Optical Character Recognition

Project on Multilingual Optical Character Recognition

Tasks

• Process image - Focus on the text area, removing all corners and extra features
• Extract text - Using tesseract, extract text from the image
• Translate - Translate English text to Spanish
• Server - Flask server to integrate Python modules with the frontend
• Website - Basic website to upload image and display OCR text and translated text
• Document server code

APIs

• Yandex API

   # Url
   URL = "https://translate.yandex.net/api/v1.5/tr/translate"
   
   # Parameters
   # key: API KEY
   # text: Text to translate
   # land: from - to language
   PARAMS = {
       'key': api_key,
       'text': text,
       'lang': "en-es"
   }
   
   # Request
   r = requests.get(url=URL, params=PARAMS)
   
   # Parse response
   data = r.text
   data = re.sub('<[^>]+>', '', data)[1:]

Run

• Install requirements

pip install -r requirements.txt

• Start the server

python server.py

• Run the website
• Upload image

Contributors

😍 Nishant Rodrigues