problem statement :
you have to create bounding
boxes for each line in the Sanskrit text that is provided in the attached
file. After the bounding box is created for the entire line (and not the
words). You need
to save each line in separate jpg files and send us the output.
Additionally, you need to save the coordinates of the bounding box in a JSON
format as shown below:
{
"box1": {
"top_left": [0, 0],
"top_right": [0, 10],
"bottom_left": [10, 0],
"bottom_right": [10, 10]
},
"box2": {
....
}
}
"output_with_rectangles.pdf" contains the pdf with rectangluar boxes around sanskrit texts
Sanskrit_italic.TTF is the devanagiri font used while printing the sanskrit text in the image .
Coordinate_&_draw_lines contains script for finding out coordinates and draw the bounding boxes for sanskrit text.
write_to_json.py contains the function used to write the json file that is annot_data.json
images folder contains all the sanskrit texts in an jpg format
pip list
Package Version
------------------- ----------
annotated-types 0.6.0
blis 0.7.11
catalogue 2.0.10
certifi 2023.11.17
cffi 1.16.0
charset-normalizer 3.3.2
click 8.1.7
cloudpathlib 0.16.0
confection 0.1.4
contourpy 1.2.0
cycler 0.12.1
cymem 2.0.8
fasttext 0.9.2
fasttext-langdetect 1.0.5
filelock 3.13.1
fonttools 4.47.2
frozenlist 1.4.1
fsspec 2023.12.2
idna 3.6
importlib-resources 6.1.1
Jinja2 3.1.3
kiwisolver 1.4.5
langcodes 3.3.0
langid 1.1.6
MarkupSafe 2.1.4
mpmath 1.3.0
multidict 6.0.4
murmurhash 1.0.10
networkx 3.2.1
numpy 1.26.3
packaging 23.2
pillow 10.2.0
pip 23.3.2
preshed 3.0.9
pybind11 2.11.1
pycparser 2.21
pydantic 2.6.0
pydantic_core 2.16.1
PyMuPDF 1.23.20
PyMuPDFb 1.23.9
pyparsing 3.1.1
python-dateutil 2.8.2
pytz 2023.4
pyvips 2.2.2
PyYAML 6.0.1
requests 2.31.0
scipy 1.12.0
setuptools 58.0.4
six 1.16.0
smart-open 6.4.0
soupsieve 2.5
spacy 3.7.2
spacy-legacy 3.0.12
spacy-loggers 1.0.5
srsly 2.4.8
sympy 1.12
thinc 8.2.2
torch 2.2.0
torchvision 0.17.0
tqdm 4.66.1
typer 0.9.0
typing_extensions 4.9.0
tzdata 2023.4
unicode 2.9
Unidecode 1.3.8
urllib3 2.2.0
wasabi 1.1.2
weasel 0.3.4
yarl 1.9.4
zipp 3.17.0
these were all the packages i did install at the time , a lot of them were not used .adi-lb-phoenix/optical_character_recognition
To detect sanskrit texts from the given pdf file , draw a rectangular bounding box around them , write the coordinates in json file and print those sanskrit texts in an image
Python