Automated object identification and processing based on digital imaging and physical attributes
A method and associated systems for object identification and subsequent processing based on digital imaging and physical attributes. An object-identification system receives, in a materials-handling environment, a digital image and physical attributes that characterize an unidentified object. An attempt is made to identify the object by matching the image and attributes to those of known objects stored in an image database, an attribute database, or another external source. The object is associated with a label that identifies the actual object, associates the object with a similar object that may be substituted for the actual object in a desired application, or designates the object as unidentifiable. The digital image, label, and external sources used to identify the object may be updated by associating them with metadata gathered during the identification process. Subsequent processing is governed by business rules that operate as functions of the label data.
1. A computerized object-identification system comprising a processor, a memory coupled to the processor, an interface to an imaging device, a set of interfaces to a set of electronic sensors, and a computer-readable hardware storage device coupled to the processor, the storage device containing program code configured to be run by the processor via the memory to implement a method for automated object identification and processing based on at least one of a digital image of an object and a physical attribute of the object, the method comprising:
the computerized object-identification system receiving notice that the object has not been identified;
the system receiving information that describes an unidentified object;
the system identifying the object as a function of the received information;
the system further identifying a set of related objects as a further function of the received information;
the system ranking the related objects as a function of a value of a degree of similarity of each object of the set of related objects to the unidentified object, wherein a higher-ranked object of the set of related objects is more similar to the unidentified object than is a lower-ranked object of the set of related objects object;
the system, at a time after the receiving notice, directing a labeling device to generate a unique machine-readable species label that associates the unidentified object with at least one species of object selected from a group comprising: an actual species of the unidentified object, a default species that indicates that the object cannot be identified, and a species of a highest-ranked related object of the set of related objects; and
the system directing that the species label be affixed to the unidentified object such that a downstream business rule may determine how the object should be processed as a function of the species identified by the species label.
2. The system of claim 1 , wherein the received information comprises a digital image of the unidentified object received from the imaging device, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function;
the system receiving, in response to the submitting, a set of match scores, wherein each score of the set of match scores identifies a degree of similarity between the submitted image and a stored image that is known to identify a known object, and wherein the degree of similarity is a function of a number of characteristics of the submitted image that each match an analogous characteristic of the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of match scores; and
the system identifying the unidentified object as being the most acceptable known object.
3. The system of claim 2 , further comprising:
the system directing the image-matching function to update itself as a function of the selecting the most acceptable known object.
4. The system of claim 1 , wherein the received information comprises a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting, a set of attribute matches, wherein each match of the set of attribute matches identifies a known object as a function of a degree of similarity between the submitted subset and a stored set of physical attributes that is known to identify the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of attribute matches; and
the system identifying the unidentified object as being the most acceptable known object.
5. The system of claim 4 , further comprising:
the system directing the attribute-matching function to update itself as a function of the selecting the most acceptable known object.
6. The system of claim 4 ,
wherein the set of physical attributes are received from the set of electronic sensors;
wherein each sensor of the set of electronic sensors is selected from a group comprising: a weight scale, a reflectance meter, a digital-imaging device, a laser, a light curtain, an optical character reader, and a digital scanner;
wherein an attribute of the set of the physical attributes is selected from a group comprising: a weight, a mass, a linear dimension, a shape, a marking or other identifier, a color, a binding, a volume, a reflectance, a visual pattern, a texture, a trade-dress format or design, a title, a value, a cost, and a page count.
7. The system of claim 1 , wherein the received information comprises a digital image of the unidentified object and a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function and further submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting and to the further submitting, one or more messages indicating that the image-matching function and the attribute-matching function failed to identify an acceptable known object associated with the submitted information;
the system, in response to the one or more messages, requesting that an external search agency identify the unidentified object by searching extrinsic sources of information; and
the system, in response to the requesting, receiving from the external search agency an identification of the unidentified object.
8. The system of claim 7 , further comprising:
the system directing the external search agency to update an extrinsic source of information as a function of the identification of the unidentified object.
9. The system of claim 1 , wherein the correlated characteristic is selected from a group comprising a sales rank, a value, a current inventory level, a target inventory level, a publication date, a manufacture date, a title, an edition, an author, a number of sales offers received through at least one sale channel, a range of values of sales offers received through at least one sale channel, and a customer requirement.
10. The system of claim 1 , wherein the identifying comprises receiving an identification of a set of candidate objects and a set of match scores, wherein each match score indicates a degree of similarity between the unidentified object and one candidate object of the set of candidate objects, and wherein the identifying is performed as a function of the set of match scores.
11. The system of claim 1 , further comprising:
the system associating the digital image with metadata that is associated with the species of object identified by the species label, wherein the metadata is selected from a group comprising: a title, an author, a number of sales offers received through at least one sale channel, a range of values of sales offers received through at least one sale channel, an inventory level, and a customer requirement.
12. The system of claim 1 , wherein the ranking is performed as a further function of a business priority.
13. The system of claim 1 , wherein the downstream business rule is selected from a group comprising: directing the object to an area of inventory, offering the object to a customer, and discarding the object.
