Handwritten Mathematical Expressions Recognition using Back Propagation Artificial Neural Network

Abstract

Handwritten mathematical expressions recognition is yet challenging task due to its intricate spatial structure, tangled semantics and 2-dimensional layout of the characters. There is a still room for enhancement in recognition rate. Artificial neural network is superior to disentangle classification problems. In this paper, feedforward back propagation neural network is implemented to achieve both character recognition and mathematical structure recognition with upgrade in effective performance in addition to accuracy of the experimental results including lessen efforts. System proves its potency by recognizing expressions in analysis of math documents.

Fault Diagnosis and Prognosis for Satellite Formation Flying: A Survey

“…The remaining useful lifetime (RUL) prediction can also be used to monitor the system's condition. In terms of practicality and feasibility, RUL prediction is more attractive [43].…”

Fault Detection, Isolation and Prognosis of Aerospace Systems Using Adaptive Growing Recurrent Neural Networks

“…The hidden layers in a neural network have different types that can give different results. Convolutional neural networks (CNN) were firstly proposed by LeCun [52] and they aim to learn abstract features by alternating and stacking convolutional and pooling layers. Joint loss CNN were studied and used to make RUL predictions for bearings [53].…”

A Survey on Evaluating Handwritten Iterative Mathematical Expressions

“…The recognition rate achieved is 97.58% and 98.40%. Sagar Shinde, Rajendra Waghulade [8] has made an effort to implement feed-forward back propagation neural network for the recognition of handwritten mathematical expressions. The system presents an approach to recognize handwritten straight line equation and quadratic equation with the focus on improving recognition rate, performance, processing time and accuracy.…”