Historical documents present many challenges for offline handwriting recognition systems, among them, the segmentation and labeling steps. Carefully annotated text lines are needed to train an HTR system. In some scenarios, transcripts are only available at the paragraph level with no text-line information. In this work, we demonstrate how to train an HTR system with few labeled data. Specifically, we train a deep convolutional recurrent neural network (CRNN) system on only 10% of manually labeled text-line data from a dataset and propose an incremental training procedure that covers the rest of the data. Performance is further increased by augmenting the training set with specially crafted multi scale data. We also propose a model-based normalization scheme which considers the variability in the writing scale at the recognition phase. We apply this approach to the publicly available READ dataset. Our system achieved the second best result during the ICDAR2017 competition [1].

The digitization of historical handwritten document images is important for the preservation of cultural heritage. Moreover, the transcription of text images obtained from digitization is necessary to provide efficient information access to the content of these documents. Handwritten Text Recognition (HTR) has become an important research topic in the areas of image and computational language processing that allows us to obtain transcriptions from text images. State-of-the-art HTR systems are, however, far from perfect. One difficulty is that they have to cope with image noise and handwriting variability. Another difficulty is the presence of a large amount of Out-Of-Vocabulary (OOV) words in ancient historical texts. A solution to this problem is to use external lexical resources, but such resources might be scarce or unavailable given the nature and the age of such documents. This work proposes a solution to avoid this limitation. It consists of associating a powerful optical recognition system that will cope with image noise and variability, with a language model based on sub-lexical units that will model OOV words. Such a language modeling approach reduces the size of the lexicon while increasing the lexicon coverage. Experiments are first conducted on the publicly available Rodrigo dataset, which contains the digitization of an ancient Spanish manuscript, with a recognizer based on Hidden Markov Models (HMMs). They show that sub-lexical units outperform word units in terms of Word Error Rate (WER), Character Error Rate (CER) and OOV word accuracy rate. This approach is then applied to deep net classifiers, namely Bi-directional Long-Short Term Memory (BLSTMs) and Convolutional Recurrent Neural Nets (CRNNs). Results show that CRNNs outperform HMMs and BLSTMs, reaching the lowest WER and CER for this image dataset and significantly improving OOV recognition.

Arabic handwritten documents present specific challenges due to the cursive nature of the writing and the presence of diacritical marks. Moreover, one of the largest labeled database of Arabic handwritten documents, the OpenHart-NIST database includes specific noise, namely guidelines, that has to be addressed. We propose several approaches to process these documents. First a guideline detection approach has been developed, based on K-means, that detects the documents that include guidelines. We then propose a series of preprocessing at text-line level to reduce the noise effects. For text-lines including guidelines, a guideline removal preprocessing is described and existing keystroke restoration approaches are assessed. In addition, we propose a preprocessing that combines noise removal and deskewing by removing line fragments from neighboring text lines, while searching for the principal orientation of the text-line. We provide recognition results, showing the significant improvement brought by the proposed processings.

Modern mobile devices and networks permit to develop applications that reduce the language barrier for tourists and visitors in foreign countries. This paper describes a mobile application built on a distributed architecture that allows tourists to obtain additional information about location names and menu entries in Arabic language. By simply taking a photo, translation and additional information are displayed to users in their preferred language. While several mobile applications exist to translate the texts into photos, the present application offers the advantage of translating both handwritten and printed texts. The work was focused on two important aspects: the recognition system and the mobile application. The major challenge was to build a recognition system able to recognize handwritten or printed texts of various writing styles or fonts. Therefore, we propose an original approach consisting of recognizing handwritten and printed words using the Balamand-ENST handwriting recognition system, trained on handwritten texts. Experiments conducted on a collected database have shown that the recognition system trained on handwritten text provides excellent performance when used in recognizing printed text. This recognition system and its performance in both modalities are briefly described in this paper. In addition, a full description on how the mobile application was designed and its functionality is presented.

Historical documents present many challenges for offline handwriting recognition systems, among them, the segmentation and labeling steps. Carefully annotated textlines are needed to train an HTR system. In some scenarios, transcripts are only available at the paragraph level with no text-line information. In this work, we demonstrate how to train an HTR system with few labeled data. Specifically, we train a deep convolutional recurrent neural network (CRNN) system on only 10% of manually labeled text-line data from a dataset and propose an incremental training procedure that covers the rest of the data. Performance is further increased by augmenting the training set with specially crafted multiscale data. We also propose a model-based normalization scheme which considers the variability in the writing scale at the recognition phase. We apply this approach to the publicly available READ dataset. Our system achieved the second best result during the ICDAR2017 competition.