In the factory floor, the Engineer enters data on the sheet. This sheet at the end of the repair cycle, manufacturing process, etc. is manually entered into the database. This manual process is timeconsuming and repetitive. Currently, there are OCR technologies that do not perform well with Cursive Handwritten data. Build a Deep learning OCR and Azure services that which can work with handwritten and image-text data. Since each individual has a different handwriting style.
In many automobile and manufacturing industries, a huge amount of data is handwritten or in the image. Still, in many government sectors in India, the document are handwritten. Optical character recognition makes it possible to convert many kinds of texts or photos into editable, searchable, and analyzable data. Researchers have been using AI and ML methods to automatically evaluate printed and handwritten documents for the past ten years in order to digitize them. Industries use paper sheets to enter data collected during work. This data is then manually entered into a database for further usage. But manually entering the data can be tedious work. So to solve this problem we can use the Optical Character Recognition technique, which is widely used to detect handwritten cursive text. We need this technology because if the data entered is misinterpreted by humans it can lead to failure or loss to the industry.
-
Clone this repository
git clone https://github.com/TheNobody-12/OCR-APP.git -
Install python version 3.9 or higher version
-
Install Virtual Environment using Pip in command line.
pip install virtualenv -
Create a Virtual Environment in command line.
python<version> -m venv <virtual-environment-name> -
Activate the Environment in command line.
.\<virtual-environment-name>\Scripts\activate -
In activated environment, install all python dependencies.
pip install -r requirements.txt -
Run code file using below command.
python app.py
Sarthak Kapaliya(TheNobody-12) 👑 Admin |