Skip to content

Latest commit

 

History

History
42 lines (29 loc) · 1.45 KB

EquationSolver.md

File metadata and controls

42 lines (29 loc) · 1.45 KB
Raw

Equation Solver

Team members

  • Sidhya Jain
  • Chaitravi Chalke
  • Kalpana Yadav
  • Ananya Pawar
  • Rakshita Gond
  • Samina Attari
  • Neha Binwal

Mentors

  • Priyesh Vakharia

Project Description

Equation solver enables users to solve handwritten equations. The scanned image can be in any colour format (The colour of the background and digits can be according to the users convenience). We are still working on its webapp. The users for now can solve the equations on a local machine. The digit and operator recognition is done by our ML model which we have built using our knowledge of CNN, Tensorflow and OpenCV. The detected equation is solved using python libraries.

Tech Stack

  • CNN (Convolutional Neural Network)
  • Tensorflow
  • Python
  • OpenCV

Architecture

Layer 1 - Zero Padding | Conv | Batch Normalization | Zero Padding | Conv | Batch Normalisation | Max Pooling

Layer 2 - Zero Padding | Conv | Batch Normalization | Zero Padding | Conv | Batch Normalisation | Max Pooling

Layer 3 - Zero Padding | Conv | Batch Normalization | Zero Padding | Conv | Batch Normalisation | Max Pooling

Layer 4 - Batch Normalization | Flatten | Dense | Dense | Batch Normalization | Dense

Github link

https://github.com/sidhya3112/Equation-Solver-Using-CNN

Future Scope

For now we are running and testing on local machine. Our future goals are:

  • To improve the accuracy of prediction of our ML model.
  • To deploy our project in form of a web-app.