Handwritten Text to Editable Text Document
One of the major problem faced by every organization is management of old records of handwritten text that are likely to be subjected to further deterioration in the future. These old records are difficult to manage because of the sheer volume they exist in. This issue can be rectified if there existed a softcopy of those records but the Analysis of Handwritten text has been one of the major challenge in the field of image processing because of various writing styles, background lighting and text orientation. All the existing technique have failed when the letters in a text cannot be segmented properly. In our project, the text will be segmented and then segmented letters will be analysed and classified into appropriate class using a neural network .With the aid of Natural Language processing, we can analyse the context of a misclassified word and predict its occurrence in reference to the context. With a proper blend of image processing, Natural language processing and Convolutional Neural Network, we will recognize the text with a high degree of accuracy.