This project utilizes optical character recognition (OCR) and translation to translate text within images from one language to another. It performs the following steps:
- OCR Processing: The project extracts text and its bounding boxes from input images using the EasyOCR library.
- Translation: It translates the extracted text using the Google Translator API.
- Text Replacement: The translated text is then overlaid onto the image, replacing the original text while maintaining its position and style.
- Output: Finally, the modified image with translated text is saved to an output folder.
- Clone this repository to your local machine.
- Install the required Python dependencies using
pip install pipenv && pipenv install.
- Place your input images in the
inputfolder. - Run the script
main.py. - Translated images will be saved in the
outputfolder.
- Supported languages for OCR can be seen here
- Supported languages for Google Translate can be obtained using the following code:
from deep_translator.constants import GOOGLE_LANGUAGES_TO_CODES print(GOOGLE_LANGUAGES_TO_CODES)
- Adjustments to text languages, recognition thresholds, translation services, or image processing parameters can be made within the script.
- EasyOCR - For OCR processing.
- Google Translator - For text translation.
- Pillow (PIL Fork) - For image manipulation.