Merge branch 'integrate-training-from-sbb_pixelwise_segmentation' of https://github.com/qurator-spk/eynollah into integrate-training-from-sbb_pixelwise_segmentation

This commit is contained in:
kba 2025-10-01 18:02:09 +02:00
commit 95bb5908bb
15 changed files with 78 additions and 287 deletions

View file

@ -22,7 +22,7 @@ Added:
Fixed:
* allow empty imports for optional dependencies
* avoid Numpy warnings (empty slices etc)
* avoid Numpy warnings (empty slices etc.)
* remove deprecated Numpy types
* binarization CLI: make `dir_in` usable again

View file

@ -11,23 +11,24 @@
![](https://user-images.githubusercontent.com/952378/102350683-8a74db80-3fa5-11eb-8c7e-f743f7d6eae2.jpg)
## Features
* Support for up to 10 segmentation classes:
* Support for 10 distinct segmentation classes:
* background, [page border](https://ocr-d.de/en/gt-guidelines/trans/lyRand.html), [text region](https://ocr-d.de/en/gt-guidelines/trans/lytextregion.html#textregionen__textregion_), [text line](https://ocr-d.de/en/gt-guidelines/pagexml/pagecontent_xsd_Complex_Type_pc_TextLineType.html), [header](https://ocr-d.de/en/gt-guidelines/trans/lyUeberschrift.html), [image](https://ocr-d.de/en/gt-guidelines/trans/lyBildbereiche.html), [separator](https://ocr-d.de/en/gt-guidelines/trans/lySeparatoren.html), [marginalia](https://ocr-d.de/en/gt-guidelines/trans/lyMarginalie.html), [initial](https://ocr-d.de/en/gt-guidelines/trans/lyInitiale.html), [table](https://ocr-d.de/en/gt-guidelines/trans/lyTabellen.html)
* Support for various image optimization operations:
* cropping (border detection), binarization, deskewing, dewarping, scaling, enhancing, resizing
* Text line segmentation to bounding boxes or polygons (contours) including for curved lines and vertical text
* Detection of reading order (left-to-right or right-to-left)
* Textline segmentation to bounding boxes or polygons (contours) including for curved lines and vertical text
* Text recognition (OCR) using either CNN-RNN or Transformer models
* Detection of reading order (left-to-right or right-to-left) using either heuristics or trainable models
* Output in [PAGE-XML](https://github.com/PRImA-Research-Lab/PAGE-XML)
* [OCR-D](https://github.com/qurator-spk/eynollah#use-as-ocr-d-processor) interface
:warning: Development is currently focused on achieving the best possible quality of results for a wide variety of
historical documents and therefore processing can be very slow. We aim to improve this, but contributions are welcome.
:warning: Development is focused on achieving the best quality of results for a wide variety of historical
documents and therefore processing can be very slow. We aim to improve this, but contributions are welcome.
## Installation
Python `3.8-3.11` with Tensorflow `<2.13` on Linux are currently supported.
For (limited) GPU support the CUDA toolkit needs to be installed.
For (limited) GPU support the CUDA toolkit needs to be installed. A known working config is CUDA `11` with cuDNN `8.6`.
You can either install from PyPI
@ -56,26 +57,27 @@ make install EXTRAS=OCR
Pretrained models can be downloaded from [zenodo](https://zenodo.org/records/17194824) or [huggingface](https://huggingface.co/SBB?search_models=eynollah).
For documentation on methods and models, have a look at [`models.md`](https://github.com/qurator-spk/eynollah/tree/main/docs/models.md).
For documentation on models, have a look at [`models.md`](https://github.com/qurator-spk/eynollah/tree/main/docs/models.md).
Model cards are also provided for our trained models.
## Training
In case you want to train your own model with Eynollah, have see the
In case you want to train your own model with Eynollah, see the
documentation in [`train.md`](https://github.com/qurator-spk/eynollah/tree/main/docs/train.md) and use the
tools in the [`train` folder](https://github.com/qurator-spk/eynollah/tree/main/train).
## Usage
Eynollah supports five use cases: layout analysis (segmentation), binarization,
image enhancement, text recognition (OCR), and (trainable) reading order detection.
image enhancement, text recognition (OCR), and reading order detection.
### Layout Analysis
The layout analysis module is responsible for detecting layouts, identifying text lines, and determining reading order
using both heuristic methods or a machine-based reading order detection model.
The layout analysis module is responsible for detecting layout elements, identifying text lines, and determining reading
order using either heuristic methods or a [pretrained reading order detection model](https://github.com/qurator-spk/eynollah#machine-based-reading-order).
Note that there are currently two supported ways for reading order detection: either as part of layout analysis based
on image input, or, currently under development, for given layout analysis results based on PAGE-XML data as input.
Reading order detection can be performed either as part of layout analysis based on image input, or, currently under
development, based on pre-existing layout analysis results in PAGE-XML format as input.
The command-line interface for layout analysis can be called like this:
@ -108,15 +110,15 @@ The following options can be used to further configure the processing:
| `-sp <directory>` | save cropped page image to this directory |
| `-sa <directory>` | save all (plot, enhanced/binary image, layout) to this directory |
If no option is set, the tool performs layout detection of main regions (background, text, images, separators
If no further option is set, the tool performs layout detection of main regions (background, text, images, separators
and marginals).
The best output quality is produced when RGB images are used as input rather than greyscale or binarized images.
The best output quality is achieved when RGB images are used as input rather than greyscale or binarized images.
### Binarization
The binarization module performs document image binarization using pretrained pixelwise segmentation models.
The command-line interface for binarization of single image can be called like this:
The command-line interface for binarization can be called like this:
```sh
eynollah binarization \
@ -127,16 +129,16 @@ eynollah binarization \
### OCR
The OCR module performs text recognition from images using two main families of pretrained models: CNN-RNNbased OCR and Transformer-based OCR.
The OCR module performs text recognition using either a CNN-RNN model or a Transformer model.
The command-line interface for ocr can be called like this:
The command-line interface for OCR can be called like this:
```sh
eynollah ocr \
-i <single image file> | -di <directory containing image files> \
-dx <directory of xmls> \
-o <output directory> \
-m <path to directory containing model files> | --model_name <path to specific model> \
-m <directory containing model files> | --model_name <path to specific model> \
```
### Machine-based-reading-order
@ -172,22 +174,20 @@ If the input file group is PAGE-XML (from a previous OCR-D workflow step), Eynol
(because some other preprocessing step was in effect like `denoised`), then
the output PAGE-XML will be based on that as new top-level (`@imageFilename`)
ocrd-eynollah-segment -I OCR-D-XYZ -O OCR-D-SEG -P models eynollah_layout_v0_5_0
ocrd-eynollah-segment -I OCR-D-XYZ -O OCR-D-SEG -P models eynollah_layout_v0_5_0
Still, in general, it makes more sense to add other workflow steps **after** Eynollah.
In general, it makes more sense to add other workflow steps **after** Eynollah.
There is also an OCR-D processor for the binarization:
There is also an OCR-D processor for binarization:
ocrd-sbb-binarize -I OCR-D-IMG -O OCR-D-BIN -P models default-2021-03-09
#### Additional documentation
Please check the [wiki](https://github.com/qurator-spk/eynollah/wiki).
Additional documentation is available in the [docs](https://github.com/qurator-spk/eynollah/tree/main/docs) directory.
## How to cite
If you find this tool useful in your work, please consider citing our paper:
```bibtex
@inproceedings{hip23rezanezhad,
title = {Document Layout Analysis with Deep Learning and Heuristics},

