Data processing: financial – business practice – management – or co – For cost/price
Reexamination Certificate
1998-08-31
2002-03-26
Trammell, James P. (Department: 2161)
Data processing: financial, business practice, management, or co
For cost/price
C382S110000
Reexamination Certificate
active
06363366
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to checkout systems such as employed for supermarkets and other retail outlets. The present invention further relates to pattern recognition systems and methods.
2. Description of the Prior Art and Related Information
Bar code scanning systems are ubiquitous in supermarket checkouts and other checkout systems in various retail outlets. This type of product identification and pricing for checkout systems is extremely effective and has revolutionized retail checkouts both in speed and accuracy. Bar code handling at retail checkouts has also enabled detailed inventory control. Nonetheless bar codes are not well-suited for certain types of articles and certain types of applications. Most notable of these applications where bar codes are not well-suited is the identification and pricing of produce, i.e., fruits, vegetables and other agricultural products, at supermarket checkouts. The handling of produce is not well suited to bar code systems since supermarket shoppers are typically free to purchase produce in varying quantities based on weight or number of items. Therefore, while some produce is prepackaged and bar coded the majority of produce is handled as loose collections of variable numbers of items.
To handle such variable collections of produce, checkout personnel must either memorize a lengthy list of prices or produce codes or consult printed tables. Both procedures are potentially prone to error and much slower than the handling of bar coded products. Therefore, produce handling at supermarket checkouts represents both a time bottleneck and a source of error in the handling of a high volume of supermarket checkout items. This time problem may require additional checkout aisles to be opened, and additional employees to be occupied at the checkouts, to provide a desired degree of throughput. Also, to reduce error the training involved in handling of produce at supermarket checkouts is greater than required for handling of bar coded products. Where a high turnaround of supermarket checkout personnel is experienced, this additional training can introduce higher employee costs and possible lost revenues.
Therefore, it will be appreciated that a need presently exists for an improved approach to the handling of identification and pricing of produce at supermarket checkouts. In particular, it would be highly desirable to provide the same degree of handling efficiency for produce that is available for bar coded items. Also, any solution to this problem should introduce a minimum of additional cost and hardware and provide a high degree of reliability.
SUMMARY OF THE INVENTION
The present invention provides a system and method for identification, pricing and inventory monitoring of produce at supermarket checkouts which automates the checkout process to a degree comparable to the handling of bar coded products. Furthermore, the present invention provides such a system and method having a high degree of reliability and without adding substantial costs to the overall checkout system.
In accordance with a preferred embodiment of the present invention, a produce identification and pricing system includes an imager which provides image data from produce being passed through the checkout system. A signal processor provides a real-time identification of the produce employing at least a spectral matching operation between a spectral signature vector derived from the image data from the produce and stored spectral signature vectors. An imager which is enhanced for detecting spectral information in the near infrared region may be employed. If a suitable match is not found using the spectral image data, the identification operation proceeds using more complicated signatures derived from two dimensional image data. The more complex signatures may be compared to a truncated target set which is dramatically reduced from the initial target set based on the use of the spectral image data. If a unique best match is not found within a predetermined time window the best candidate matches are displayed and the checkout operator selects the correct match through a touch screen, push button or voice input.
The system also includes a CPU which receives the produce identification from the signal processor and accesses a pricing database to determine the pricing structure for the identified produce. If the produce is priced on a weight basis an input from a scale is enabled and the weight is used to compute the total price for the collection of produce. If the produce is priced on a number of item basis, a count operation for the signal processor is enabled and a count estimate is made based on the image data. The count operation employs a three-dimensional image of the produce collection and preferably a properly positioned second imager is employed to provide the three-dimensional image data. The number of items determined from the count operation is then used by the CPU to compute total price.
The system may improve the produce identification accuracy using the corrected identification information from the operator. Preferably, a plurality of like checkout systems may be networked together in a typical large supermarket implementation. Information from the networked systems may be shared to allow rapidly updating the matching algorithms based on inputted correction information from the various checkouts. In this way variations of produce due to new shipments, on the shelf ripening, bruising or other factors may be rapidly included into the matching algorithms so that updated signatures for the produce provide highly accurate matching despite day-to-day variations in the produce characteristics. Also, to facilitate matching of new produce shipments a calibration station may be provided, e.g., in the supermarket warehouse area, to allow sample produce from new shipments to be entered into the networked produce recognition system as the new shipments of produce are received.
Various other novel features and advantages of the present invention are provided which will be appreciated from the following detailed description of the present invention.
REFERENCES:
patent: 3867041 (1975-02-01), Brown et al.
patent: 4645080 (1987-02-01), Scopatz
patent: 4741042 (1988-04-01), Throop et al.
patent: 5083638 (1992-01-01), Schneider
patent: 5426282 (1995-06-01), Humble
patent: 5546475 (1996-08-01), Bolle et al.
patent: 5555464 (1996-09-01), Hatlestad
patent: 5561526 (1996-10-01), Huber et al.
patent: 5621460 (1997-04-01), Hatlestad et al.
patent: 5631976 (1997-05-01), Bolle et al.
patent: 5649070 (1997-07-01), Connell et al.
patent: 5661817 (1997-08-01), Hatlestad et al.
patent: 6005959 (1999-12-01), Mohan et al.
patent: 40 05 081 (1991-08-01), None
no author, Self-checkout systems add ‘on-line’ efficiency, Jun. 1998; Discount Store News, v37 n11, p70+; DialogWeb copy pp. 1-3.
Dixon Thomas A.
Trammell James P.
LandOfFree
Produce identification and pricing system for checkouts does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Produce identification and pricing system for checkouts, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Produce identification and pricing system for checkouts will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2830180