Optical character recognition utilizing hashed templates

A computing device for identifying a template associated with an image, the computing device comprising:

• a memory;
• and
• a processor configured to:
• receive a digital image, wherein a portion of the digital image depicts an entire first document including a plurality of bordered text sections comprising respective text values, and wherein at least one of the plurality of bordered text sections is adjacent to a text label;
• extract the portion of the digital image depicting the entire first document;
• remove the text values from the plurality of bordered text sections in the extracted portion of the digital image;
• generate a scaled image by scaling the portion of the digital image from a first resolution to a second resolution, wherein the second resolution is less than the first resolution, and wherein the scaled image includes the plurality of bordered text sections and the text label, but does not include the text values;
• generate a first hash from the scaled image;
• compare the first hash to a set of hashes, each hash of the set of hashes corresponding to a template of a set of templates, wherein each template of the set of templates is associated with data indicative of a size and location of one or more sections designated for text of a corresponding document;
• select a first template as corresponding to the first document based on the first hash best matching one of the hashes of the set of hashes that corresponds to the first template;
• extract one or more sections of the portion of the digital image based on the selected first template comprising data indicative of a size and location of the one or more sections;
• and
• perform optical character recognition on the extracted one or more sections of the portion of the digital image.

A non-transitory computer-readable medium comprising instructions that when executed by a computing device cause the computing device to perform a method for identifying a template associated with an image, the method comprising:

• receiving a digital image, wherein a portion of the digital image depicts an entire first document including a plurality of bordered text sections comprising respective text values, and wherein at least one of the plurality of bordered text sections is adjacent to a text label;
• extracting the portion of the digital image depicting the entire first document;
• removing the text values from the plurality of bordered text sections in the extracted portion of the digital image;
• generating a scaled image by scaling the portion of the digital image from a first resolution to a second resolution, wherein the second resolution is less than the first resolution, and wherein the scaled image includes the plurality of bordered text sections and the text label, but does not include the text values;
• generating a first hash from the scaled image;
• comparing the first hash to a set of hashes, each hash of the set of hashes corresponding to a template of a set of templates, wherein each template of the set of templates is associated with data indicative of a size and location of one or more sections designated for text of a corresponding document;
• selecting a first template as corresponding to the first document based on the first hash best matching one of the hashes of the set of hashes that corresponds to the first template;
• extracting one or more sections of the portion of the digital image based on the selected first template comprising data indicative of a size and location of the one or more sections;
• and
• performing optical character recognition on the extracted one or more sections of the portion of the digital image.

A computing device for identifying a template associated with an image, the computing device comprising:

• means for receiving a digital image, wherein a portion of the digital image depicts an entire first document including a plurality of bordered text sections comprising respective text values, and wherein at least one of the plurality of bordered text sections is adjacent to a text label;
• means for extracting the portion of the digital image depicting the entire first document;
• means for removing the text values from the plurality of bordered text sections in the extracted portion of the digital image;
• means for generating a scaled image by scaling the portion of the digital image from a first resolution to a second resolution, wherein the second resolution is less than the first resolution, and wherein the scaled image includes the plurality of bordered text sections and the text label, but does not include the text values;
• means for generating a first hash from the scaled image;
• means for comparing the first hash to a set of hashes, each hash of the set of hashes corresponding to a template of a set of templates, wherein each template of the set of templates is associated with data indicative of a size and location of one or more sections designated for text of a corresponding document;
• means for selecting a first template as corresponding to the first document based on the first hash best matching one of the hashes of the set of hashes that corresponds to the first template;
• means for extracting one or more sections of the portion of the digital image based on the selected first template comprising data indicative of a size and location of the one or more sections;
• and
• means for performing optical character recognition on the extracted one or more sections of the portion of the digital image.