navaneeth-sharma / akshara-jaana

A OCR Project for Reading New and Old Kannada Texts

Hello there 👋 It's Navaneeth here

Simple Developer with Full stack skills who

• respects clean code, good documentation and simplicity
• working on Vue.js , Fastapi, Djnago , Alpine.js, Tailwind, UI/UX, Figma Designs
• learning advanced Python, Rust, Deep Learning and design principles
• into Transformers, Diffusion, LLMs and much more

akshara-jaana's Issues

Add `paddleocr` Model for OCR

Currently only tesseract model is present, which is not that efficient for old Kannada scripts (printed), So implement the OCR in paddleocr. No Extra Installation Required

Modules required

pip install paddleocr paddlepaddle

Sample Code

import cv2
from paddleocr import PaddleOCR, draw_ocr

# Creating OCR object
ocr = PaddleOCR(lang="ka")

# Reading image
img_path = "/content/test_image.jpeg"
img = cv2.imread(img_path)

# Running OCR
result = ocr.ocr(img)

# Printing results
for line in result:
    print(line)


boxes = [line[0] for line in result]  # the box locations in the image (just like cv2)
txts = [line[1][0] for line in result] # the extracted text 
scores = [line[1][1] for line in result] # confidence score

Reference

• https://github.com/PaddlePaddle/PaddleOCR
• https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/doc/doc_en/algorithm_overview_en.md

Do not use bare except

https://realpython.com/the-most-diabolical-python-antipattern/

Feature Request: Add a spell correction model for in the pipeline

To improve the accuracy of the overall pipeline, it is better to add a spell correction Large Language model. Use Opensource models like Mistral, LLama, Falcon and others

Add `easyocr` Model for OCR

Modules required

pip install easyocr

Sample Code

import easyocr

img_path = "/content/test_image.jpeg"
reader = easyocr.Reader(['kn','en'])

result = reader.readtext(img_path, paragraph=True) 

References

• https://www.jaided.ai/easyocr/
• https://github.com/JaidedAI/EasyOCR

Unicode Error while reading PDF

Unicode Error while reading PDF

To Reproduce
Steps to reproduce the behavior:

1. Install dependencies
2. run below commands

from AksharJaana.main import ocr_engine
ocr_engine('path.pdf')

Expected behavior
The below error is occurring while reading the pdf

Desktop (please complete the following information):

• OS: Windows
• Version : (github clone)