Research On Handwritten Character Recognition Based On Deep Learning

Optical character recognition has always been one of the most interesting topics.It is designed to recognize text in natural images.Although this subject has been studied for many years,it is still a great challenge to develop optical character recognition technology which is equivalent to human ability.In recent years,with the needs of production and living,there is an urgent need to convert paper documents into digital documents for search and preservation.Optical character recognition can be described as converting an image that may be a handwritten text image or a printed text image.Research in this field has been going on for more than half a century.In recent years,many technologies have been applied to text extraction in document images.In recent years,deep learning has been frequently used to solve hot topics in the field of machine learning,such as image recognition,classification,retrieval and so on.Deep learning learns the characteristics of samples by building a multi-layer neural network,this greatly avoids the tedious and complex manual feature extraction.Although deep learning has been used in optical character recognition,handwritten optical character recognition is still a hot research topic in the field of machine learning because of the complexity of handwritten characters.This thesis will use the method of deep learning to study the recognition of handwritten Chinese characters.This thesis is to further improve the accuracy of handwritten Chinese character recognition based on convolution neural network and combined with the advantage of Recurrent neural network.