68 questions
3
votes
1
answer
49
views
PaddleOCR is leaking extracted data from a previous document
I've implemented an OCR queue using PaddleOCR and noticed that some documents are receiving the same extracted data as the previously processed document.
So far, this behaviour seems to be isolated to ...
- 4,959
-1
votes
0
answers
38
views
Are there any way to use paddleocr on ppc64le cpu ibm power 9 cpu?
i have tried build package, conda, pip etc. can't do it please help me for any solution for install paddleocr on ppc64le cpu ibm power 9 cpu. thanks.
1
vote
0
answers
67
views
PP-OCRv5 lang="japan" Setting Outputs Chinese: Is This a Bug or a Feature?
I am currently using PP-OCRv5 with the PP-OCRv5_server_det model. I've set the language parameter to lang="japan" in an attempt to process Japanese text, but the output frequently contains ...
- 11
0
votes
0
answers
270
views
PaddleOCR OCR analyzes Left-to-Right instead of Right-to-Left for Arabic- How to process RTL languages correctly?
I am using PaddleOCR with the Arabic language model (lang='ar') to perform OCR on Arabic images. While PaddleOCR correctly recognizes the Arabic characters, it processes the text in a Left-to-Right (...
0
votes
1
answer
210
views
Paddle OCR fail to extract single digit
I’m using Paddle OCR to extract prices from some products price tags, in most cases it give good results but I noticed that it fails to extract prices with single digit
When I edit the image and add ...
- 173
0
votes
0
answers
534
views
Paddle Ocr not detecting text when too much of the image is empty
I got a multitude of envelopes captured like in this screenshot (address hidden by me for confidentiality). For most of them Paddle OCR is working great. But for a couples it doesn't captures much. If ...
- 83
0
votes
0
answers
628
views
Easy way to create dataset for PaddleOCR custom font training?
I want to use Paddle to OCR a really fancy custom font. It's good but i know it could be better if I fine tuned it.
I've run into this tutorial on Medium but it assuems the reader already has training ...
- 19
1
vote
0
answers
1k
views
RuntimeError: (PreconditionNotMet) Cannot Load cuDNN Shared Library in Docker Container with PaddleOCR
I am currently working on an OCR task using PaddleOCR within a Docker container. I am encountering the following error when trying to execute my application:
RuntimeError: (PreconditionNotMet) Cannot ...
- 31
4
votes
2
answers
569
views
Paddle OCR not able to extract single digits from image
I am doing an image OCR using Paddle OCR. Below is the sample code I am using:
from paddleocr import PaddleOCR
import os
image_file = "3_496.png"
ocr = PaddleOCR(use_gpu=True)
image_path = ...
- 7,596
0
votes
1
answer
1k
views
Is it possible to get the angle of rotation of the text area via PaddleOCR?
PaddleOCR can correctly recognize 90, 180 and even 270 degree rotated text in a mode use_angle_cls=True, but it doesnt provide any information about the angle of rotation in a result.As a result, area ...
- 23
1
vote
1
answer
694
views
How to сorrectly recognize 90 and 270 degrees rotated text?
I need to do OCR task on images containing ordinary horizontal text, and, besides this, text with 90 and 270 degree rotation Like this
I use python and PaddleOcr .PaddleOCR can recognize rotated text, ...
- 23
1
vote
1
answer
740
views
What type of image processing required to make an image a good input for ONNX converted paddleOCR inference model?
I'm trying to switch a PaddleOCR model to ONNX for better performance. It normally takes just the image path or base64 and gives results with preprocessing handled internally. But in ONNX, it suddenly ...
- 11
0
votes
1
answer
161
views
PaddleOCR for Tajik Language
Tajik language uses cyrillic, do I have to create new model from scratch or fine tune Russian cyrillic model?
Tajik has extra letters ӣ, ҳ, қ, ҷ, ӯ, ғ
Tajik and the Russian language are different
- 139
1
vote
0
answers
540
views
%env not update after downgrade CUDA from 12.2 to 11.7 on Google Colab
When i use my Colab GPU to train Handwriting Recognition based on PaddleOCR pipeline, i have to downgrade its CUDA to 11.7 so it can work well with PaddleOCR, i check !nvcc -V its got updated but when ...
- 11
0
votes
1
answer
239
views
I've been trying to run this paddleocr code but it keeps on saying result is undefined even though it's already been defined in the first line
%timeit result = ocr.ocr(img_path, cls =True)
for line in result:
print(line)
I couldn't figure anything out as I do not understand what is wrong with it
0
votes
0
answers
2k
views
Import paddleHub: AttributeError: module 'paddle' has no attribute 'version'
I try to run paddleHub in my colab, but I face a question
import paddlehub as hub
the error
AttributeError: module 'paddle' has no attribute 'version'
I try to import 2.0 up version but it's still ...
0
votes
1
answer
4k
views
Not able to import paddleocr library on Google Colab
I am not able to import paddleocr library on Google Colab after the install of paddlepaddle and paddleocr successfully. It hits the error as shown below:
**from paddleocr import PaddleOCR, draw_ocr**
...
- 3
1
vote
1
answer
4k
views
How do I use a pre-trained PaddleOCR model within code?
I intend to use PaddleOCR to detect image orientation and plan to use their Text Angle Classification Model. However, I got stuck trying to figure out how to install it and integrate it into my code (...
- 801
1
vote
1
answer
3k
views
How to save and load PaddleOCR model?
How to save PaddleOCR model? How to load it afterwards from the saved file?
I'd like to have something like this:
import PaddleOCR
model = PaddleOCR(use_angle_cls=True, lang='en')
save_path = '/some/...
- 445
0
votes
1
answer
962
views
Reading 7 segment digits using Paddle OCR
I am trying to read the 7 segment display using Paddle OCR, sometimes it reads correctly otherwise it reads poorly given the image is clear and visible to read
For example:
This image is clear and ...
- 145
1
vote
6
answers
15k
views
Error: Can not import paddle core while this file exists
I am currently using intel i3 with no gpu,
I created a virtual environment in python with python version 3.10.11, while my current python version is 3.11.3
In the virtual env
I tried installing ...
- 145
0
votes
1
answer
4k
views
How to load non-default PaddleOCR detection model in a python script?
PaddleOCR seems to support different algorithms/models for text detection, as written in their algorithms overview. However, when trying to load a different model from "DB" (for example &...
- 11
2
votes
1
answer
2k
views
PaddleOCR installation failure
when I am trying to install PaddleOCR, I'm running into an installation error while trying to install pymupdf. I'm currently on Python 3.8.16
error: command 'swig' failed with exit status 1
[end of ...
2
votes
2
answers
1k
views
PaddleOCR Error flag 'flagfile' was defined more than once
Am encountering an issue running PaddleOCR on M1 Macbook
ERROR: flag 'flagfile' was defined more than once (in files '/Users/paddle/xly/workspace/f2bafd01-b80e-4ac8-972c-1652775b2e51/Paddle/build/...
- 443
1
vote
1
answer
4k
views
Paddle OCR gives "device id must be less than GPU count" error
I am trying to use Paddle OCR for reading numbers from images, but it gives me this error:
"(InvalidArgument) Device id must be less than GPU count, but received id is: 0. GPU count is: 0.
[Hint: ...
- 95