InvoiceMate

<div align="center"> <h1>InvoiceMate</h1>

<br> <br>

</div>

Update: License key expired, will renew sometime later.(you can fork and add your own key to run the app.)

Features:

Scanning UI

 User guidance to the best position for capturing the document with manual + automatic photo capture.

Pre-Processing of images.

 Inbuilt function from Tensorflow for poets are used.

Perspective Correction

 4 Point getPerspective Transform

Cropping UI

 User interface to crop pages with magnetic handles.

Text Recognition (OCR)

Text Rekognition is done using Amazon Textract.
The app also supports on device OCR

Screenshots

<div align="center"> <img src="https://github.com/Techno-Disaster/Invoice/blob/master/screenshots/Screenshot_20200308-084953_InvoiceMate.png" width="210"> &nbsp; <img src="https://github.com/Techno-Disaster/Invoice/blob/master/screenshots/Screenshot_20200308-084957_InvoiceMate.png" width="210"> &nbsp; <img src="https://github.com/Techno-Disaster/Invoice/blob/master/screenshots/Screenshot_20200308-085007_InvoiceMate.png" width="210"> &nbsp; <img src="https://github.com/Techno-Disaster/Invoice/blob/master/screenshots/Screenshot_20200308-103458_InvoiceMate.png" width="210"> &nbsp; <img src="https://github.com/Techno-Disaster/InvoiceMate/blob/master/screenshots/preprocessing.png" width="210"> &nbsp; </div>

Contributing

We're are open to enhancements & bug-fixes :smile: Also do have a look here

Note

• This project was done under 36 hours with little to no pre-preparation
• We use the Scanbot SDK for cropping the image and Amazon Textract to perform OCR.
• We use Flutter for our frontend for cross platform support.

Contributors

• @Techno-Disaster
• @neelansh15