An inclusive review on deep learning techniques and their scope in handwriting recognition

Abstract

Deep learning encompasses a set of machine learning algorithms that can merge raw inputs into intermediate feature layers. These algorithms have shown impressive results across various fields. In particular, deep learning has achieved human-level performance in computer vision and pattern recognition across multiple domains. To achieve state-of-the-art performance in different areas, various architectures and activation functions are used to compute between hidden and output layers. This paper provides a survey of current research on deep learning in the field of handwriting recognition. Despite recent advancements in deep learning methods for improving speed and accuracy in handwriting recognition, extensive literature review reveals that deep learning has not completely transformed the field and still needs to address significant challenges. Nevertheless, there have been promising developments compared to previous advancements. Moreover, the lack of labeled data for training poses a challenge in this domain. However, the current analysis of handwriting recognition predicts that deep learning will bring about significant changes in various fields such as image processing, speech recognition, computer vision, machine translation, robotics and control, medical imaging, medical information processing, bio-informatics, natural language processing, cyber security, and more.