Handwriting recognition by using deep learning to extract meaningful features

Article type: Research Article

Authors: Pastor-Pellicer, Joan^a | Castro-Bleda, María José^{a; *} | España-Boquera, Salvador^a | Zamora-Martínez, Francisco^b

Affiliations: [a] Universitat Politècnica de València, Camino Vera s/n, Valencia 46021, Spain | [b] R&D Department, Veridas S.L., Pol. Ind. Talluntxe II, Tajonar 31192, Spain

Correspondence: [*] Corresponding author. E-mail: [email protected].

Abstract: Recent improvements in deep learning techniques show that deep models can extract more meaningful data directly from raw signals than conventional parametrization techniques, making it possible to avoid specific feature extraction in the area of pattern recognition, especially for Computer Vision or Speech tasks. In this work, we directly use raw text line images by feeding them to Convolutional Neural Networks and deep Multilayer Perceptrons for feature extraction in a Handwriting Recognition system. The proposed recognition system, based on Hidden Markov Models that are hybridized with Neural Networks, has been tested with the IAM Database, achieving a considerable improvement.

Keywords: Handwriting recognition, deep learning, convolutional neural networks

DOI: 10.3233/AIC-170562

Journal: AI Communications, vol. 32, no. 2, pp. 101-112, 2019