OCR of Legacy Documents as a Building Block in Industrial Disaster Prevention

Legacy text documenting or recording operational details from industrial or other human-made facilities which constitute potential hazards may contain important safety critical information. In these cases it is desirable to make such safety critical information as is implicitly present in older records readily available and accessible to modern day authorities which have to deal with the facility from an industrial disaster prevention point of view. An important first step in such an analysis, we argue, is a robust optical character recognition (OCR) stage, for digitising often decade old records containing valuable information for industrial disaster prevention or management which may otherwise be lost or unavailable at the right time. In this paper we present an overview of a project concerned with the study and analysis of legacy records from the operational history of a deep geological repository for nuclear waste and present a preliminary study on segmenting letterhead information in legacy correspondence concerning this particular facility.