Automated method of decomposing scanned documents
AssigneeApple Inc.
InventorHans-Werner Neubrand
Life of this patent
Estimating the term…
Abstract
Some embodiments produce an image capture application that implements a novel auto scan feature. The auto scan feature directs a scanner to perform an initial scan and then decomposes the scanned document into component regions. Some embodiments then identify a set of attributes of each region from the initial scan, select a set of optimal scanning parameters for each region based on the identified set of attributes, and then direct the scanner to perform a detailed secondary scan of each region with the identified set of scanning parameters. Following the secondary scan, some embodiments perform post-scan operations on the image of the scanned region.
Claims38
A non-transitory machine readable medium storing an image capture application for execution on a device, the image capture application comprising sets of instructions for:
- receiving a scanned electronic document based on a first scan;
- and
- automatically and without user input:
- identifying a plurality of component regions within the scanned electronic document that contain content, wherein a first component region is identified as containing content of a first type and a second component region is identified as containing content of a second type that is different from the first type of content based on attributes for a corresponding component region,
- identifying a set of scan parameters for the first and second component regions based on the attributes for the corresponding component region, and
- obtaining a second scan of the first and second component regions based on the set of scan parameters identified for the corresponding component region.
A non-transitory machine readable medium for storing an image capture application for execution on a device, the image capture application comprising sets of instructions for:
- receiving a scanned electronic document that was produced by a first scan:
- and
- automatically and without user input:
- identifying a plurality of component regions within the scanned electronic document that contain content, wherein a first component region is identified as containing content of a first type and a second component region is identified as containing content of a second type that is different from the first type of content based on attributes for a corresponding component region;
- rotating at least one of the first and second component regions with respect to the scanned electronic document based on the attributes for the at least one of the first and second component regions;
- and
- extracting and storing each component region separately.
A method of enabling control of a scanner, the method comprising:
- defining an image capture engine;
- and
- defining application programming interfaces (APIs) for commanding said image capture engine to obtain an initial scanned electronic document, the APIs further commanding said image capture engine to automatically and without user input identify a plurality of component regions within the initial scanned electronic document that contain content, wherein a first component region is identified as containing content of a first type and a second component region is identified as containing content of a second type that is different from the first type of content based on attributes for a corresponding region, identity a set of scan parameters for the first and second component regions based on the attributes for the corresponding component region, and obtain, based on the set of scan parameters, a second scan of the first and second component regions by performing a plurality of different individual secondary scan operations, wherein each different secondary scan operation is performed for a different component region.
A non-transitory machine readable medium storing an image capture application for execution by at least one processing unit, the image capture application comprising sets of instructions for:
- receiving a scanned electronic document comprising a first type of content and a second different type of content;
- and
- automatically and without user input:
- identifying, in the scanned electronic document, a first component region that contains the first type of content and a second component region that contains the second type of content based on attributes for the corresponding component region;
- obtaining a second scan of each of the first and second component regions using first and second set of parameters determined by attributes for the corresponding component region;
- and
- post processing the second scan of each of the first and second component regions by performing a first set of operations on the second scan of the first component region and performing a second different set of operations on the second scan of the second component region.
For an image capture application, a method of scanning a document, the method comprising:
- receiving a scanned electronic document computing text content and image content;
- and
- automatically and without user input:
- identifying, in the scanned electronic document, a first component region that contains the text content and a second component region that contains the image content based on attributes for corresponding component region
- obtaining a second scan of each of the first and second component regions using first and second set of parameters determined by the attributes for the corresponding component region, and
- post processing the second scan of each of the first and second component regions by performing optical character recognition on the text content and performing a rotation operation that at least partially rotates the image content with respect to the scanned electronic document.
Dependencies are inferred from the claim text.