Checkout counter

1. A classification device for identification of articles in an automated checkout counter, comprising:

a memory unit configured to store digital reference signatures, each of which digital reference signatures corresponds to an article identity, and

a processor connected to the memory unit;

wherein the classification device further comprises a first spectroscopy sensor which is connected to the processor, and a light source for illuminating an article wherein light emitted from the light source has a wavelength distribution covering at least operating wavelengths of the spectroscopy sensor;

wherein the first spectroscopy sensor is configured to determine a measured signature of an article from a unique reflection spectrum when the article is placed before, on or after a weight sensor;

wherein the processor is configured to compare the measured signature with the digital reference signature in order to identify the article as an existing article identity in the memory unit; and

wherein the first spectroscopy sensor is an NIR sensor configured to measure light having a wavelength between about 850 nm and about 2500 nm.

2. The classification device according to claim 1 , further comprising a second spectroscopy sensor configured to measure light having a wavelength between 200 nm and 1100 nm.

3. The classification device according to claim 2 , wherein each digital reference signature is created by at least one of the first and second spectroscopy sensors.

4. The classification device according to claim 2 , wherein at least one of the first and second spectroscopy sensors is a single array spectrometer.

5. The classification device according to claim 2 , wherein said second spectroscopy sensor is a visual spectroscopy sensor configured to measure light having a wavelength between 200 nm and 400 nm, and between 660 nm and 1100 nm.

6. The classification device according to claim 5 , wherein said visual spectroscopy sensor is configured to measure light having a wavelength in the complete interval between 200 nm and 1100 nm.

7. The classification device according to claim 6 , wherein at least one of the first and second spectroscopy sensors is arranged to measure light being reflected by said article.

8. The classification device according to claim 7 , further comprising at least one light source arranged to illuminate the article at a position at which at least one of the first and second spectroscopy sensors is directed.

9. The classification device according to claim 8 , wherein said at least one light source is configured to emit light having a wavelength between 200 nm and 2500 nm.

10. The classification device according to claim 1 , further comprising at least one of a contour sensor, a bar-code sensor, a symbol reading sensor configured for optical character recognition and machine text interpretation, and at least one of a color texture sensor, a color histogram sensor, and a further spectroscopy sensor, which co-operate together with the first spectroscopy sensor such that if the sensors as well as the first spectroscopy sensor are activated in one of the following combinations:

weight sensor and contour sensor and symbol reading sensor, or

contour sensor and symbol reading sensor, or

weight sensor and symbol reading sensor, or

weight sensor and color histogram sensor and contour sensor, or

weight sensor and contour sensor and color texture sensor and symbol reading sensor, or

weight sensor and barcode sensor, or

only symbol reading sensor, or

color histogram sensor and color texture sensor, or

color-histogram sensor, or

color texture sensor, or

the further spectroscopy sensor together with any of the above mentioned combinations, or

only a further spectroscopy sensor,

the remaining sensors are allowed to be deactivated or not active, depending on the article being identified.

11. The classification device according to claim 1 , further comprising an initial sensor which identifies the article by 100% and is arranged to be used during learning of the device by first identifying the article and subsequently conveying the article through the classification device where all of the sensors identify properties of the article, which properties will subsequently be stored in a database as digital reference signatures.

12. The classification device according to claim 11 , wherein the initial sensor is a barcode reader.

13. The classification device according to claim 1 , further comprising a tunnel shaped structure which shields a part of the classification device and at least some of the sensors completely or partly, in order to prevent unauthorized people from affecting the classification process.

14. The classification device according to claim 1 , wherein the first spectroscopy sensor is configured to read the article and to determine the digital reference signature during a learning procedure.

15. An automated checkout counter, comprising a classification device according to claim 1 .

16. An automated checkout counter according to claim 15 , further comprising a weight sensor for weighing the article.

17. The automated checkout counter according to claim 16 , wherein the weight sensor comprises a conveyor scale that conveys and weighs the article automatically.

18. The automated checkout counter according to claim 15 , wherein the classification device is arranged to weigh the article by means of the weight sensor before, during or after determining the measured signature, wherein the processor is arranged to use the weight of the article and the article identity for deciding the price of the article.

19. The automated checkout counter according to claim 15 , further comprising an interactive display unit connected to the processor for displaying information about the article identity, weight and price, which display unit is arranged to be used by a user for approving the displayed information.

20. A method for classifying articles in an automated checkout counter, comprising the steps of:

providing a classification device comprising a memory unit capable of storing digital reference signatures, each of which digital reference signatures corresponds to an article identity, a processor connected to the memory unit and to a weight sensor, and a spectroscopy sensor connected to the processor wherein the spectroscopy sensor is an NIR sensor configured to measure light having a wavelength between about 850 nm and about 2500 nm, and a light source for illuminating an article wherein light emitted from the light source has a wavelength distribution covering at least operating wavelengths of the spectroscopy sensor;

determining a measured signature from a unique reflection spectrum of an article by means of the spectroscopy sensor when the article is placed before, on or after the weight sensor; and

classifying the article by comparing the measured signature with the digital reference signature by means of the processor in order to identify the article as an existing article identity in the memory unit.