System and method for quantifying reflection e.g. when analyzing laminated documents
AssigneeAu10Tix Limited
InventorsRon Atzmon, Sergey Markin, Benjamin Neeman, Yael Moscovitz
Life of this patent
Estimating the term…
Abstract
A system for computerized authentication of a laminated object, the system comprising a digital medium storing a digital image of at least a portion of the laminated object in a computer-implemented memory; a shininess analyzer operative, using a processor, to generate shininess data quantifying shininess of the digital image; and a parameterized computerized authentication sub-system operative to differentially perform at least one laminated object authentication operation based on the shininess data.
Claims15
A system for computerized authentication of a laminated object, the system comprising:
- a digital image of at least a portion of the laminated object stored in a computer-implemented memory;
- a shininess analyzer operative, using a processor, to generate shininess data quantifying shininess of said digital image;
- and
- a parameterized computerized authentication sub-system operative, using the processor, to differentially perform at least one laminated object authentication operation based on said shininess data,
- wherein said image is represented by a multiplicity of pixels and said shininess analyzer is operative for quantifying shininess of the image by:
- generating a 2-dimensional histogram array representing frequency of occurrence of each of a plurality of n-tuplets of color component values, in the image,
- wherein said shininess analyzer is also operative to identify, for each individual diagonal from among a plurality of diagonals covering the histogram array, a maximum array element corresponding to an n-tuplet of color component values which occurs more frequently than other color component values within the same individual diagonal.
A method for quantifying shininess of the image by:
- generating a 2-dimensional histogram array representing frequency of occurrence of each of a plurality of n-tuplets of color component values, in the image;
- for each individual diagonal from among a plurality of diagonals covering the histogram array, identifying a maximum array element corresponding to an n-tuplet of color component values which occurs more frequently than other color component values within the same individual diagonal;
- determining a shininess value for each individual pixel, defining an individual n-tuplet of color component values, from among the multiplicity of pixels, by finding said individual n-tuplet of color component values in said 2-dimensional histogram array, thereby to identify a specific diagonal from among the plurality of diagonals to which said individual n-tuplet belongs, and computing a distance between said individual n-tuplet of color component values and the maximum array element defined for said specific diagonal;
- and
- outputting a function of said distance,
- wherein said identifying an n-tuplet which occurs more frequently is operative to disregard at least one frequently occurring color which exceeds a threshold selected to disqualify initially bright colors.
A non-transitory computer readable medium having computer readable program code stored therein, said computer readable program code adapted to be executed to implement a method for quantifying shininess of the image by:
- generating a 2-dimensional histogram array representing frequency of occurrence of each of a plurality of n-tuplets of color component values, in the image;
- for each individual diagonal from among a plurality of diagonals covering the histogram array, using at least one processor for identifying a maximum array element corresponding to an n-tuplet of color component values which occurs more frequently than other color component values within the same individual diagonal;
- determining a shininess value for each individual pixel, defining an individual n-tuplet of color component values, from among the multiplicity of pixels, by finding said individual n-tuplet of color component values in said 2-dimensional histogram array, thereby to identify a specific diagonal from among the plurality of diagonals to which said individual n-tuplet belongs, and computing a distance between said individual n-tuplet of color component values and the maximum array element defined for said specific diagonal;
- and
- outputting a function of said distance,
- wherein said identifying an n-tuplet which occurs more frequently is operative to disregard at least one frequently occurring color lying along an individual diagonal whose distance from an initial color along the individual diagonal indicates that said frequently occurring color is such a dark color that the initial color could not be a shiny version of that dark color.
Dependencies are inferred from the claim text.