A Novel Hybrid CNN-LSTM Approach for Handwritten Text Recognition for the Washington Database

The handwritten text recognition (HTR) is a challenging task in the field of pattern recognition and machine learning. In this paper, a novel hybrid approach based on CNN and LSTM networks is proposed for HTR on the Washington database. The proposed approach consists of two stages. In the first stage, a CNN is used to extract features from the input images. The output of the CNN is then fed into a bidirectional LSTM network in the second stage for recognizing the handwritten text. The proposed approach was evaluated on the Washington database, which contains a large number of handwritten documents with various writing styles and languages. The experimental results showed that the proposed approach outperforms in terms of Training loss, validation loss and testing loss. Overall, the proposed hybrid CNN-LSTM approach offers a promising solution for HTR tasks on complex and diverse datasets like the Washington database.