IDA: A System for Automated Sorting, Indexing, and Classification of Documents.
Gerd Maderlechner, Thomas M. Bruckner, Peter Suda · 1996
IDA (Intelligent Document Analysis) is a modular software system, which assists to automate paper document entry. IDA consists of the following components: layout analysis, preclassification, OCR interface, fuzzy string matching, text categorization, lexical, syntactical and semantic analysis. The system has been applied to a variety of tasks: Presorting of forms, reports and letters, index extraction for archiving and retrieval, text column analysis in real estate register documents, in-house mail distribution, and classification of business letters by text content. This paper presents an overview of the architecture and applications of the system. Therefore focus of attention techniques have been developed to limit OCR evaluation to few regions of interest. 2 Architecture of the system IDA has a modular and open architecture with three main components, layout analysis, OCR, and content analysis. The layout analysis consists of the following modules (Fig. 1): 1