Checkout counter
First Claim
1. A classification device for identification of articles in an automated checkout counter, 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, andat least two different sensors configured to determine a measured signature of an article, wherein at least one said sensor is a spectroscopy sensor in the form of a spectrometer configured to operate in a wave length interval of 850-2500 nm, wherein the measured signature determined from the at least one sensor is a digital representation of a reflectance spectrum in the wave length interval;
wherein said processor is configured to compare said measured signature with the digital reference signatures, and to calculate a matching probability of a predetermined number of article identities, wherein the processor is configured to determine the article identities by comparing the matching probability from the different sensors and selecting the article identities having the highest matching probability from at least one of the different sensors regardless of whether any of the article identities associated with the different sensors are the same.
6 Assignments
0 Petitions
Accused Products
Abstract
A classification device (2) for identification of articles (3) in an automated checkout counter is presented. The device comprises a memory unit (5) capable of storing digital reference signatures, each of which digital reference signatures corresponds to an article identity, a processor (6) connected to the memory unit (5), and at least one sensor (4, 7, 14, 15, 16, 17, 18, 24) configured to determine a measured signature of an article (3) wherein said processor (6) is configured to compare said measured signature with the digital reference signatures, and to calculate a matching probability of a predetermined number of article identities.
-
Citations
11 Claims
-
1. A classification device for identification of articles in an automated checkout counter, 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 at least two different sensors configured to determine a measured signature of an article, wherein at least one said sensor is a spectroscopy sensor in the form of a spectrometer configured to operate in a wave length interval of 850-2500 nm, wherein the measured signature determined from the at least one sensor is a digital representation of a reflectance spectrum in the wave length interval; wherein said processor is configured to compare said measured signature with the digital reference signatures, and to calculate a matching probability of a predetermined number of article identities, wherein the processor is configured to determine the article identities by comparing the matching probability from the different sensors and selecting the article identities having the highest matching probability from at least one of the different sensors regardless of whether any of the article identities associated with the different sensors are the same. - View Dependent Claims (2, 3, 4, 5)
-
-
6. 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 at least two different sensors configured to determine a measured signature of an article, wherein said sensor is a spectroscopy sensor in the form of a spectrometer configured to operate in a wave length interval of 850-2500 nm, wherein the measured signature determined from the at least one sensor is a digital representation of a reflectance spectrum in the wave length interval, comparing said measured signature with the digital reference signatures calculating a matching probability of a predetermined number of article identities; and determining the article identities by comparing the matching probability from the different sensors and selecting the article identities having the highest matching probability from at least one of the different sensors regardless of whether any of the article identities associated with the different sensors are the same. - View Dependent Claims (7, 8, 9, 10, 11)
-
Specification