Sonar imaging
AssigneeSeeByte, Ltd.
InventorOliver Daniell
Life of this patent
Estimating the term…
Abstract
A method for recognizing a target in a sonar image, the method comprising: normalizing a sonar image; using/defining multiple test objects; rotating each test object between multiple positions; using a projection of each test object in each position as a template, so that multiple templates are provided for each test object, each template corresponding to a different rotational position; applying the multiple templates for the multiple test objects to the normalized image; and creating at least one feature vector for the image for use in target recognition.
Claims19
A method for recognising a target in a sonar image, the method comprising the steps of:
- normalising a sonar image;
- at least one of using or defining multiple three-dimensional test objects;
- rotating each three-dimensional test object between multiple positions;
- using a two-dimensional projection of each three-dimensional test object in each position as a template, so that multiple templates are provided for each test object, each template corresponding to a different rotational position;
- applying the multiple templates for the multiple test objects to the normalised image to determine a response for each template;
- and
- using the response determined for each template to create at least one feature vector representative of the image for use in target recognition,
- wherein:
- each template is at least one of the following three classes of templates:
- a standard object template, a bounded front template, or a bounded side template;
- the template response for each standard object template present is calculated as a mean of a sum of an intensity of the pixels in both highlight and shadow regions of the standard object template;
- the template response for each bounded front template present is calculated as a mean of a sum of an intensity of the pixels in a two-dimensional region the same size as the highlight region of the standard object template subtracted from a mean of a sum of the pixels in the standard template response;
- and
- the template response for each bounded side template present is calculated as a mean of a sum of an intensity of the pixels in each side region subtracted from the standard template response.
Dependencies are inferred from the claim text.