Text_Recognition

• In this project we can send an image using the android app (apk file already provided) consisting of text, and run a local server (flask).
• When image is sent to server, the server does the processing and revert back the image with bounding box along with the recognized text.

Environment

• Make a separed environment to install dependencies and run flask server for backend
• USE: python3 -m venv my-venv for making environment named as "my-venv"
• To activate environment use source ./my-venv/bin/activate

Installation

• Install the packages and dependencies using pip install -r requirements.txt

• sudo apt-get install tesseract-ocr

• Download and install the .apk file in your android mobile.

Run

• python api.py
• Note the host Ipv4 Address given in the terminal
• (example- http://192.169.30.52:8082/) Here IP: 192.169.30.52, PORT:8082

Input

• Open the Text Recognition app installed using the given .apk
• Select any image from galary for applying text detection+recognition to it.
• Enter IPv4 address and PORT number noted in above step.
• Click 'CONNECT TO SERVER'

Output

• You will receive the file in the backend
• The detection and recognition process will take place
• You will get the output image in the app's image view and text as caption.

Regards: Himanshu Khairajani