Scanning labels to extract and store structured information

1. A system, comprising:

a communication interface; and

a processor coupled to the communication interface and configured to:

receive a video stream of a food or beverage package comprising a plurality of image frames;

recognize in each of at least a subset of the image frames a line of text comprising a nutritional ingredient and a corresponding frame-specific quantity, wherein recognizing comprises mapping a structural context element based on a standard;

wherein the standard includes at least one of the following: an administrative standard, a regulatory standard, a legal standard, a business standard, and a de facto standard;

wherein the structural context element comprises an element associated with the standard based at least in part on one of following: color, number of colors, columnar format, typeface, type size, relative position on label, number of occurrences, rounding rules, keywords, serving size, standards of identity, and % Daily Value correlation;

preprocess the respective corresponding frame-specific quantities using the structural context element; and

determine a reported quantity for the nutritional ingredient based at least in part on a statistical confidence score based at least in part on aggregation of the preprocessed frame-specific quantities.

2. The system of claim 1 , wherein recognizing comprises using optical character recognition (OCR) analysis.

3. The system of claim 2 , wherein OCR analysis comprises using at least one of the following: naive dictionary text matching, Levenshtein algorithm, autocorrection, and regexp matching.

4. The system of claim 2 , wherein OCR analysis comprises using % Daily Value correlation and rounding rules.

5. The system of claim 1 , wherein the processor is further configured to receive a server OCR analysis and wherein recognizing comprises using the server OCR analysis.

6. The system of claim 1 , wherein determining comprises assessing a confidence score.

7. The system of claim 6 , wherein the confidence score is based at least in part on a dimensional order.

8. The system of claim 6 , wherein the confidence score is based at least in part on a bounding box analysis.

9. The system of claim 1 , further comprising a camera.

10. The system of claim 9 , wherein the processor is further configured to coach a user on repositioning the camera for a second plurality of image frames.

11. The system of claim 1 , wherein the processor is further configured to transmit the reported quantity for the nutritional ingredient to a mobile app server.

12. The system of claim 1 , wherein the nutritional element is at least one of: macronutrients, list of ingredients, calories, serving size, micronutrients, and allergens.

13. The system of claim 1 , wherein the plurality of image frames is received from a mobile app client.

14. A method, comprising:

receiving a video stream of a food or beverage package comprising a plurality of image frames;

recognizing in each of at least a subset of the image frames a line of text comprising a nutritional ingredient and a corresponding frame-specific quantity, wherein recognizing comprises mapping a structural context element based on a standard;

wherein the standard includes at least one of the following: an administrative standard, a regulatory standard, a legal standard, a business standard, and a de facto standard;

wherein the structural context element comprises an element associated with the standard based at least in part on one of following: color, number of colors, columnar format, typeface, type size, relative position on label, number of occurrences, rounding rules, keywords, serving size, standards of identity, and % Daily Value correlation;

preprocessing the respective corresponding frame-specific quantities using the structural context element; and

determining a reported quantity for the nutritional ingredient based at least in part on a statistical confidence score based at least in part on aggregation of the preprocessed frame-specific quantities.

15. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a video stream of a food or beverage package comprising a plurality of image frames;

recognizing in each of at least a subset of the image frames a line of text comprising a nutritional ingredient and a corresponding frame-specific quantity, wherein recognizing comprises mapping a structural context element based on a standard;

wherein the standard includes at least one of the following: an administrative standard, a regulatory standard, a legal standard, a business standard, and a de facto standard;

wherein the structural context element comprises an element associated with the standard based at least in part on one of following: color, number of colors, columnar format, typeface, type size, relative position on label, number of occurrences, rounding rules, keywords, serving size, standards of identity, and % Daily Value correlation;

preprocessing the respective corresponding frame-specific quantities using the structural context element; and

determining a reported quantity for the nutritional ingredient based at least in part on a statistical confidence score based at least in part on aggregation of the preprocessed frame-specific quantities.