| Many industries, from turnpikes and railroads to shipping and supply-line management, have all recently shown the need for vehicle, package, or product tracking and identification. Several companies have tried to fill this need with RFID technologies, relying on expensive RF readers at strategic locations, and inexpensive RF transmitter tags placed on the tracked item. However, for systems with existing infrastructure, the effort to place these tags on each item becomes a costly and mired operational endeavor. Furthermore, most of the items that have a need to be tracked are already uniquely marked visually, usually with a number plate (e.g. automobile or train car), a barcode or otherwise. An identification system based on visual recognition is less intrusive, less expensive, less wasteful, and easier to maintain. The Modular Auto-Identification by the Recognition of Existing Markings (MAREM) system has been developed and described in this thesis as a response to these needs. |
