Apparatus and method for spectroscopic product recognition and identification

• US 5,867,265 A
• Filed: 04/17/1997
• Issued: 02/02/1999
• Est. Priority Date: 08/07/1995
• Status: Expired due to Term

• Current Assignee: NCR Corporation (NCR Voyix Corporation)
• Original Assignee: NCR Corporation (NCR Voyix Corporation)
• Inventors: Thomas, Gordon Albert
• Primary Examiner(s): HANTIS, KAREN
• Application Number:
  US08/844,671
• Time in Patent Office: 656 Days
• Field of Search: 356/328, 356/326, 356/430, 356/428, 356/446, 356/240, 356/416, 356/419, 356/402-411, 250/576, 250/226, 209/587, 364/526, 186/61, 235/383, 235/462, 702/134-136, 702/28, 702/84
• US Class Current: 356/328
• CPC Class Codes: G01N 2021/8466 Investigation of vegetal material, e.g. leaves, plants, fruits
• CPC Class Codes: G01N 21/25 Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
• CPC Class Codes: G01N 21/31 Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry G01N21/72 takes precedence
• CPC Class Codes: G07G 1/0054 with control of supplementary check-parameters, e.g. weight or number of articles

Abstract

An optical identification system includes a light source with a broad wavelength spectrum that is directed on an object to be identified. Suitable optical components, such as, one or more collimating lenses gather light that is reflected from the object and direct this light into a spectrometer. The spectrometer disperses the collimated light using a dispersing element, such as one or more gratings, prisms or a combinations of both, onto an array of detectors. The array of detectors may be comprised of a linear diode array or a charge-coupled device (CCD) array which indicates the amount of light at each of a finely-spaced set of wavelengths covering a wide spectral range. The detectors are sensitive over a wavelength region, for example, in the case of silicon detectors from near-infrared plus the visible region, e.g., from 250 nm to 1100 nm. The set of signals from the detectors is read with an analog to digital converter, and transferred to a computer in the form of a spectrum. A set of known spectra determine the reference spectra and the unknown test spectrum is compared with the reference sets. A software program in the computer compares the test spectrum with reference spectra sets utilizing a statistical program. The program takes into account how much the known spectra vary from one another in addition to the average values. A display reads out a list of possible matches in rank order that have a probability of match greater than a predetermined threshold. An operator checks that the first listed item is correct and either accepts the first choice or indicates the correct choice. As an alternative, the system could automatically accept the first choice.

Claims

1. A produce recognition system comprising:

a checkout counter;

a bar code reader at the checkout counter, including a scale for producing a weight signal representing a weight of a produce item to be purchased during a transaction at the checkout counter;

a light source within the checkout counter for illuminating the produce item with light for identifying the produce item;

a spectrometer within the checkout counter which receives a reflected portion of the light from the produce item, including a dispersion element which disperses the reflected portion of the light into a plurality of finely spaced wavelengths of light, and a photodetector array which converts the plurality of finely spaced wavelengths of light to corresponding proportional electrical signals to produce a first spectra;

storage means for storing a set of reference spectra;

a computer coupled to the storage means, the bar code reader, and the spectrometer, which compares the first spectra from the photodetector array to the set of reference spectra to identify a second spectra within the set of reference spectra which most closely resembles the first spectra, and which determines a price of the produce item associated with the second spectra based upon the weight of the produce item.

2. The system as recited in claim 1, wherein the light source comprises a broad wavelength spectrum light source which emits light in a useful spectral range including part of the infrared spectral range, all of the visible spectral range, and part of the ultra-violet spectral range.

3. The system as recited in claim 1, wherein the light source comprises a tungsten-halogen lamp.

4. The system as recited in claim 1, wherein the photodetector array comprises a linear diode array.

5. The system as recited in claim 1, wherein the photodetector array comprises a charge-coupled device array.

6. The system as recited in claim 1, wherein the dispersion element resolves between 100 and 1100 individual wavelengths.

7. The system as recited in claim 1, wherein the computer is a transaction processing computer located at the checkout counter.

8. The system as recited in claim 1, wherein the computer displays a plurality of produce item identification choices including a choice associated with the second spectra and additional choices associated with additional spectra within the set of reference spectra, records entry of a single choice of the plurality of produce item identification choices by an operator, and adds a price associated with the single choice to the transaction.

9. The system as recited in claim 1, wherein the computer is located at the checkout counter.

10. The system as recited in claim 1, wherein the computer is located remotely from the checkout counter.

11. A produce recognition system comprising:

a checkout counter;

a bar code reader at the checkout counter, including a scale for producing a weight signal representing a weight of a produce item to be purchased during a transaction at the checkout counter;

a light source within the checkout counter for illuminating the produce item with light for identifying the produce item while the produce item is being weighed by the scale;

a spectrometer within the checkout counter which receives a reflected portion of the light from the produce item, including a dispersion element which disperses the reflected portion of the light into a plurality of finely spaced wavelengths of light, and a photodetector array which converts the plurality of finely spaced wavelengths of light to corresponding proportional electrical signals to produce a first spectra;

storage means for storing a set of reference spectra;

a computer coupled to the storage means, the bar code reader, and the spectrometer, which compares the first spectra from the photodetector array to the set of reference spectra to identify a second spectra within the set of reference spectra which most closely resembles the first spectra, and which determines a price of the produce item associated with the second spectra based upon the weight of the produce item.