View file

@ -4886,9 +4886,9 @@ class Eynollah:
textline_mask_tot_ea_org[img_revised_tab==drop_label_in_full_layout] = 0
text_only = ((img_revised_tab[:, :] == 1)) * 1
text_only = (img_revised_tab[:, :] == 1) * 1
if np.abs(slope_deskew) >= SLOPE_THRESHOLD:
text_only_d = ((text_regions_p_1_n[:, :] == 1)) * 1
text_only_d = (text_regions_p_1_n[:, :] == 1) * 1
#print("text region early 2 in %.1fs", time.time() - t0)
###min_con_area = 0.000005

View file

@ -12,7 +12,7 @@ from .utils import crop_image_inside_box
from .utils.rotate import rotate_image_different
from .utils.resize import resize_image
class EynollahPlotter():
class EynollahPlotter:
"""
Class collecting all the plotting and image writing methods
"""

View file

@ -138,8 +138,7 @@ def return_x_start_end_mothers_childs_and_type_of_reading_order(
min_ys=np.min(y_sep)
max_ys=np.max(y_sep)
y_mains=[]
y_mains.append(min_ys)
y_mains= [min_ys]
y_mains_sep_ohne_grenzen=[]
for ii in range(len(new_main_sep_y)):
@ -493,8 +492,7 @@ def find_num_col(regions_without_separators, num_col_classifier, tables, multipl
# print(forest[np.argmin(z[forest]) ] )
if not isNaN(forest[np.argmin(z[forest])]):
peaks_neg_true.append(forest[np.argmin(z[forest])])
forest = []
forest.append(peaks_neg_fin[i + 1])
forest = [peaks_neg_fin[i + 1]]
if i == (len(peaks_neg_fin) - 1):
# print(print(forest[np.argmin(z[forest]) ] ))
if not isNaN(forest[np.argmin(z[forest])]):
@ -662,8 +660,7 @@ def find_num_col_only_image(regions_without_separators, multiplier=3.8):
# print(forest[np.argmin(z[forest]) ] )
if not isNaN(forest[np.argmin(z[forest])]):
peaks_neg_true.append(forest[np.argmin(z[forest])])
forest = []
forest.append(peaks_neg_fin[i + 1])
forest = [peaks_neg_fin[i + 1]]
if i == (len(peaks_neg_fin) - 1):
# print(print(forest[np.argmin(z[forest]) ] ))
if not isNaN(forest[np.argmin(z[forest])]):
@ -1211,7 +1208,7 @@ def order_of_regions(textline_mask, contours_main, contours_header, y_ref):
##plt.plot(z)
##plt.show()
if contours_main != None:
if contours_main is not None:
areas_main = np.array([cv2.contourArea(contours_main[j]) for j in range(len(contours_main))])
M_main = [cv2.moments(contours_main[j]) for j in range(len(contours_main))]
cx_main = [(M_main[j]["m10"] / (M_main[j]["m00"] + 1e-32)) for j in range(len(M_main))]
@ -1222,7 +1219,7 @@ def order_of_regions(textline_mask, contours_main, contours_header, y_ref):
y_min_main = np.array([np.min(contours_main[j][:, 0, 1]) for j in range(len(contours_main))])
y_max_main = np.array([np.max(contours_main[j][:, 0, 1]) for j in range(len(contours_main))])
if len(contours_header) != None:
if len(contours_header) is not None:
areas_header = np.array([cv2.contourArea(contours_header[j]) for j in range(len(contours_header))])
M_header = [cv2.moments(contours_header[j]) for j in range(len(contours_header))]
cx_header = [(M_header[j]["m10"] / (M_header[j]["m00"] + 1e-32)) for j in range(len(M_header))]
@ -1235,17 +1232,16 @@ def order_of_regions(textline_mask, contours_main, contours_header, y_ref):
y_max_header = np.array([np.max(contours_header[j][:, 0, 1]) for j in range(len(contours_header))])
# print(cy_main,'mainy')
peaks_neg_new = []
peaks_neg_new.append(0 + y_ref)
peaks_neg_new = [0 + y_ref]
for iii in range(len(peaks_neg)):
