Handwriting Recognition with Artificial Neural Networks a Decade Literature Review

Abstract

Deep Learning Artificial Neural Networks has pushed forward researches in the field of pattern recognition, furthermore in human handwriting recognition. From online to offline approach, signature verification, writing or writer identification, segmentation or features extraction, a multitude of Artificial Neural Networks (ANNs) models are applied in the process. This paper focuses on the literature review of human handwriting recognition with ANN's over the last decade. We propose an exploratory analysis of 294 research papers collected from five indexed research engines: ACM Digital Library, IEEE digital library, Science Direct, Scopus and Web of Science. Our aim is to provide a research papers distribution across years and journals, a Keywords frequency analysis using cloud visualization, and a Natural Language Processing Topic Modeling using Non-Negative Matrix Factorization (NMF). The results of this study show that the number of research papers reached noticeably a peak in the 2010 with 44 published papers; also Pattern Recognition was the top publishing journal with 12 published papers. As for the topic modeling using NMF we obtained 3 topics listed as follows: 1) Feature Extraction and segmentation techniques for Handwritten Texts Recognition; 2) Signature Verification in Biometric security for Off-line Authentication; 3) Assessment Systems for Student Identification