A Survey on Handwritten Text Recognition

Handwritten Text Recognition (HTR) is the detection of characters from images. HTR is a complex task due to the variability and diversity of handwritten characters in the script. CNNs are a type of deep learning algorithm that can automatically learn features from images and are widely used in image recognition tasks. This paper presents a CNN-based approach for HTR that achieves state-of-theatre performance on a benchmark dataset. The proposed approach involves a pre-processing step to normalize and segment the input images, followed by a CNN architecture that consists of several convolutional layers and fully connected layers. The network is trained using a massive character labelled dataset. The outcomes demonstrate that the suggested method achieves excellent accuracy in recognizing characters and can be applied to real-world applications such as document digitization and text-to-speech conversion.