peaks_neg_new.append(peaks_neg[iii] + y_ref)
peaks_neg_new.append(textline_mask.shape[0] + y_ref)
if len(cy_main) > 0 and np.max(cy_main) > np.max(peaks_neg_new):
cy_main = np.array(cy_main) * (np.max(peaks_neg_new) / np.max(cy_main)) - 10
if contours_main != None:
if contours_main is not None:
indexer_main = np.arange(len(contours_main))
if contours_main != None:
if contours_main is not None:
len_main = len(contours_main)
else:
len_main = 0
@ -1271,11 +1267,11 @@ def order_of_regions(textline_mask, contours_main, contours_header, y_ref):
top = peaks_neg_new[i]
down = peaks_neg_new[i + 1]
indexes_in = matrix_of_orders[:, 0][(matrix_of_orders[:, 3] >= top) &
((matrix_of_orders[:, 3] < down))]
(matrix_of_orders[:, 3] < down)]
cxs_in = matrix_of_orders[:, 2][(matrix_of_orders[:, 3] >= top) &
((matrix_of_orders[:, 3] < down))]
(matrix_of_orders[:, 3] < down)]
cys_in = matrix_of_orders[:, 3][(matrix_of_orders[:, 3] >= top) &
((matrix_of_orders[:, 3] < down))]
(matrix_of_orders[:, 3] < down)]
types_of_text = matrix_of_orders[:, 1][(matrix_of_orders[:, 3] >= top) &
(matrix_of_orders[:, 3] < down)]
index_types_of_text = matrix_of_orders[:, 4][(matrix_of_orders[:, 3] >= top) &
@ -1404,8 +1400,7 @@ def combine_hor_lines_and_delete_cross_points_and_get_lines_features_back_new(
return img_p_in[:,:,0], special_separators
def return_points_with_boundies(peaks_neg_fin, first_point, last_point):
peaks_neg_tot = []
peaks_neg_tot.append(first_point)
peaks_neg_tot = [first_point]
for ii in range(len(peaks_neg_fin)):
peaks_neg_tot.append(peaks_neg_fin[ii])
peaks_neg_tot.append(last_point)
@ -1413,7 +1408,7 @@ def return_points_with_boundies(peaks_neg_fin, first_point, last_point):
def find_number_of_columns_in_document(region_pre_p, num_col_classifier, tables, pixel_lines, contours_h=None):
t_ins_c0 = time.time()
separators_closeup=( (region_pre_p[:,:,:]==pixel_lines))*1
separators_closeup= (region_pre_p[:, :, :] == pixel_lines) * 1
separators_closeup[0:110,:,:]=0
separators_closeup[separators_closeup.shape[0]-150:,:,:]=0
@ -1452,7 +1447,7 @@ def find_number_of_columns_in_document(region_pre_p, num_col_classifier, tables,
gray = cv2.bitwise_not(separators_closeup_n_binary)
gray=gray.astype(np.uint8)
bw = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, \
bw = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
cv2.THRESH_BINARY, 15, -2)
horizontal = np.copy(bw)
vertical = np.copy(bw)
@ -1588,8 +1583,7 @@ def find_number_of_columns_in_document(region_pre_p, num_col_classifier, tables,
args_cy_splitter=np.argsort(cy_main_splitters)
cy_main_splitters_sort=cy_main_splitters[args_cy_splitter]
splitter_y_new=[]
splitter_y_new.append(0)
splitter_y_new= [0]
for i in range(len(cy_main_splitters_sort)):
splitter_y_new.append( cy_main_splitters_sort[i] )
splitter_y_new.append(region_pre_p.shape[0])
@ -1663,8 +1657,7 @@ def return_boxes_of_images_by_order_of_reading_new(
num_col, peaks_neg_fin = find_num_col(
regions_without_separators[int(splitter_y_new[i]):int(splitter_y_new[i+1]),:],
num_col_classifier, tables, multiplier=3.)
peaks_neg_fin_early=[]
peaks_neg_fin_early.append(0)
peaks_neg_fin_early= [0]
#print(peaks_neg_fin,'peaks_neg_fin')
for p_n in peaks_neg_fin:
peaks_neg_fin_early.append(p_n)

