Published Paper Details:

Machine learning and deep learning have become increasingly important as a result of technological advancement. Handwriting recognition, robotics, artificial intelligence, and many more industries are now using machine learning and deep learning approaches. Such systems require data training, allowing our machine to learn and make the necessary predictions. In this research, a Handwritten Equation solver with a respectable accuracy of 98 percent is demonstrated. It was trained on handwritten digits and mathematical symbols using a Convolutional Neural Network and certain image processing techniques. The images of the digits 0 to 9, plus and minus signs (+), and handwritten symbols * make up the dataset. For the purpose of extracting the features, we shall use contour extraction. In this Project we build the model using convolutional neural network and train the model to evaluate the equations written by hand in this we use the dataset of numbers and operators written by hand. The input image of hand written equation is given the image is converted into gray background for this we use contour extraction for obtaining the features. The output is produced by the model by evaluating the equation.