ANN based Handwritten Maths Equation Solving System

Using Convolutional Neural Network to produce a robust handwritten equation solver is a delicate task in image processing. One of the most delicate challenges in computer vision exploration is handwritten fine expression recognition. The work is made more delicate by the fact that certain characters are segmented and classified. This project looks at quadratic equations, linear equations, cubic equations as well as differential equations. These equations must be honored and answered. Vertical compact protuberance analysis is used for segmentation. We use both connected element analysis and integrated connected element analysis methodologies. The characters are classified using the Convolutional Neural Network. Each appropriate is needed for the result of the problem. Char-acter string operation is used for discovery. Eventually, the results of the trial show that the strategy described is relatively effective. Equation solves are able of dealing with a wide range of fine equations.