HANDWRITTEN TEXT RECOGNITION USING MACHINE LEARNING

In this rapidly evolving world everything is getting digitalized, Exchange of information plays a crucial role in any work. Most of the times we may not find the facility to create a document so we use pen and paper but people may not understand everything expressed on paper due to many reasons line bad penmanship etc. So we are building a platform through which one could convert the handwritten text to editable text. Our main aim is to classify the words directly and character segmentation. We use CNN for this process. We use long term memory networks (LSTM) with convolution to construct bounding boxes for each character. We divide the each given character and classify them and reconstruct for each word of neural networks for Handwriting character recognition.