Document image orientation, assessment and correction

• US 10,019,772 B1
• Filed: 10/06/2017
• Issued: 07/10/2018
• Est. Priority Date: 10/06/2017
• Status: Active Grant

• Current Assignee: Vertifi Software, LLC
• Original Assignee: Vertifi Software, LLC
• Inventors: Smith, Christopher Edward
• Primary Examiner(s): Krasnic, Bernard
• Application Number:
  US15/727,051
• Time in Patent Office: 277 Days
• Field of Search: None
• CPC Class Codes: G06F 16/51 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06Q 40/128 Check-book balancing, updating or printing arrangements
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06T 3/40 Scaling of whole images or parts thereof, e.g. expanding or contracting
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: G06V 10/242 by image rotation, e.g. by 90 degrees
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: G06V 30/412 Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text

Abstract

The present invention relates to methods and systems for automated processing of document images using mobile and desktop communications devices. Systems and methods are provided for expediting analysis of the accuracy and quality of captured images by such devices before transmission to a server for processing. The processes use on-device software to provide immediate feedback to the user as to whether the quality of the document photograph is sufficient for processing, without the need for intermediate communication with a server. The processes utilize component labeling, MICR character recognition methods, automatic image rotation, and weighted image scaling. The invention increases speed of processing, reduces rates of rejection, increases processor efficiency, limits memory use, identifies images with likely errors, and provides the opportunity for correction of errors at the earliest possible time.

Claims

1. A method for scaling an image of a document consisting of a plurality of pixels, said scaling comprising:

a. acquiring on a device an image of a document consisting of a plurality of pixels;

b. determining a luminosity of a plurality of the pixels;

c. creating distinctive label groups by assigning a unique label to each group of contiguous locations of pixels of a first specified luminosity, whereby each pixel location in a group has a common label;

d. selecting a specified set of known characters of a defined font, each said known character comprising a set of character defining pixels of known relative location and known luminosity;

e. selecting a second specified luminosity, wherein said second specified luminosity may be the same as or different from said first specified luminosity;

f. for each pixel location within each said distinctive label group, determining the luminosity of the character defining pixel at the similar relative pixel location within each set of character defining pixels;

g. for each pixel location within each said distinctive label group and for each set of character defining pixels, determining whether each said pixel location matches the similar relative pixel location of a character defining pixel of a second specified luminosity within each set of character defining pixels;

h. determining a number of said matches for each distinctive label group and each known character;

i. for each distinctive label group with a said match in excess of a threshold number, assigning to the distinctive label group the known character with a highest number of said matches;

j. calculating a matching score for each distinctive label group with an assigned known character by determining a proportion of pixel locations within the distinctive label group with a label-candidate character pixel match;

k. calculating a height of each distinctive label group with an assigned known character;

l. weighing said calculated height of each said distinctive label group by its said matching score;

m. determining an average weighted height of all distinctive label groups with an assigned known character using said calculated weighted heights;

n. determining a width of the document image based upon said average weighted height of all distinctive label groups with an assigned known character; and

o. scaling a size of the document image based upon said average weighted height.