Mobile check deposit
Methods and systems for remote check deposit are disclosed. A check for deposit is processed without the need for a server to receive any image of the check initially. Instead, optical character recognition (OCR) data is received at the server from a mobile device. Verification processing for the check is then performed using the OCR data. If the verification process is successful, a confirmation notification is sent to the mobile device. Subsequently, after sending the confirmation notification, a check image is received, from which the OCR data was determined. The check is, in turn, processed for deposit using the received check image.
1 . A method for processing an image of an object, the method comprising:
receiving an object image of an object, the object image captured by an image capture device of a mobile device, wherein the object image includes:
a first region having a first set of characters of a first font; and
a second region having a second set of characters of a second font;
obtaining a second region snippet that includes data corresponding to the second region in the object image;
isolating the second set of characters within the second region from a background, wherein the isolating includes:
calculating a vertical projection of the second region snippet; and
calculating a horizontal projection of the second region snippet;
generating one or more character snippets from the isolated second set of characters;
determining a string of characters based on determining a character represented by each of the one or more character snippets, wherein the determining the character includes comparing each of the one or more character snippets to one or more templates; and
returning the string of characters to a calling function.
2 . The method of claim 1 ,
wherein calculating the vertical projection includes summing intensity values along columns within the second region snippet; and
wherein calculating the horizontal projection includes summing intensity values across rows within the second region snippet.
3 . The method of claim 1 , wherein isolating the second set of characters further includes applying a vertical mask based on the vertical projection and applying a horizontal mask based on the horizontal projection.
4 . The method of claim 3 ,
wherein applying a vertical mask includes, for each respective summed intensity for a respective column, masking off that respective column responsive to the respective summed intensity failing to satisfy a vertical masking threshold; and
wherein applying a horizontal mask includes, for each respective summed intensity for a respective row, masking off that respective row responsive to the respective summed intensity failing to satisfy a horizontal masking threshold.
5 . The method of claim 4 , further comprising:
estimating an amount of noise in the second region snippet or the object image; and
selecting or modifying the vertical masking threshold and the horizontal masking threshold based on the estimated amount of noise.
6 . The method of claim 1 , further comprising:
after isolating the second set of characters, performing one or more post-projection processing steps selected from the group consisting of: applying an erosion operation, applying a dilation operation, applying a normalization operation including scaling intensities or adjusting dynamic range, and applying a histogram equalization operation to enhance contrast.
7 . The method of claim 1 , wherein comparing each of the one or more character snippets to one or more templates includes:
for each respective character snippet of the one or more character snippets:
for each respective template of the one or more templates:
calculating an exclusive-or operation between the respective character snippet and the respective template; and
summing the resulting values to form a difference score;
identifying a character-template pair having the lowest difference score, wherein the character-template pair comprises the respective character snippet and one of the one or more templates;
assigning to the respective character snippet a character associated with the template of the character-template pair.
8 . The method of claim 7 , further comprising:
calculating an acceptance ratio based on a dividing the difference score associated with the character-template pair by a sum of values in the character snippet of the character-template pair.
9 . The method of claim 1 , wherein determining the character represented by each of the one or more character snippets includes determining an acceptance ratio.
10 . The method of claim 9 , wherein responsive to the acceptance ratio failing to satisfy a threshold, performing a remediation.
11 . The method of claim 9 , wherein responsive to the acceptance ratio satisfying the threshold, selecting a processing path based on whether the determined character is a digit or a special character.
12 . The method of claim 1 , wherein comparing each of the one or more character snippets to one or more templates includes:
generating a conforming outer perimeter of a respective character within a respective character snippet; and
comparing the conforming outer perimeter of the respective character with one or more templates of a set of template conforming outer perimeters for each integer zero through nine of a specific font.
13 . The method of claim 1 , wherein obtaining the second region snippet includes cropping the object image to the second region to form the second region snippet.
14 . The method of claim 1 , wherein the object is a check.
15 . The method of claim 1 , wherein the object is an identification card.
16 . The method of claim 1 , further comprising:
determining a number of characters present in the second region based on a length of the string;
determining a length of the second region in pixels;
using the number of characters present in the second region and the length of the second region in pixels to calculate a scaling factor; and
scaling the object image based on the scaling factor.
17 . The method of claim 16 , further comprising artificially increasing the number of characters by one.
18 . The method of claim 1 , wherein the method is performed entirely on the mobile device before the object image is sent to a server for processing.
19 . A non-transitory computer readable medium having a string stored thereon, wherein the string was created by a process including executing instructions with a set of one or more processors that causes the set of one or more processors to:
isolate a set of characters within an image, wherein to isolate the set includes to:
calculate a vertical projection, wherein calculating the vertical projection includes summing intensity values along columns within the image;
calculate a horizontal projection, wherein calculating the horizontal projection includes summing intensity values across rows within the image;
apply a vertical mask based on the vertical projection, wherein applying the vertical mask includes, for each respective summed intensity for a respective column, masking off that respective column responsive to the respective summed intensity failing to satisfy a vertical masking threshold; and
apply a horizontal mask based on the horizontal projection, wherein applying a horizontal mask includes, for each respective summed intensity for a respective row, masking off that respective row responsive to the respective summed intensity failing to satisfy a horizontal masking threshold;
generate one or more character snippets from the isolated set of characters; and
determine the string based on determining a character represented by each of the one or more character snippets, wherein the determining includes comparing each of the one or more character snippets to one or more templates.
20 . An optical character recognition method comprising:
isolating a set of characters within an image, wherein the isolating includes:
calculating a vertical projection, wherein calculating the vertical projection includes summing intensity values along columns within the image;
calculating a horizontal projection, wherein calculating the horizontal projection includes summing intensity values across rows within the image;
applying a vertical mask based on the vertical projection, wherein applying the vertical mask includes, for each respective summed intensity for a respective column, masking off that respective column responsive to the respective summed intensity failing to satisfy a vertical masking threshold; and
applying a horizontal mask based on the horizontal projection, wherein applying a horizontal mask includes, for each respective summed intensity for a respective row, masking off that respective row responsive to the respective summed intensity failing to satisfy a horizontal masking threshold;
generating one or more character snippets from the isolated set of characters; and
determining a string of characters based on determining a character represented by each of the one or more character snippets, wherein the determining includes comparing conforming outer perimeters of each of the one or more character snippets to one or more template conforming outer perimeters.