View file

@ -239,8 +239,7 @@ def do_back_rotation_and_get_cnt_back(contour_par, index_r_con, img, slope_first
cont_int, _ = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
if len(cont_int)==0:
cont_int = []
cont_int.append(contour_par)
cont_int = [contour_par]
confidence_contour = 0
else:
cont_int[0][:, 0, 0] = cont_int[0][:, 0, 0] + np.abs(img_copy.shape[1] - img.shape[1])

View file

@ -3,7 +3,7 @@ from collections import Counter
REGION_ID_TEMPLATE = 'region_%04d'
LINE_ID_TEMPLATE = 'region_%04d_line_%04d'
class EynollahIdCounter():
class EynollahIdCounter:
def __init__(self, region_idx=0, line_idx=0):
self._counter = Counter()

View file

@ -76,7 +76,7 @@ def get_marginals(text_with_lines, text_regions, num_col, slope_deskew, light_ve
peaks, _ = find_peaks(text_with_lines_y_rev, height=0)
peaks=np.array(peaks)
peaks=peaks[(peaks>first_nonzero) & ((peaks<last_nonzero))]
peaks=peaks[(peaks>first_nonzero) & (peaks < last_nonzero)]
peaks=peaks[region_sum_0[peaks]<min_textline_thickness ]

