Semi-automatic methodology for unconstrained structured handwriting recognition in hospital bed head ticket automation

dc.contributor.authorSenaratne, GC
dc.date.accessioned2013-10-21T02:12:50Z
dc.date.available2013-10-21T02:12:50Z
dc.date.issued2010
dc.description.abstractThis paper first provides an introduction to die semi-automatic methodology for unconstrained structured handwritten scripts recognition in hospital bed head ticket recognition. Then based on extensive studies of the practices and procedures in a hospital system, this paper presents a realistic attainable solution to digitalize handwritten scripts. . I structure definition language featured by XML can be used to generalize this approach to any sort of handwriting convention We also provide an implementation of handwriting recognition algorithms with customization to suit this extreme case of handwriting recognition. This implementation is followed by a generalized statistical rule engine. Each word is interpreted by this rule engine and proper set of rules can gel the overall system accuracy even for 100%. We introduced self learning capability to the entire system so that the system would converge with lime enabling higher hit ratio. The digitalized information produced by the recognition process can be used with any existing medical record systems to provide value added hospital automation features He also suggest future enhancements to the system that would increase the robustness and accuracy without adding any overhead to the system.
dc.identifier.conferenceResearch for Industry
dc.identifier.pgnospp. 214-218
dc.identifier.placeFaculty of Engineering, University of Moratuwa
dc.identifier.proceeding16th Annual symposium on Research and Industry
dc.identifier.urihttp://dl.lib.mrt.ac.lk/handle/123/8232
dc.identifier.year2010
dc.languageen
dc.titleSemi-automatic methodology for unconstrained structured handwriting recognition in hospital bed head ticket automation
dc.typeConference-Extended-Abstract

Files

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Plain Text
Description:

Collections