Handwritten Gujarati Script Recognition

Their are numerous languages in this world and they are spoken and written every day. In modern times everyone has recognized the need for the recognition of handwritten scripts by machines for various purposes. Handwriting recognition has been one of the most demanding and interesting research areas in the field of image processing and machine learning since a long time. It finds its use in various applications which include conversion of handwritten documents into structural text form, reading aid for the blind, bank cheques, etc. Handwritten Script recognition can also play a significant role in preserving the culture, famous works, and literature from old times that were handwritten. For this reason, it is particularly important to have a handwritten script recognition system that can be used irrespective of the document type.The project aims at developing an offline handwritten Gujarati script recognition system which can be effectively used for recognizing handwritten Gujarati scripts. There is a significant difference in the handwriting of different people and the contours formed by characters of the Gujarati Script and these factors poses a stiff challenge in this process. This process involves the creation of a neural network that will take input as an image (.jpg,.jpeg,.png, etc.) and then the features will be extracted from it, the neural network will recognize this text and output it on the machine. This will help in the recognition of Gujarati scripts of different handwriting and will help reduce human error and save the time of individuals.