Methods, systems, apparatus and articles of manufacture for receipt decoding

1. An apparatus for processing a receipt associated with a user, the apparatus comprising:

optical character recognition circuitry to generate bounding boxes including a first bounding box, a second bounding box and a third bounding box, respective ones of the bounding boxes associated with groups of characters detected in the receipt, locations of the bounding boxes defined by a set of coordinates;

word connecting circuitry to:

connect the first bounding box to the second bounding box based on (1) an adjacency of the first bounding box to the second bounding box and (2) a difference value from a comparison of a location of the first bounding box to a location of the second bounding box, the comparison including determining whether the location of the first bounding box and the location of the second bounding box satisfy a connection criterion;

in response to determining the first bounding box and the second bounding box satisfy the connection criterion, move the first bounding box and the second bounding box into a list of connected words; and

in response to determining the first bounding box and the second bounding box do not satisfy the connection criterion, move the first bounding box and the second bounding box into a list of non-connected words;

line connecting circuitry to form a line of the bounding boxes by connecting the third bounding box to the second bounding box based on a relationship between the first bounding box and the second bounding box, the line of the bounding boxes indicative of related receipt fields; and

creditor circuitry to generate a report based on the line.

2. The apparatus of claim 1 , wherein the optical character recognition circuitry is to define the location of the first bounding box via a first set of coordinates and the location of the second bounding box via a second set of coordinates.

3. The apparatus of claim 1 , further including range determining circuitry to determine a quantity of the bounding boxes in range of the second bounding box.

4. The apparatus of claim 3 , wherein the line connecting circuitry is to, in response to the range determining circuitry determining the quantity is two:

determine a first association between the first bounding box and the third bounding box and a second association between the third bounding box and the second bounding box; and

connect the second bounding box to the third bounding box based on the first association and the second association.

5. The apparatus of claim 3 , wherein the line connecting circuitry is to, in response to the range determining circuitry determining the quantity is two:

determine a spatial relationship of the first bounding box and the second bounding box; and

connect the second bounding box to the third bounding box based on the spatial relationship.

6. A method for processing a receipt associated with a user, the method comprising:

generating, by executing an instruction with at least one processor, bounding boxes, the bounding boxes including a first bounding box, a second bounding box and a third bounding box, respective ones of the bounding boxes associated with groups of characters detected in the receipt, locations of the bounding boxes defined by a set of coordinates;

connecting, by executing an instruction with the at least one processor, the first bounding box to the second bounding box based on (1) an adjacency of the first bounding box to the second bounding box and (2) a difference value from a comparison of a location of the first bounding box to a location of the second bounding box, the comparison corresponding to satisfaction of a connection criterion;

in response to determining the first bounding box and the second bounding box satisfy the connection criterion, moving the first bounding box and the second bounding box into a list of connected words;

in response to determining the first bounding box and the second bounding box do not satisfy the connection criterion, moving the first bounding box and the second bounding box into a list of non-connected words;

forming, by executing an instruction with the at least one processor, a line of the bounding boxes by connecting the third bounding box to the second bounding based on a relationship between the first bounding box and the second bounding box, the line of the bounding boxes indicative of related receipt fields; and

generating, by executing an instruction with the at least one processor, a report based on the line.

7. The method of claim 6 , wherein the location of the first bounding box is defined by a first set of coordinates and the location of the second bounding box is defined by a second set of coordinates.

8. The method of claim 6 , further including determining a quantity of the bounding boxes in range of the second bounding box.

9. The method of claim 8 , wherein (1) the quantity is two and (2) forming a line of bounding boxes by connecting the third bounding box to the second bounding based on a relationship between the first bounding box and the second bounding box includes:

determining a first association between the first bounding box and the third bounding box and a second association between the third bounding box and the second bounding box; and

connecting the second bounding box to the third bounding box based on the first association and the second association.

10. The method of claim 8 , wherein (1) the quantity is two and (2) forming a line of bounding boxes by connecting the third bounding box to the second bounding box based on a relationship between the first bounding box and the second bounding box includes:

determining a spatial relationship of the first bounding box and the second bounding box; and

connecting the second bounding box to the third bounding box based on the spatial relationship.

11. A non-transitory computer readable medium including instructions, which when executed, cause a processor to at least:

generate bounding boxes, the bounding boxes including a first bounding box, a second bounding box and a third bounding box, respective ones of the bounding boxes associated with groups of characters detected in a receipt, a location of the bounding boxes defined by a set of coordinates;

connect the first bounding box to the second bounding box based on (1) an adjacency of the first bounding box to the second bounding box and (2) a difference value from a comparison of a location of the first bounding box to a location of the second bounding box, the comparison including determining if the first set of coordinates and the second set of coordinates satisfy a connection criterion;

in response to determining the first set of coordinates and the second set of coordinates satisfy the connection criterion, move the first set of coordinate and the second set of coordinates to a list of connected words;

in response to determining the first set of coordinates and the second set of coordinates do not satisfy the connection criterion, moving the first set of coordinate and the second set of coordinates to a list of non-connected words;

form a line of the bounding boxes by connecting the third bounding box to the second bounding based on a relationship between the first bounding box and the second bounding box, the line of the bounding boxes indicative of related receipt fields; and

generate a report based on the line.

12. The non-transitory computer readable medium of claim 11 , wherein the location of the first bounding box is defined by a first set of coordinates and the location of the second bounding box is defined by a second set of coordinates.

13. The non-transitory computer readable medium of claim 11 , wherein the instructions further cause the processor to determine a quantity of the bounding boxes in range of the second bounding box.

14. The non-transitory computer readable medium of claim 13 , wherein the quantity is two and the instructions further cause the processor to

determine a spatial relationship of the first bounding box and the second bounding box; and

connect the second bounding box to the third bounding box based on the spatial relationship.

15. The apparatus of claim 1 , further including bounding box correcting circuitry to:

determine if one of the bounding boxes is missing a coordinate of the set of coordinates; and

in response to determining one of the bounding boxes is missing the coordinate, assign a default value to the coordinate.

16. The method of claim 6 , further including correcting the respective bounding boxes by:

determining, by executing an instruction with the at least one processor, if one of the bounding boxes is missing a coordinate of the set of coordinates; and

in response to determining one of the bounding boxes is missing the coordinate, assigning, by executing an instruction with the at least one processor, a default value to the coordinate.