Advancements in Handwriting Recognition: Deep Learning Techniques Applied to Kannada Language

Numerous applications use handwriting recognition, such as optical character recognition, document analysis, and automated form generation. Computer vision and machine learning techniques have been extensively explored to recognize handwritten letters due to the abundance of handwritten text. It has been shown that convolutional neural networks (CNNs) and recurrent neural networks (RNNs) perform exceptionally well when classifying vintage images in the computer vision field. This work introduce a robust, dynamic, and efficient approach to handwritten character recognition in Kannada. Leveraging the strengths of advanced architectures and pre-processing free deep learning techniques, our method offers enhanced performance and adaptability. Additionally, the paper provides a comparative analysis of the utilization of CNNs, LSTMs, and vision transformers in the context of handwritten character recognition.