Overview
FrançaisABSTRACT
This article presents a panorama of research on handwriting recognition, from the recognition of isolated letters up to that of full sentences. It presents the issue and principles associated to handwriting recognition, without detailing the most advanced approaches but by providing, however numerous, references allowing for a deepened knowledge of the notions dealt with. The main constitutive elements of a recognition system are also reviewed: pretreatment, segmentation, modeling, learning, recognition, integration of linguistic knowledge, etc. The various approaches are presented and in particular the recognition of so-called "on-line" handwriting recognition. All this information concerns the systems developed for peripherals or new generation computers which are equipped with tactile or sensitive screens: pen-based computer (tablet), new generation cell phone (Smartphone) or even Interactive Whiteboard (IWB).
Read this article from a comprehensive knowledge base, updated and supplemented with articles reviewed by scientific committees.
Read the articleAUTHOR
-
Éric ANQUETIL: Senior lecturer, IRISA laboratory, INSA Rennes
INTRODUCTION
Handwriting is a primordial mode of communication for human beings. It is both universal and highly personal. Each person develops his or her own handwriting style, which evolves constantly throughout life. Studies in graphotherapy have shown the central role that handwriting can play in a person's development.
Until now, computerization has tended to diminish the role of handwriting in communication: the keyboard/mouse device is gradually replacing the pen, to the detriment of handwriting. Even more recently, the widespread use of e-mail and SMS has only amplified this phenomenon, systematizing it earlier and earlier in everyday life.
The retreat of handwriting from keyboards is, in a way, forced by the need to access new technologies underpinned by increasingly powerful computer functions. There's no denying that all these new technologies represent real progress. However, it should be noted that, most of the time, man is asked to absorb these technologies at the cost of a relatively major adaptation effort. This effort is motivated by the attraction exerted by these new services: we are then ready to enter an "SMS-coded" text using the ten keys of a telephone keypad to communicate. However, not everyone can easily master the constraints imposed as prerequisites for access to these new services. Not to mention the fact that, even if the youngest people have a remarkable ability to adapt, we may not be giving enough consideration to the impact this may have on the growing rejection of handwriting skills and, by extension, of mother tongue and intellectual training.
Recently, new pen-oriented technologies have come to maturity: new-generation cell phones (Smartphones), tablet computers (Tablet PCs or UMPCs) and interactive whiteboards (TBIs). The trend is therefore gradually shifting, repositioning handwriting at the heart of man-machine communication. Coupling the written and gestural modalities with these new technologies aims to make them more user-friendly by introducing new uses that will facilitate access for the greatest number of people. Research into automatic handwriting recognition has made considerable progress over the last ten years. Even if not all the problems have yet been solved, "on-line" handwriting recognition engines are making progress and now benefit from more computing power; they are now sufficiently powerful to envisage direct interaction with the machine via handwriting.
Of course, this does not mean doing away with traditional peripherals (keyboard/mouse), which are highly effective in many contexts. Rather, the idea is to put the user back at the center of interaction with the machine, so that he or she once again becomes the main actor. In certain mobility contexts, and in many document...
Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference
This article is included in
Digital documents and content management
This offer includes:
Knowledge Base
Updated and enriched with articles validated by our scientific committees
Services
A set of exclusive tools to complement the resources
Practical Path
Operational and didactic, to guarantee the acquisition of transversal skills
Doc & Quiz
Interactive articles with quizzes, for constructive reading
On-line handwriting recognition
Bibliography
Websites
•Imadoc, research team on writing and documents http://www.irisa.fr/imadoc/
•INRIA's Interstices on-line scientific popularization magazine: handwritten line recognition http://interstices.info/display.jsp?id=c_17397
Events
National conferences
•CIFED (colloque international francophone sur l'écrit et le document): francophone conference supported by GRCE (groupe de recherche en communication écrite)
•RFIA (pattern recognition and artificial intelligence): conference supported by AFRIF (Association française pour la reconnaissance et l'interprétation de formes) and AFIA (Association française d'intelligence artificielle)....
Standards and norms
InkML, W3C standard format for electronic ink representation http://www.w3.org/2002/mmi/ink.html
UPX, electronic ink annotation format. This format is associated with InkML.
Directory
Manufacturers – Suppliers – Distributors (non-exhaustive list)
Sites to find information on pen-oriented computers (Tablet, Tablet PC, smartphone...) :
...Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference