Reduction of incorrectly identified document scanning defects
View Patent ↗In general, this disclosure describes techniques of reducing the possibility of an image evaluation device incorrectly identifying defects in images of documents. Some defects may be apparent in the images of both sides of a document. For instance, a tear in a document could be apparent in an image of a front side and an image of a rear side of the document. However, the image evaluation device could erroneously identify such a tear in an image of one side of the document. In this case, the tear would not be apparent in the image of the other side of the document. To reduce the possibility that the image evaluation device erroneously identifies such a defect, the image evaluation device may determine whether defects identified in the image of the front side of the document correspond to defects identified in the image of the rear side of the document.
1. A method comprising:
receiving a first image, wherein the first image is a digital image of a front surface of a document;
receiving a second image, wherein the second image is a digital image of a rear surface of the document;
identifying defects in the first image that could hinder use of the first image and the second image to process the document in an automated document processing system;
identifying defects in the second digital image that could hinder use of the first image and the second image to process the document in the automated document processing system;
determining whether a first defect corresponds to any of the defects in the second image, wherein the first defect is one of the identified defects in the first image; and
outputting data that indicates that the first image and the second image include a defect that could hinder use of the first image and the second image in the automated document processing system when the first defect corresponds to one of the defects in the second image.
2. The method of claim 1 , further comprising causing a digital version of the document to be used in the automated document processing system when the data indicates that the first defect does not correspond to any of the defects in the second image, wherein the digital version of the document includes the first image and the second image.
3. The method of claim 2 ,
wherein the document is a check;
wherein the automated document processing system is an automated check clearing system;
wherein receiving the first image comprises receiving the first image from an enterprise;
wherein receiving the second image comprises receiving the second image from the enterprise; and
wherein causing the digital version of the document to be used in the automated document processing system comprises submitting the digital version of the check to a check processing authority.
4. The method of claim 2 , wherein the document is a type of document selected from a group consisting of: forms, remittance slips, legal documents, tax returns, orders, and invoices.
5. The method of claim 1 ,
wherein the method further comprises:
performing one or more graphics processing operations on the first image prior to identify the defects in the first image; and
performing one or more of the graphics processing operations on the second image prior to identifying the defects in the second image; and
wherein the graphics processing operations are selected from a group that consists of: a compression operation, a despeckling operation, and a deskewing operation.
6. The method of claim 1 ,
wherein identifying defects in the first image comprises:
using the first image to determine whether the document includes a document deficiency; and
identifying the first defect when it is determined using the first image that the document includes the document deficiency;
wherein identifying defects in the second image comprises:
using the second image to determine whether the document includes the document deficiency; and
identifying a second defect when it is determined using the second image that the document includes the document deficiency; and
wherein determining whether the first defect corresponds to any of the defects in the second image comprises determining whether the first defect and the second defect are associated with corresponding parts of the document,
wherein the document deficiency belongs to a group of document deficiencies that consists of: a fold and a tear.
7. The method of claim 1 ,
wherein identifying defects in the first image comprises:
determining whether the first image is excessively skewed; and
identifying the first defect when it is determined that the first image is excessively skewed;
wherein identifying defects in the second image comprises:
determining whether the second image is excessively skewed; and
identifying a second defect when it is determined that the second image is excessively skewed; and
wherein determining whether the first defect corresponds to any of the defects in the second image comprises determining whether both the first defect and the second defect are identified.
8. The method of claim 1 ,
wherein identifying defects in the first image comprises identifying a non-symmetric defect in the first image; and
wherein the method further comprises:
raising a flag associated with the non-symmetric defect, wherein the flag is a flag in a set of flags defined by a standards institution; and
determining whether any flags in the set of flags have been raised; and
wherein outputting data comprises outputting the flag associated with the non-symmetric defect.
9. A system that comprises an image evaluation device that includes:
a device comprising:
a front defect detection module that identifies defects in a first image that could hinder use of a first image to process a document in an automated electronic document processing system, wherein the first image is a digital image of a front surface of the document;
a rear defect detection module that identifies defects in a second image that could hinder use of the second image to process the document in the automated electronic document processing system, wherein the second image is a digital image of a rear surface of the document;
a defect correspondence module that determines whether a first defect corresponds to any of the defects in the second image, wherein the first defect is one of the identified defects in the first image; and
an output module that outputs data that indicates that the first image and the second image include a defect that could hinder use of the first image and the second image in the automated document processing system when the first defect corresponds to one of the defects in the second image.
10. The system of claim 9 , further comprising a storage device that causes a digital version of the document to be used in the automated electronic document processing system when the data indicates that the first defect does not correspond to any of the defects in the second image, wherein the digital version of the document includes the first image and the second image.
11. The system of claim 10 ,
wherein the document is a check;
wherein the automated document processing system is an automated check clearing system;
wherein the image evaluation device further comprises a network interface that receives the digital version of the check from a financial institution; and
wherein the image verification module causes the digital version of the check to be used in the automated check clearing process by performing the automated check clearing process on the digital version of the check.
12. The system of claim 10 ,
wherein the document is a check;
wherein the automated document processing system is an automated check clearing system; and
wherein the storage device further comprises:
a digital version module that generates the digital version of the check; and
a network interface that automatically transmits the digital version of the check to a financial institution that uses the digital version of the check to perform the automated check clearing process.
13. The system of claim 9 , wherein the system further comprises an image capture device that comprises:
a front image processing module that performs one or more graphics processing operations on the first image before the front defect detection module identifies the defects in the first image; and
a rear image processing module that performs one or more graphics processing operations on the second image before the rear defect detection module identifies the defects in the second image,
wherein the graphics processing operations are selected from a group that consists of: a compression operation, a despeckling operation, a normalization operation, a grayscale remapping operation, a thresholding operation, and a deskewing operation.
14. The system of claim 9 ,
wherein the front defect detection module uses the first image to determine whether the document includes a document deficiency and identifies the first defect when the front defect detection module determines that the document includes the document deficiency;
wherein the rear defect detection module uses the second image to determine whether the document includes the document deficiency and identifies a second defect when the rear defect detection module determines that the document includes the document deficiency; and
wherein in order to determine whether the first defect corresponds to any of the defects in the second image, the defect correspondence module determines whether the first defect and the second defect are associated with corresponding parts of the document,
wherein the document deficiency belongs to a group of document deficiencies that consists of: a fold and a tear.
15. The system of claim 9 ,
wherein the front defect detection module determines whether the first image is excessively skewed aid identifies the first defect when it is determined that the first image is excessively skewed;
wherein the rear defect detection module determines whether the second image is excessively skewed and identifies a second defect when it is determined that the second image is excessively skewed; and
wherein in order to determine whether the first defect corresponds to any of the defects in the second image, the defect correspondence module determines whether the first defect and the second defect are identified.
16. The system of claim 9 ,
wherein the front defect detection module identifies a non-symmetric defect in the first image;
wherein the defect correspondence module raises a flag associated with the non-symmetric defect, wherein the flag is a flag in a set of flags defined by a standards institution; and
wherein the output module outputs the set of flags, thereby indicating that the first image includes a defect that could binder use of the first image and the second image in the automated document processing system.
17. A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:
receive a first image, wherein the first image is a digital image of a front surface of a document;
receive a second image, wherein the second image is a digital image of a rear surface of the document;
identify defects in the first image that could hinder use of the first image and the second image to process the document in an automated document processing system;
identify defects in the second digital image that could hinder use of the first image and the second image to process the document in the automated document processing system;
determine whether a first defect corresponds to any of the defects in the second image, wherein the first defect is one of the identified defects in the first image; and
generate data that indicates that the first image and the second image include a defect that could hinder use the first image and the second image in the automated document processing system when the first defect corresponds to one of the defects in the second image.
18. The computer-readable medium of claim 17 , wherein the instructions further cause the one or more processors to automatically transmit the digital version of the document to a financial institution that uses the digital version of the document to perform the automated document processing system.
19. The computer-readable medium of claim 17 ,
wherein the instructions cause the one or more processors to identify defects in the first image at least in part by causing the one or more processors to:
use the first image to determine whether the document includes a document deficiency; and
identify the first defect when it is determined using the first image that the document includes the document deficiency;
wherein the instructions cause the one or more processors to identify defects in the second image at least in part by causing the one or more processors to:
use the second image to determine whether the document includes the document deficiency; and
identify a second defect when it is determined using the second image that the document includes the document deficiency; and
wherein the instructions cause the one or more processors to determine whether the first defect corresponds to any of the defects in the second image at least in part by causing the one or more processors to determine whether the first defect and the second defect are associated with corresponding parts of the document
wherein the document deficiency belongs to a group of document deficiencies that consists of: a fold and a tear.
20. The computer-readable medium of claim 17 ,
wherein the instructions cause the one or more processors to identify defects in the first image at least in part by causing the one or more processors to:
determine whether the first image is excessively skewed; and
identify the first defect when it is determined that the first image is excessively skewed;
wherein the instructions cause the one or more processors to identify defects in the second image at least in part by causing the one or more processors to:
determine whether the second image is excessively skewed; and
identify a second defect when it is determined that the second image is excessively skewed; and
wherein the instructions cause the one or more processors to determine whether the first defect corresponds to any of the defects in the second image at least in part by causing the one or more processors to determine whether both the first defect and the second defect are identified.