104 questions
- Bountied 0
- Unanswered
- Frequent
- Score
- Trending
- Week
- Month
- Unanswered (my tags)
1
vote
1
answer
185
views
Is it possible to extract character-level coordinates from EasyOCR?
I am trying to extract text from documents using ocr, but for my use case I need character-level coordinates.
I am getting following output from EasyOCR:
[
[
[
60,
...
2
votes
0
answers
155
views
Read Arabic text in Python with easyOCR
I have a problem reading the text in Arabic. I did some processing and contrast, but the text is not fully readable.
!pip install easyocr
!pip install --upgrade easyocr opencv-python
import easyocr
...
3
votes
0
answers
5k
views
"UserWarning: 'pin_memory' argument is set as true but no accelerator is found, then device pinned memory won't be used."
I am trying to use EasyOCR,
reader = easyocr.Reader(['en'], gpu=False)
def read_sign(sign):
result = reader.readtext(sign)
return 'STOP' in result`
However while it did work at the start, it ...
0
votes
0
answers
107
views
PaddleOCR fails to correctly detect German diacritics (ä, ö, ü, ß) in text recognition
I'm using PaddleOCR to detect German text from images, but I've noticed that it consistently fails to correctly recognize German diacritics. For example:
"Spaß" is detected as "SpafS&...
0
votes
0
answers
97
views
EasyOCR is not detecting Arabic letters in a license plate image
I am trying to extract text from a Moroccan license plate using EasyOCR, but the Arabic letter in the middle is not being detected.
Problem:
When processing an image of a license plate, EasyOCR ...
1
vote
1
answer
506
views
Running EasyOCR on streamlit cloud, Neither CUDA nor MPS are available - defaulting to CPU. Downloading detection model, please wait
The process takes long time and there is no limit to how long it will takes, it just does not ends.
Neither CUDA nor MPS are available - defaulting to CPU. Note: This module is much faster with a GPU.
...
0
votes
0
answers
217
views
easyocr readtext ends with illegal instruction exception
I'm trying to process a very clear image with python's easyocr package.
This is my image: phonenum.png
Here is my code:
import easyocr
reader = easyocr.Reader(['en'])
result = reader.readtext('images/...
0
votes
0
answers
641
views
Could not initialize NNPACK! Reason: Unsupported hardware while deploying using docker file
We are using Easy OCR ultra analytics modal for one of the applications. We are taking video input and processing it using the modal to identify the text.
The code works fine on GPU & CPU. I ...
0
votes
0
answers
83
views
Why "PyRun_SimpleString("from PIL import Image")" does not work in debug mode?
I'm writing a c++ program that use Python library EasyOCR to read some text. I wrote such code:
Py_Initialize();
PyRun_SimpleString("import sys");
PyRun_SimpleString("import easyocr&...
0
votes
1
answer
135
views
EasyOCR inconsistent in recognizing letter "e" when by itself
I have a function in Python which reads text in Portuguese with EasyOCR. For some reason it doesn't always recognize the "e" between bigger words, which is a common connector word in this ...
2
votes
0
answers
648
views
AWS Textract Too Slow
We are using AWS Textract to OCR PDFs. We don't use any of their advanced form/table detection, queries, etc. features. We only want to take a PDF (or PNG/JPEG), and get back a list of all "...
0
votes
0
answers
65
views
Id Card Reader is not adding the texts in their labels
So, I am working on an id reader app using python and kivy and opencv and easyocr, the reads is nice but I want to put the reads in their label like the name in the name label and the national id ...
0
votes
0
answers
53
views
Draw on top of see-through hole in window
So I recently found out about the concept of "see through hole" on windows.
Managed to do it in PyQT5 using following code:
import sys
from PyQt5.QtCore import Qt, QRect, QPoint
from PyQt5....
0
votes
0
answers
61
views
RuntimeError: Error(s) in loading state_dict for Model:
I am getting
RuntimeError: Error(s) in loading state_dict for Model:
Missing key(s) in state_dict:
when I try to load my fine-tuned model that I tuned by the easyocr trainer.
import easyocr
...
0
votes
0
answers
128
views
Text extraction with OCR
The provided image has various elements of a monitor. I want to extract the text of each of them. I have tried various approaches, but there is a lot of text that I cannot identify.enter image ...