View file

@ -1174,8 +1174,7 @@ def separate_lines_new_inside_tiles(img_path, thetha):
if diff_peaks[i] > cut_off:
if not np.isnan(forest[np.argmin(z[forest])]):
peaks_neg_true.append(forest[np.argmin(z[forest])])
forest = []
forest.append(peaks_neg[i + 1])
forest = [peaks_neg[i + 1]]
if i == (len(peaks_neg) - 1):
if not np.isnan(forest[np.argmin(z[forest])]):
peaks_neg_true.append(forest[np.argmin(z[forest])])
@ -1195,8 +1194,7 @@ def separate_lines_new_inside_tiles(img_path, thetha):
if diff_peaks_pos[i] > cut_off:
if not np.isnan(forest[np.argmax(z[forest])]):
peaks_pos_true.append(forest[np.argmax(z[forest])])
forest = []
forest.append(peaks[i + 1])
forest = [peaks[i + 1]]
if i == (len(peaks) - 1):
if not np.isnan(forest[np.argmax(z[forest])]):
peaks_pos_true.append(forest[np.argmax(z[forest])])
@ -1430,9 +1428,9 @@ def separate_lines_new2(img_path, thetha, num_col, slope_region, logger=None, pl
img_int = np.zeros((img_xline.shape[0], img_xline.shape[1]))
img_int[:, :] = img_xline[:, :] # img_patch_org[:,:,0]
img_resized = np.zeros((int(img_int.shape[0] * (1.2)), int(img_int.shape[1] * (3))))
img_resized[int(img_int.shape[0] * (0.1)) : int(img_int.shape[0] * (0.1)) + img_int.shape[0],
int(img_int.shape[1] * (1.0)) : int(img_int.shape[1] * (1.0)) + img_int.shape[1]] = img_int[:, :]
img_resized = np.zeros((int(img_int.shape[0] * 1.2), int(img_int.shape[1] * 3)))
img_resized[int(img_int.shape[0] * 0.1): int(img_int.shape[0] * 0.1) + img_int.shape[0],
int(img_int.shape[1] * 1.0): int(img_int.shape[1] * 1.0) + img_int.shape[1]] = img_int[:, :]
# plt.imshow(img_xline)
# plt.show()
img_line_rotated = rotate_image(img_resized, slopes_tile_wise[i])
@ -1444,8 +1442,8 @@ def separate_lines_new2(img_path, thetha, num_col, slope_region, logger=None, pl
img_patch_separated_returned[:, :][img_patch_separated_returned[:, :] != 0] = 1
img_patch_separated_returned_true_size = img_patch_separated_returned[
int(img_int.shape[0] * (0.1)) : int(img_int.shape[0] * (0.1)) + img_int.shape[0],
int(img_int.shape[1] * (1.0)) : int(img_int.shape[1] * (1.0)) + img_int.shape[1]]
int(img_int.shape[0] * 0.1): int(img_int.shape[0] * 0.1) + img_int.shape[0],
int(img_int.shape[1] * 1.0): int(img_int.shape[1] * 1.0) + img_int.shape[1]]
img_patch_separated_returned_true_size = img_patch_separated_returned_true_size[:, margin : length_x - margin]
img_patch_ineterst_revised[:, index_x_d + margin : index_x_u - margin] = img_patch_separated_returned_true_size
@ -1473,7 +1471,7 @@ def return_deskew_slop(img_patch_org, sigma_des,n_tot_angles=100,
img_int[:,:]=img_patch_org[:,:]#img_patch_org[:,:,0]
max_shape=np.max(img_int.shape)
img_resized=np.zeros((int( max_shape*(1.1) ) , int( max_shape*(1.1) ) ))
img_resized=np.zeros((int(max_shape * 1.1) , int(max_shape * 1.1)))
onset_x=int((img_resized.shape[1]-img_int.shape[1])/2.)
onset_y=int((img_resized.shape[0]-img_int.shape[0])/2.)
@ -1538,7 +1536,7 @@ def return_deskew_slop_old_mp(img_patch_org, sigma_des,n_tot_angles=100,
img_int[:,:]=img_patch_org[:,:]#img_patch_org[:,:,0]
max_shape=np.max(img_int.shape)
img_resized=np.zeros((int( max_shape*(1.1) ) , int( max_shape*(1.1) ) ))
img_resized=np.zeros((int(max_shape * 1.1) , int(max_shape * 1.1)))
onset_x=int((img_resized.shape[1]-img_int.shape[1])/2.)
onset_y=int((img_resized.shape[0]-img_int.shape[0])/2.)

