DEV Community


Kirana automated invoice generator

ayushbathrey profile image Ayush Bathrey ・1 min read

My Final Project

Edge computing-based invoice generator and can differentiate up to 46 classes with respect to the size of the product.

Demo Link

Link to Code

[Note]: #

How To Test Object Detection Classifier for Multiple Objects Using TensorFlow (GPU) on Windows 10

Brief Summary

Last updated: 10/18/2019 with TensorFlow v1.14

What this model can predict/detect ?

It can check upto 46 different items , which may have brand but different size

Brands it can detect:

  • 1. Coke Glass bottle , Can , Plastic bottle 300ml to 2L.
  • 2. Colgate Small , Medium , Size , Plax.
  • 3. Nescafe Cappuccino , Gold , Classic availabe in small and medium size.
  • 4. Toothbush.
  • 5. Gooday biscuit small , medium , large.
  • 6. Lays small , medium , large/family.
  • 7. Brittania Toast small , medium, large.
    • I also made a YouTube video that walks through this tutorial. Any discrepancies between the video and this written tutorial are due to updates required for using newer versions of TensorFlow

      If there are differences between this written tutorial and the video, follow the

How I built it (what's the stack? did I run into issues or discover something new along the way?)

  1. Tensorflow object detection API.
  2. Flask Web development Framework.
  3. HTML, CSS, JavaScript, Bootstrap, Jquery.


This was project was pretty exciting to work, the most time-consuming part of the development was dataset collection (as no official dataset was present) and image labeling part.

[Final Note]: # Finally my hard work paid off and was called for an interview via this project.

Discussion (0)

Editor guide