Image analysis – Applications – Mail processing
Reexamination Certificate
1999-12-08
2004-08-17
Mehta, Bhavesh M. (Department: 2623)
Image analysis
Applications
Mail processing
C209S584000, C209S900000, C235S462070, C235S470000, C382S190000, C382S317000, C715S252000, C715S252000
Reexamination Certificate
active
06778683
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method and apparatus for reading and decoding information and more particularly to systems for automatically imaging and processing information contained on the surface of an object traveling on a moving conveyor belt.
DESCRIPTION OF THE PRIOR ART
Throughout the world, competition among express package delivery services has reached epoch proportions. In order to maintain or even increase their market share, these organizations are under intense pressure to process packages more efficiently and deliver them to their destinations quicker than their competitors. Although recent hardware and software advances have been made in the area of automated package sorting and routing, further improvements are desirable in order to achieve the economic benefits that would naturally flow from a complete and fully automated process.
Today, modem package processing systems rely on the use of conveyor belts to introduce packages into a processing facility, and to convey them through the facility as they are imaged by a high resolution camera and processed for eventual delivery to a final location. One such system is disclosed in U.S. Pat. No. 5,770,841 to Moed et al., entitled “System and Method for Reading Package Information.” Upon arriving at a distribution center, the packages are unloaded from a delivery truck onto a conveyor belt and arranged in single file by a singulator. A package height sensor determines the height of each package and passes the information onto the high resolution camera's focusing system. As each package passes beneath the high resolution camera, an image of the package surface is captured. The digitized image is passed onto an image processing system for automated analysis of the information. Conventional package imaging systems, like that discussed in Moed et al., typically require the stream of packages to be at least singularized (single file) and sometimes even justified (oriented along one side of the conveyor) before they are imaged because they have a very limited in-focus object plane and field of view (FOV). In essence, when two adjacent articles lie in two different focus planes (i.e., they are different distances from the camera), the two articles must be imaged separately or they both will not appear clearly in a single image. Similarly, when two adjacent articles are in the same focus plane but transversely separated, again the articles must be imaged separately or it is likely that only a partial surface of one or both articles will be captured in a single image. Unfortunately, the process of properly singularizing and justifying packages takes valuable time and severely impacts processing throughput. Moreover, the number of packages that can actually be placed on a conveyor belt for processing at any one time is severely restricted when the packages must be placed in single file. Given the speed at which many delivery service conveyor belts operate, the process of focusing, imaging and then re-focusing for each package that must be processed places a significant burden on even the most sophisticated imaging systems.
After the imaging system captures an image of the package, experience has proven that the most error-prone aspect of the process is actually recognizing the handwritten textual information on the package and determining in a timely manner, the package's proper destination from that information. Compounding this already difficult task is the reality that the volume of packages processed by major package delivery companies has reached the level of several million parcels per day and climbing.
In an effort to improve the processing throughput without correspondingly increasing manpower levels, package delivery companies have resorted to the use of high-speed imaging systems optical and intelligent character recognition techniques to quickly image the packages and process the printed and cursive information contained on them. Unfortunately, many of these conventional character recognition engines can only recognize handwritten characters at approximately a 98% recognition rate, which still results in 2-3 errors per package.
Thus, there is a need in the art for an imaging system that allows more latitude in the placement of packages on a conveyor. Furthermore, there is a need for a processing system that provides the capability to quickly and accurately recognize printed and cursive characters beyond that of current systems.
SUMMARY OF THE INVENTION
Systems and methods consistent with this invention provide a apparatus for reading and decoding information extracted from a form. In the system of the present invention, packages are randomly placed on a conveyor belt, with their labels visible to a package imaging camera. As they move along the conveyor belt, the package imaging camera takes an image of a portion of the belt at an instant in time. A digital image of the packages within the field of view of the camera is then transferred to the processing system for analysis. The processing system identifies individual packages in the image, extracts them and then analyzes the information written, or printed on the package labels. The analysis process utilizes conventional Optical Character Recognition (OCR) and Intelligent Character Recognition (ICR) techniques to evaluate the information written on the package label, including check boxes. Once the information is decoded, the system accesses a package record database entry associated with the decoded machine-readable code, or creates a new database entry if necessary. The retrieved database entry is then populated with verified information extracted from the package shipping form. When an unknown word image is encountered, the field-specific recognition process utilizes client-specific or installation-specific lexicon information to aid in the recognition process. The lexicon information is continuously revised and optimized based on processed form information. In a preferred embodiment, verified destination addresses associated with a client are alphabetized or rank-ordered based on frequency of occurrence. Every time an address associated with a client is verified, it is added to the database and the order of the associated destination addresses modified, if appropriate. It is only after the system determines that the originating client is not stored in the database does it resort to a ZIP+4 or similar database to verify a destination address.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
REFERENCES:
patent: 5384452 (1995-01-01), He
patent: 5737438 (1998-04-01), Zlotnick et al.
patent: 5880451 (1999-03-01), Smith et al.
patent: 6123361 (2000-09-01), Cohen
patent: 6152371 (2000-11-01), Schwartz et al.
patent: 6370259 (2002-04-01), Hobson et al.
patent: 32 06 267 (1982-02-01), None
Bonner Brett Bracewell
Jankevics Andris
Skaaksrud Ole-Petter
Desire Gregory
Federal Express Corporation
Finnegan Henderson Farabow Garrett & Dunner LLP
Mehta Bhavesh M.
LandOfFree
Method and apparatus for reading and decoding information does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for reading and decoding information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for reading and decoding information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3294362