Automatic Ground-truth Generation for Document Image Analysis and Understanding

Performance evaluation for document image analysis and understanding is a recurring problem. Many groundtruthed document image databases are now used to evaluate general algorithms, but these databases are less useful for the design of a complete system in a precise context. This paper proposes an approach for the automatic generation of ground-truth information using a derivation of publishing tools. An implementation of this approach illustrates the richness of the produced information.

• Type: Conference
• Year: 2007
• Where: ICDAR
• Authors: Pierre Héroux, Eugen Barbu, Sébastien Adam, Éric Trupin
• Keywords: Document Analysis, Document Image, Document Image Analysis, Document Image Databases, ICDAR 2007