Skip to content

Latest commit

 

History

History
132 lines (103 loc) · 6.35 KB

README.md

File metadata and controls

132 lines (103 loc) · 6.35 KB
Raw

Multilingual Handwriting Recognition and Translation

To create a model that recognises written text from an image file uploaded by the user and translates the recognised text into the languages provided.

Table of Contents

Objectives

  • To accurately recognise the text contained in the image uploaded.
  • To provide accurate translations for both target languages.
  • To provide appropriate translation, with respect to using region specific and colloquial words in the translations.
  • To have a flexible dataset which can be altered to suit any use case by appending or deleting words/sentences provided in the dataset for translation.
  • To make our project applicable to many real-life based use cases (see Objectives - pt. 4).

Objectives completed

  1. Accuracy achieved:

    • Translation to French: 93.76%
    • Translation to Spanish: 96.86%

  2. Since the datasets are stored in .txt file format, they can be infinitely edited to suit multiple use cases such as:

    • A program that translates ancient texts or manuscripts.
    • Travel-based translator to help people communicate when they are abroad.
    • Digitisation of documents in several languages in order to go paperless.
    • A system which converts handwritten text into digital format and translates it into audio for visually impaired individuals, enabling them to access and understand written content.

Usage

Installation

Installation Steps 
1. Clone the Repository: 
      - Open the GitHub repository of the project. 
      - Click on the "Code" button and copy the repository URL. 
      - Open your terminal or command prompt. 
      - Navigate to the directory where you want to clone the repository. 
      - Run the following command to clone the repository:
            git clone <repository_url>

2. Setting up the Environment: 
      - Make sure Python is installed on your system. If not, download and install Python from the official website (https://www.python.org). 
      - Create a virtual environment (optional but recommended) to isolate the project dependencies using the following command:
            python -m venv env

      - Activate the virtual environment:
            -- For Windows: Run the command: env\Scripts\activate
            -- For macOS/Linux: Run the command: source env/bin/activate. 
            -- Install the required dependencies by running the following command:
                pip install -r requirements.txt

3. Saving Files to Drive: 
      - Make sure you have a Google account and access to Google Drive. 
      - Follow the instructions in the documentation or code to integrate the project with Google Drive. 
      - Authenticate the project with your Google account and grant the necessary permissions to access Google Drive. 
      - Modify the code to specify the Google Drive directory path where you want to save the files.

4. Google Colab Integration: 
      - Upload the project files (or the entire cloned repository folder) to your Google Drive. 
      - Open Google Colab in your web browser.
      - Mount your Google Drive in Colab by running the following code snippet:
            from google.colab import drive
            drive.mount('/content/drive')

      - Navigate to the project folder using the file navigation panel in Google Colab.
      - Open the project notebook (.ipynb file) and start working with it.

5. PyCharm Integration: 
      - Open PyCharm and create a new project. 
      - Set the project interpreter to the virtual environment you created earlier. 
      - Add the project files (or the entire cloned repository folder) to your PyCharm project. 
      - Open the project notebook (.ipynb file) and start working with it.

Navigation

Show instructions
1. Accessing the website: 
    - Open your web browser and navigate to the URL of the hosted website.
    - The main page of the website will be displayed.

2. Uploading the handwritten text image: 
    - On the page, locate the "Choose file" button.
    - Click on the "Choose file" button to open the file selection window. 
    - Browse your computer to find the handwritten text image file that you want to process. 
    - Select the file and click "Open" to initiate the upload process. 

3. Performing handwriting recognition: 
    - Locate the "Predict" button on the main page. 
    - Click on the "Predict" button to initiate the handwriting recognition process. 
    - Wait for the system to process the uploaded image and recognize the handwritten text. 
    - Once the recognition process is completed, the recognized text will be displayed on the screen.

4. Translating the recognized text: 
    - Locate the "Translate to French/Spanish" button on the main page. 
    - Click on the "Translate to French/Spanish" button to initiate the translation process. 
    - The system will use the selected translation language to translate the recognized text. 
    - Once the translation process is completed, the translated text will be displayed on the screen.

Note:
    - The Choose file button opens up a window for the user to select an image file from their device.
    - The Predict button gives the user the text contained within the image as an output.
    - Translate to French / Translate to Spanish buttons invoke the translation models in the program and show the translated output.

Team

Mentors:

Irfan Siddavatam ( [email protected] )
Ashwini Dalvi ( [email protected] )

Members:

Sr No. Name e-mail git-profile
1. Vedant Hire [email protected] vedanthire14
2. Pranav Mahulkar [email protected] Pranav-Mahulkar
3. Chinmay Maitre [email protected] Chinmay-Maitre08
4. Hiral Patel [email protected] hiral25p