Authentication of dendritic structures
AssigneeArizona Board of Regents on behalf of Arizona State University
InventorMichael N. Kozicki
Life of this patent
Estimating the term…
Abstract
Assessing the authenticity of a dendrite on a substrate includes illuminating the dendrite and the substrate with radiation, obtaining a first image of the dendrite on the substrate at a first polarization angle, obtaining a second image of the dendrite on the substrate at a second polarization angle, and assessing a difference in intensity of a multiplicity of corresponding pixels of the candidate structure in the first image and the second image. If the difference in intensity of each pair of pixels of the multiplicity of corresponding pixels exceeds a preselected value, the dendrite is identified as authentic.
Claims25
A method of assessing the authenticity of a dendrite on a substrate, the method comprising:
- illuminating the dendrite and the substrate with radiation;
- obtaining a first image of the dendrite on the substrate at a first polarization angle;
- obtaining a second image of the dendrite on the substrate at a second polarization angle;
- assessing a difference in intensity of a multiplicity of corresponding pixels of the candidate structure in the first image and the second image;
- and
- if the difference in intensity of each pair of pixels of the multiplicity of corresponding pixels exceeds a preselected value, identifying the dendrite as authentic,
- wherein the preselected value is determined using a machine learning technique based on a multiplicity of images of authentic and non-authentic dendrites taken over a multiplicity of polarization angles.
A method of assessing the authenticity of a dendrite on a substrate, the method comprising:
- illuminating the dendrite and the substrate with radiation;
- obtaining a multiplicity of images of the dendrite on the substrate at a multiplicity of polarization angles;
- obtaining a degree of linear polarization image, an angle of linear polarization image, or both from the multiplicity of images;
- comparing the degree of linear polarization image, the angle of linear polarization image, or both with an unfiltered image of the dendrite and the substrate;
- assessing a difference between an intensity of a pixel corresponding to the substrate and a pixel corresponding to the dendrite in the degree of linear polarization image, the angle of linear polarization image, or both;
- and
- if the difference exceeds a preselected value, identifying the dendrite as authentic.
A method of assessing the authenticity of a dendrite on a substrate, the method comprising:
- illuminating the dendrite and the substrate with radiation;
- obtaining a multiplicity of images of the dendrite on the substrate at a multiplicity of polarization angles;
- obtaining a degree of linear polarization image, an angle of linear polarization image, or both from the multiplicity of images;
- assessing a difference between an intensity of a pixel corresponding to the substrate and a pixel corresponding to the dendrite in the degree of linear polarization image, the angle of linear polarization image, or both;
- and
- if the difference exceeds a preselected value, identifying the dendrite as authentic,
- wherein the preselected value is determined using a machine learning technique based on a multiplicity of images of authentic and non-authentic dendrites taken over a multiplicity of polarization angles.
Dependencies are inferred from the claim text.