Compensating for geometric distortion of images in constrained processing environments
AssigneeDigimarc Corporation
InventorVojtech Holub
Life of this patent
Estimating the term…
Abstract
An image processing method determines a geometric transform of a suspect image by efficiently evaluating a large number of geometric transform candidates in environments with limited processing resources. Processing resources are conserved by using complementary methods for determining a geometric transform of an embedded signal. One method excels at higher geometric distortion, and specifically, distortion caused by greater tilt angle of a camera. Another method excels at lower geometric distortion, for weaker signals. Together, the methods provide a more reliable detector of an embedded data signal in image across a larger range of distortion while making efficient use of limited processing resources in mobile devices.
Claims20
A method of reading an embedded digital payload in an image, the method comprising:
- obtaining a suspect image;
- transforming the suspect image into an image feature space;
- for a first set of geometric transform candidates, applying a first complementary process for evaluating the first set, wherein the first complementary process applies a fitting process that produces first refined geometric transform candidates from the first set having detection metrics for an embedded signal in the image feature space that satisfy predetermined criteria;
- applying a second complementary process for evaluating geometric transform candidates, the second complementary process applying a correlation process on a geometric parameter space mapping of the suspect image and producing second refined geometric transform candidates having detection metrics for the embedded signal that satisfy predetermined criteria;
- selecting a refined candidate geometric transform from the first and second refined geometric transform candidates based on detection metrics;
- and
- using the refined candidate geometric transform to extract a digital payload from the embedded signal.
A reader device comprising:
- an imager operable to capture an image;
- memory configured to store the image from the imager;
- an image pre-processor configured to execute on a processor, the image pre-processor configured to convert the image in the memory into a first domain for evaluating an embedded signal to determine a geometric transform of the embedded signal;
- a first geometric transform module configured to execute on a processor, the first geometric transform module configured to evaluate a first set of geometric transform candidates by using the first set of geometric candidates as seeds to a least squares process and producing a first set of refined geometric transform candidates with detection metrics;
- a second geometric transform module configured to execute on a processor, the second geometric transform module configured to evaluate a second set of geometric transform candidates by applying a correlation process on a rotation and scale mapping of the image and producing a second set of refined geometric transform candidates with detection metrics;
- a processor configured to execute instructions to select a refined candidate geometric transform from first and second refined geometric transform candidates based on detection metrics;
- and
- a processor configured to execute instructions to extract a digital payload from the embedded signal using the selected refined candidate geometric transform.
A non-transitory computer readable medium on which is stored instructions, which when executed by a processor, perform a method of reading an embedded digital payload in an image, the method comprising:
- obtaining a suspect image;
- transforming the suspect image into a first domain for evaluating geometric distortion of an embedded signal;
- for a first set of geometric transform candidates, applying a first complementary process to evaluate the first set, wherein the first complementary process applies a least squares process that produces first refined geometric transform candidates from the first set having detection metrics for the embedded signal that satisfy predetermined criteria;
- applying a second complementary process for evaluating geometric transform candidates on a rotation and scale mapping of the suspect image and producing second refined geometric transform candidates having detection metrics for the embedded signal that satisfy predetermined criteria;
- selecting a refined candidate geometric transform from the first and second refined geometric transform candidates based on detection metrics;
- and
- using the refined candidate geometric transform to extract a digital payload from the embedded signal.
Dependencies are inferred from the claim text.