14. A method for automated object identification and processing based on at least one of a digital image of an object and a physical attribute of the object, the method comprising:
a computerized object-identification system receiving notice that the object has not been identified;
the system receiving information that describes an unidentified object;
the system identifying the object as a function of the received information;
the system further identifying a set of related objects as a further function of the received information;
the system ranking the related objects as a function of a value of a degree of similarity of each object of the set of related objects to the unidentified object, wherein a higher-ranked object of the set of related objects is more similar to the unidentified object than is a lower-ranked object of the set of related objects;
the system, at a time after the receiving notice, directing a labeling device to generate a unique machine-readable species label that associates the unidentified object with at least one species of object selected from a group comprising: an actual species of the unidentified object, a default species that indicates that the object cannot be identified, and a species of a highest-ranked related object of the set of related objects; and
the system directing that the species label be affixed to the unidentified object such that a downstream business rule may determine how the object should be processed as a function of the species identified by the species label.
15. The method of claim 14 , wherein the received information comprises a digital image of the unidentified object received from the imaging device, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function;
the system receiving, in response to the submitting, a set of match scores, wherein each score of the set of match scores identifies a degree of similarity between the submitted image and a stored image that is known to identify a known object, and wherein the degree of similarity is a function of a number of characteristics of the submitted image that each match an analogous characteristic of the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of match scores; and
the system identifying the unidentified object as being the most acceptable known object.
16. The method of claim 14 , wherein the received information comprises a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting, a set of attribute matches, wherein each match of the set of attribute matches identifies a known object as a function of a degree of similarity between the submitted subset and a stored set of physical attributes that is known to identify the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of attribute matches; and
the system identifying the unidentified object as being the most acceptable known object.
17. The method of claim 14 , wherein the received information comprises a digital image of the unidentified object and a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function and further submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting and to the further submitting, one or more messages indicating that the image-matching function and the attribute-matching function failed to identify an acceptable known object associated with the submitted information;
the system, in response to the one or more messages, requesting that an external search agency identify the unidentified object by searching extrinsic sources of information; and
the system, in response to the requesting, receiving from the external search agency an identification of the unidentified object.
18. A computer program product, comprising a computer-readable hardware storage device having a computer-readable program code stored therein, the program code configured to be executed by a computerized object-identification system comprising a processor, a memory coupled to the processor, an interface to an imaging device, a set of interfaces to a set of electronic sensors, and a computer-readable hardware storage device coupled to the processor, the storage device containing program code configured to be run by the processor via the memory to implement a method for automated object identification and processing based on at least one of a digital image of an object and a physical attribute of the object, the method comprising:
the computerized object-identification system receiving notice that the object has not been identified;
the system receiving information that describes an unidentified object;
the system identifying the object as a function of the received information;
the system further identifying a set of related objects as a further function of the received information;
the system ranking the related objects as a function of a value of a degree of similarity of each object of the set of related objects to the unidentified object, wherein a higher-ranked object of the set of related objects is more similar to the unidentified object than is a lower-ranked object of the set of related objects;
the system, at a time after the receiving notice, directing a labeling device to generate a unique machine-readable species label that associates the unidentified object with at least one species of object selected from a group comprising: an actual species of the unidentified object, a default species that indicates that the object cannot be identified, and a species of a highest-ranked related object of the set of related objects; and
the system directing that the species label be affixed to the unidentified object such that a downstream business rule may determine how the object should be processed as a function of the species identified by the species label.
19. The computer program product of claim 18 , wherein the received information comprises a digital image of the unidentified object received from the imaging device, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function;
the system receiving, in response to the submitting, a set of match scores, wherein each score of the set of match scores identifies a degree of similarity between the submitted image and a stored image that is known to identify a known object, and wherein the degree of similarity is a function of a number of characteristics of the submitted image that each match an analogous characteristic of the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of match scores; and
the system identifying the unidentified object as being the most acceptable known object.
20. The computer program product of claim 18 , wherein the received information comprises a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting, a set of attribute matches, wherein each match of the set of attribute matches identifies a known object as a function of a degree of similarity between the submitted subset and a stored set of physical attributes that is known to identify the known object;
the system selecting a most acceptable known object of the received known objects, wherein the selecting is performed as a function of the received set of attribute matches; and
the system identifying the unidentified object as being the most acceptable known object.
21. The computer program product of claim 18 , wherein the received information comprises a digital image of the unidentified object and a set of physical attributes of the unidentified object, and wherein the identifying the object comprises:
the system submitting the digital image to an image-matching function and further submitting a subset of the set of physical attributes to an attribute-matching function;
the system receiving, in response to the submitting and to the further submitting, one or more messages indicating that the image-matching function and the attribute-matching function failed to identify an acceptable known object associated with the submitted information;
the system, in response to the one or more messages, requesting that an external search agency identify the unidentified object by searching extrinsic sources of information; and
the system, in response to the requesting, receiving from the external search agency an identification of the unidentified object.