Identifying desynchronization between representations of data obtained from an ordered plurality of documents
View Patent ↗In a method for identifying desynchronization between representations of data obtained from an ordered plurality of documents, a processor can receive ordered first and second pluralities of data strings obtained from the respective plurality of documents; compare each data string in the first plurality to the corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality; based on the comparison, designate each data string in the first plurality as being one of synchronized, leading, or trailing; identify a continuous sequence of N data strings in the first plurality that all have a designation of leading or all have a designation of trailing, where N equals or exceeds a specified sequence threshold; and generate a single error signal that identifies all N of the data strings in the continuous sequence as being desynchronized.
1. A method, comprising:
receiving ordered first and second pluralities of data strings obtained from an ordered plurality of documents, the ordered first plurality of data strings obtained from an optical scanner, the ordered second plurality of data strings obtained from a magnetic scanner;
comparing each data string in the first plurality to a corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality;
based on the comparing, designating each data string in the first plurality as being one of unknown, synchronized, leading, or trailing;
when a continuous sequence of N data strings in the first plurality have a same designation as leading or trailing, generating an error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized; and
making only documents associated with data strings in the first plurality that are synchronized available for viewing.
2. The method of claim 1 , further comprising:
receiving an ordered plurality of images of the respective ordered plurality of documents; and
performing optical character recognition on the ordered plurality of images to generate the ordered first plurality of data strings.
3. The method of claim 2 , further comprising:
magnetically reading a magnetic ink recognition line from each of the plurality of documents to generate the second plurality of data strings; and
imaging at least the magnetic ink recognition line from each of the plurality of documents to form the plurality of images.
4. The method of claim 1 , wherein designating each data string in the first plurality as being one of synchronized, leading, or trailing comprises:
designating each data string in the first plurality as being one of:
synchronized, when the data string in the first plurality matches the corresponding data string in the second plurality;
leading, when the data string in the first plurality matches a data string sequentially before the corresponding data string in the second plurality; or
trailing, when the data string in the first plurality matches a data string sequentially after the corresponding data string in the second plurality.
5. The method of claim 1 , wherein comparing each data string in the first plurality to the corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality comprises:
for each data string in the first plurality:
calculating respective Levenshtein distances between said data string in the first plurality and the corresponding data string in the second plurality, and between said first data string in the first plurality and each data string sequentially before or sequentially after the corresponding data string in the second plurality;
selecting the lowest of the calculated Levenshtein distances;
selecting the data string of the second plurality corresponding to the selected Levenshtein distance; and
designating the selected data string of the second plurality as matching said data string of the first plurality.
6. The method of claim 5 , further comprising:
designating the selected data string of the second plurality as being unknown if the lowest of the calculated Levenshtein distances equals or exceeds a specified Levenshtein distance threshold.
7. The method of claim 6 , further comprising:
identifying a continuous sequence of N data strings in the first plurality that all have a designation of leading or unknown, or all have a designation of trailing or unknown, where N equals or exceeds the specified sequence threshold.
8. The method of claim 1 , wherein generating an error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized comprises:
generating a single error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized.
9. The method of claim 1 , wherein N equals or exceeds a specified sequence threshold.
10. A method, comprising:
imaging each of an ordered plurality of documents to form a respective ordered plurality of images;
performing optical character recognition on the ordered plurality of images to generate an ordered first plurality of data strings;
magnetically reading a magnetic ink recognition line from each of the plurality of documents to generate an ordered second plurality of data strings;
comparing each data string in the first plurality to a corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality;
based on the comparing, designating each data string in the first plurality as being one of unknown, synchronized, leading, or trailing;
when a continuous sequence of N data strings in the first plurality have a same designation as leading or trailing, generating an error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized, where N equals or exceeds a specified sequence threshold; and
making only images associated with data strings in the first plurality that synchronized available for viewing.
11. The method of claim 10 , wherein designating each data string in the first plurality as being one of synchronized, leading, or trailing comprises:
designating each data string in the first plurality as being one of:
synchronized, if the data string in the first plurality matches the corresponding data string in the second plurality;
leading, if the data string in the first plurality matches a data string sequentially before the corresponding data string in the second plurality; or
trailing, if the data string in the first plurality matches a data string sequentially after the corresponding data string in the second plurality.
12. The method of claim 10 , wherein comparing each data string in the first plurality to the corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality comprises:
for each data string in the first plurality:
calculating respective Levenshtein distances between said data string in the first plurality and the corresponding data string in the second plurality, and between said first data string in the first plurality and each data string sequentially before or sequentially after the corresponding data string in the second plurality;
selecting the lowest of the calculated Levenshtein distances;
selecting the data string of the second plurality corresponding to the selected Levenshtein distance; and
designating the selected data string of the second plurality as matching said data string of the first plurality.
13. The method of claim 12 , further comprising:
designating the selected data string of the second plurality as being unknown if the lowest of the calculated Levenshtein distances equals or exceeds a specified Levenshtein distance threshold.
14. The method of claim 13 , further comprising:
identifying a continuous sequence of N data strings in the first plurality that all have a designation of leading or unknown, or all have a designation of trailing or unknown, where N equals or exceeds the specified sequence threshold.
15. The method of claim 10 , wherein generating an error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized comprises:
generating a single error signal associated with the first and second pluralities identifying all N of the data strings in the continuous sequence as being desynchronized.
16. The method of claim 10 , wherein the specified sequence threshold equals three (3).
17. A system, comprising:
an optical scanner configured to capture an ordered plurality of images from a respective ordered plurality of documents;
a processor configured to perform optical character recognition on the ordered plurality of images to generate a respective ordered first plurality of data strings; and
a magnetic scanner configured to magnetically read a magnetic ink recognition line from each of the plurality of documents to generate a respective ordered second plurality of data strings;
wherein the processor is further configured to:
compare each data string in the first plurality to the corresponding data string in the second plurality and to each data string sequentially before or sequentially after the corresponding data string in the second plurality;
based on the comparison, designate each data string in the first plurality as being one of unknown, synchronized, leading, or trailing;
identify a continuous sequence of N data strings in the first plurality that all have a designation of leading or all have a designation of trailing, where N equals or exceeds a specified sequence threshold;
generate a single error signal that identifies all N of the data strings in the continuous sequence as being desynchronized; and
making only images associated with data strings in the first plurality that synchronized available for viewing.
18. The system of claim 17 , wherein the processor is further configured to:
designate each data string in the first plurality as being one of:
synchronized, if the data string in the first plurality matches the corresponding data string in the second plurality,
leading, if the data string in the first plurality matches a data string sequentially before the corresponding data string in the second plurality; or
trailing, if the data string in the first plurality matches a data string sequentially after the corresponding data string in the second plurality.
19. The system of claim 17 , wherein the processor is further configured to:
for each data string in the first plurality:
calculate respective Levenshtein distances between said data string in the first plurality and the corresponding data string in the second plurality, and between said first data string in the first plurality and each data string sequentially before or sequentially after the corresponding data string in the second plurality;
select the lowest of the calculated Levenshtein distances;
select the data string of the second plurality corresponding to the selected Levenshtein distance; and
designate the selected data string of the second plurality as matching said data string of the first plurality.
20. The system of claim 19 , wherein the processor is further configured to:
designate the selected data string of the second plurality as being unknown if the lowest of the calculated Levenshtein distances equals or exceeds a specified Levenshtein distance threshold; and
identify a continuous sequence of N data strings in the first plurality that all have a designation of leading or unknown, or all have a designation of trailing or unknown, where N equals or exceeds the specified sequence threshold.