Remote sensing of characters on 3D objects

This paper introduces a formal approach to the recognition of characters (OCR) on remote objects. Currently most applications using this technique are ad hoc. Lack of formalism creates the risk of fragmented adhocracy. This theoretical basis explores the potential of character recognition for automated reading using video and digital cameras. In order to do successful recognition the typical problems experienced in these applications should be well understood. The approach also suggests the use of commercial OCR software developed for scanners to be used in the recognition process. Finally, some experimental results of practical applications are given. This approach is pragmatic and can be used in many fields.