View file

@ -21,7 +21,7 @@ from ocrd_models.ocrd_page import (
)
import numpy as np
class EynollahXmlWriter():
class EynollahXmlWriter:
def __init__(self, *, dir_out, image_filename, curved_line,textline_light, pcgts=None):
self.logger = getLogger('eynollah.writer')

View file

@ -1,201 +0,0 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

View file

@ -173,7 +173,7 @@ class sbb_predict:
##if self.weights_dir!=None:
##self.model.load_weights(self.weights_dir)
if (self.task != 'classification' and self.task != 'reading_order'):
if self.task != 'classification' and self.task != 'reading_order':
self.img_height=self.model.layers[len(self.model.layers)-1].output_shape[1]
self.img_width=self.model.layers[len(self.model.layers)-1].output_shape[2]
self.n_classes=self.model.layers[len(self.model.layers)-1].output_shape[3]
@ -305,8 +305,7 @@ class sbb_predict:
input_1= np.zeros( (inference_bs, img_height, img_width,3))
starting_list_of_regions = []
starting_list_of_regions.append( list(range(labels_con.shape[2])) )
starting_list_of_regions = [list(range(labels_con.shape[2]))]
index_update = 0
index_selected = starting_list_of_regions[0]
@ -561,7 +560,7 @@ class sbb_predict:
if self.image:
res=self.predict(image_dir = self.image)
if (self.task == 'classification' or self.task == 'reading_order'):
if self.task == 'classification' or self.task == 'reading_order':
pass
elif self.task == 'enhancement':
if self.save:
@ -584,7 +583,7 @@ class sbb_predict:
image_dir = os.path.join(self.dir_in, ind_image)
res=self.predict(image_dir)
if (self.task == 'classification' or self.task == 'reading_order'):
if self.task == 'classification' or self.task == 'reading_order':
pass
elif self.task == 'enhancement':
self.save = os.path.join(self.out, f_name+'.png')
@ -665,7 +664,7 @@ def main(image, dir_in, model, patches, save, save_layout, ground_truth, xml_fil
with open(os.path.join(model,'config.json')) as f:
config_params_model = json.load(f)
task = config_params_model['task']
if (task != 'classification' and task != 'reading_order'):
if task != 'classification' and task != 'reading_order':
if image and not save:
print("Error: You used one of segmentation or binarization task with image input but not set -s, you need a filename to save visualized output with -s")
sys.exit(1)

View file

@ -394,7 +394,9 @@ def resnet50_unet(n_classes, input_height=224, input_width=224, task="segmentati
return model
def vit_resnet50_unet(n_classes, patch_size_x, patch_size_y, num_patches, mlp_head_units=[128, 64], transformer_layers=8, num_heads =4, projection_dim = 64, input_height=224, input_width=224, task="segmentation", weight_decay=1e-6, pretraining=False):
def vit_resnet50_unet(n_classes, patch_size_x, patch_size_y, num_patches, mlp_head_units=None, transformer_layers=8, num_heads =4, projection_dim = 64, input_height=224, input_width=224, task="segmentation", weight_decay=1e-6, pretraining=False):
if mlp_head_units is None:
mlp_head_units = [128, 64]
inputs = layers.Input(shape=(input_height, input_width, 3))
#transformer_units = [
@ -516,7 +518,9 @@ def vit_resnet50_unet(n_classes, patch_size_x, patch_size_y, num_patches, mlp_he
return model
def vit_resnet50_unet_transformer_before_cnn(n_classes, patch_size_x, patch_size_y, num_patches, mlp_head_units=[128, 64], transformer_layers=8, num_heads =4, projection_dim = 64, input_height=224, input_width=224, task="segmentation", weight_decay=1e-6, pretraining=False):
def vit_resnet50_unet_transformer_before_cnn(n_classes, patch_size_x, patch_size_y, num_patches, mlp_head_units=None, transformer_layers=8, num_heads =4, projection_dim = 64, input_height=224, input_width=224, task="segmentation", weight_decay=1e-6, pretraining=False):
if mlp_head_units is None:
mlp_head_units = [128, 64]
inputs = layers.Input(shape=(input_height, input_width, 3))
##transformer_units = [

