli/process.py", line 27, in process_cli run_tasks(mets, log_level,...
Perhaps a problem only in combination with ocrd-sbb-binarize(?) ```...> ocrd-sbb-binarize -I OCR-D-IMG -O OCR-D-BIN -P model /usr/local/ocrd_models/sbb/binarization/models (venv) jb@pers109:~/literatur_schoenen_wissenschaften1780a> ocrd-anybaseocr-crop -I OCR-D-BIN -O OCR-D-CROP 16:04:18.388 INFO OcrdAnybaseocrCropper - INPUT FILE...
Once I got the block segmentation to actually run, I was puzzled over the extremely bad results of the provided model. Here's how I gradually worked to isolate the problem....
In e941321a507ce9f4f6d6416117e441124605748a it seems 3 non-text classes arrived: ImageRegion, TableRegion and GraphicsRegion. However, the `Config.NUM_CLASSES` remained the same, and equally the provided `block_segmentation_weights.h5` still have only 1+14 classes: ``` >>>...
The way in which the trained pixel classifier for text-image segmentation is integrated here makes these predictions completely unusable: - original:  - results: | *image part* | *text part*...
The `--help` of `ocrd-anybaseocr-tiseg` states a _default wiring_ of `['OCR-D-IMG-CROP'] -> ['OCR-D-SEG-TISEG']`. ``` root@38fa7aad0b43:/data/ocrd_workspace# ocrd-anybaseocr-tiseg --help Using TensorFlow backend. Usage: ocrd-anybaseocr-tiseg [OPTIONS] separate text and non-text part with anyBaseOCR Options:...
