Deformity edge detection
AssigneeFuel 3D Technologies Limited
InventorsJamshid Dehmeshki, Adeala Zabair, Leo Navarro, George Thaw
Life of this patent
Estimating the term…
Abstract
A method for estimating a region associated with a deformity on a surface portion of the skin of a mammal, the method comprising capturing a three-dimensional image of the surface portion to produce first data, the first data comprising depth data, and generating second data from the first data. The second data comprises curvature data. The method further comprises combining the first data and the second data to produce third data, and performing region growing on the third data.
Claims25
1-25. (canceled)
A method for estimating a region associated with a deformity on or under a surface portion of the skin of a mammal, the method comprising:
- capturing an image of the surface portion to produce first data;
- generating second data from the first data;
- combining the first data and the second data to produce third data;
- and
- performing image segmentation on the third data to identify a region of interest portion of the image;
- wherein the image is a three-dimensional image, the first data comprises depth data, and the second data comprises curvature data.
A method according to any preceding claim, wherein:
- the step of performing image segmentation on the third data comprises performing region growing on the third data.
A method according to any preceding claim, wherein when the deformity has a first peak and a second peak, said region of interest portion of the image comprises a first region associated with the first peak and a second region associated with the second peak;
- and
- the method comprises combining the first region and the second region.
A method for estimating a region associated with a deformity on a surface portion of the skin of a mammal, the deformity having a first peak and a second peak, the method comprising:
- capturing an image of the surface to produce first data;
- generating modified data from the first data;
- performing image segmentation on the modified data to identify a region of interest portion of the image said region of interest image portion, the image segmentation comprising identifying a first region associated with the first peak and identifying a second region associated with the second peak;
- and
- combining the first region and the second region.
A method according to any preceding claim, comprising estimating at least one numerical characterisation parameter of the deformity.
A method according to any preceding claim, wherein the step of performing image segmentation is performed by a convolutional neural network.
A method according to any preceding claim, wherein the step of capturing said image comprises using at least one of stereoscopy and photometry.
Dependencies are inferred from the claim text.