View file

@ -269,10 +269,10 @@ def run(_config, n_classes, n_epochs, input_height,
num_patches = num_patches_x * num_patches_y
if transformer_cnn_first:
if (input_height != (num_patches_y * transformer_patchsize_y * 32) ):
if input_height != (num_patches_y * transformer_patchsize_y * 32):
print("Error: transformer_patchsize_y or transformer_num_patches_xy height value error . input_height should be equal to ( transformer_num_patches_xy height value * transformer_patchsize_y * 32)")
sys.exit(1)
if (input_width != (num_patches_x * transformer_patchsize_x * 32) ):
if input_width != (num_patches_x * transformer_patchsize_x * 32):
print("Error: transformer_patchsize_x or transformer_num_patches_xy width value error . input_width should be equal to ( transformer_num_patches_xy width value * transformer_patchsize_x * 32)")
sys.exit(1)
if (transformer_projection_dim % (transformer_patchsize_y * transformer_patchsize_x)) != 0:
@ -282,10 +282,10 @@ def run(_config, n_classes, n_epochs, input_height,
model = vit_resnet50_unet(n_classes, transformer_patchsize_x, transformer_patchsize_y, num_patches, transformer_mlp_head_units, transformer_layers, transformer_num_heads, transformer_projection_dim, input_height, input_width, task, weight_decay, pretraining)
else:
if (input_height != (num_patches_y * transformer_patchsize_y) ):
if input_height != (num_patches_y * transformer_patchsize_y):
print("Error: transformer_patchsize_y or transformer_num_patches_xy height value error . input_height should be equal to ( transformer_num_patches_xy height value * transformer_patchsize_y)")
sys.exit(1)
if (input_width != (num_patches_x * transformer_patchsize_x) ):
if input_width != (num_patches_x * transformer_patchsize_x):
print("Error: transformer_patchsize_x or transformer_num_patches_xy width value error . input_width should be equal to ( transformer_num_patches_xy width value * transformer_patchsize_x)")
sys.exit(1)
if (transformer_projection_dim % (transformer_patchsize_y * transformer_patchsize_x)) != 0:
@ -297,7 +297,7 @@ def run(_config, n_classes, n_epochs, input_height,
model.summary()
if (task == "segmentation" or task == "binarization"):
if task == "segmentation" or task == "binarization":
if not is_loss_soft_dice and not weighted_loss:
model.compile(loss='categorical_crossentropy',
optimizer=Adam(learning_rate=learning_rate), metrics=['accuracy'])
@ -365,8 +365,7 @@ def run(_config, n_classes, n_epochs, input_height,
y_tot=np.zeros((testX.shape[0],n_classes))
score_best=[]
score_best.append(0)
score_best= [0]
num_rows = return_number_of_total_training_data(dir_train)
weights=[]

View file

@ -260,7 +260,7 @@ def generate_data_from_folder_training(path_classes, batchsize, height, width, n
if batchcount>=batchsize:
ret_x = ret_x/255.
yield (ret_x, ret_y)
yield ret_x, ret_y
ret_x= np.zeros((batchsize, height,width, 3)).astype(np.int16)
ret_y= np.zeros((batchsize, n_classes)).astype(np.int16)
batchcount = 0
@ -446,7 +446,7 @@ def generate_arrays_from_folder_reading_order(classes_file_dir, modal_dir, batch
ret_y[batchcount, :] = label_class
batchcount+=1
if batchcount>=batchsize:
yield (ret_x, ret_y)
yield ret_x, ret_y
ret_x= np.zeros((batchsize, height, width, 3))#.astype(np.int16)
ret_y= np.zeros((batchsize, n_classes)).astype(np.int16)
batchcount = 0
@ -464,7 +464,7 @@ def generate_arrays_from_folder_reading_order(classes_file_dir, modal_dir, batch
ret_y[batchcount, :] = label_class
batchcount+=1
if batchcount>=batchsize:
yield (ret_x, ret_y)
yield ret_x, ret_y
ret_x= np.zeros((batchsize, height, width, 3))#.astype(np.int16)
ret_y= np.zeros((batchsize, n_classes)).astype(np.int16)
batchcount = 0