Level set tree feature detection

A non-transitory computer-readable medium encoded for level set tree feature detection for measuring changes between images, comprising:

• at least one instruction that, when executed by a computer, causes said computer to generate a fast level set tree transform of at least two pixilated images;
• at least one instruction that, when executed by a computer, causes said computer to create a binary map of said at least two pixilated images;
• at least one instruction that, when executed by a computer, causes said computer to calculate node values based on statistical differences amongst said at least two pixilated images;
• at least one instruction that, when executed by a computer, causes said computer to select a node from a set of nodes {I1 . . . In};
• at least one instruction that, when executed by a computer, causes said computer to calculate the number of white pixels in a region corresponding to said selected node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said selected node;
• at least one instruction that, when executed by a computer, causes said computer to store said number of false alarms for said selected node in an electronic storage device;
• at least one instruction that, when executed by a computer, causes said computer to calculate relative entropy, Hq({Gi}i=1q), of said selected node, wherein {Gi}i=1q are children of said selected node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as meaningful when said calculated number of false alarms for said selected node is less than a predetermined value ε;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as not meaningful when said calculated number of false alarms for said selected node is not less than said predetermined value ε;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a node from a set of nodes {I1 . . . In} instruction, said calculating the number of white pixels in a region corresponding to said node instruction, said calculating a number of false alarms for said selected node instruction, said storing said number of false alarms for said selected node instruction, said calculating relative entropy, Hq({Gi}i=1q), of said selected node wherein {Gi}i=1q are children of said selected node instruction, said labeling said selected node as meaningful when said calculated number of false alarms for said selected node is less than a predetermined value ε instruction, and said labeling said selected node as not meaningful when said calculated number of false alarms for said selected node is not less than said predetermined value ε instruction, until each node in the set of nodes I1 . . . In has been labeled either meaningful or not meaningful;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful node from a set of nodes {I1 . . . In};
• at least one instruction that, when executed by a computer, causes said computer to determine a family of sibling descendants, {Gi}i=1q, of said selected node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node;
• at least one instruction that, when executed by a computer, causes said computer to store said number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node in an electronic storage device;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as not meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is less than or equal to said number of false alarms for said selected node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is not less than or equal to said number of false alarms for said selected node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful node from a set of nodes {I1 . . . In} instruction, said determining a family of sibling descendants, {Gi}i=1q, of said selected node instruction, said calculating a number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node instruction, said storing said number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node instruction, said labeling said selected node as not meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is less than or equal to said number of false alarms for said selected node instruction, said labeling said selected node as meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is not less than or equal to said number of false alarms for said selected node instruction, until each node in the set of nodes {I1 . . . In} has been labeled either meaningful or not meaningful;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful node from a set of meaningful nodes {J1 . . . Jn};
• at least one instruction that, when executed by a computer, causes said computer to determine individual descendants, G, of said selected meaningful node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said descendants, G, of said selected meaningful node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful node as not a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected node is less than said number of false alarms for said selected meaningful node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful node as a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is not less than said number of false alarms for said selected meaningful node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful node from a set of meaningful nodes {J1 . . . Jn} instruction, said determining individual descendants, G, of said selected meaningful node instruction, said calculating a number of false alarms for said descendants, G, of said selected meaningful node instruction, said labeling said selected meaningful node as not a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is less than said number of false alarms for said selected meaningful node instruction, said labeling said selected meaningful node as a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is not less than said number of false alarms for said selected meaningful node instruction, until each meaningful node in the set of meaningful nodes {J1 . . . Jn} has been labeled either a dominant ancestor or not a dominant ancestor;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful and dominant ancestor node from a set of meaningful and dominant ancestor nodes {K1 . . . Kn};
• at least one instruction that, when executed by a computer, causes said computer to determine individual ascendants, G, of said selected meaningful and dominant ancestor node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful and dominant ancestor node as not maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is less than or equal to said number of false alarms for said selected meaningful and dominant ancestor node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful and dominant ancestor node as maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is not less than or equal to said number of false alarms for said selected meaningful and dominant node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful and dominant ancestor node from a set of meaningful and dominant ancestor nodes {K1 . . . Kn} instruction, said determining individual ascendants, G, of said selected meaningful and dominant ancestor node instruction, said calculating a number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node instruction, said labeling said selected meaningful and dominant ancestor node as not maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is less than or equal to said number of false alarms for said selected meaningful and dominant ancestor node instruction, said labeling said selected meaningful and dominant ancestor node as maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is not less than or equal to said number of false alarms for said selected meaningful and dominant node instruction, until each meaningful and dominant ancestor node in the set of meaningful and dominant ancestor nodes {K1 . . . Kn} has been labeled either maximal meaningful or not maximal meaningful;
• a readable storage device having computer readable program code stored thereon, wherein the computer readable program code includes computer executable instructions;
• and
• at least one instruction that, when executed by a computer, causes said computer to output in a tangible medium said maximal meaningful nodes.

A non-transitory computer-readable medium encoded for level set tree feature detection for measuring contrast changes in an image, comprising:

• at least one instruction that, when executed by a computer, causes said computer to generate a fast level set tree transform of one pixilated image;
• at least one instruction that, when executed by a computer, causes said computer to calculate node values based on statistical differences amongst said one pixilated image;
• at least one instruction that, when executed by a computer, causes said computer to select a node from a set of nodes {I1 . . . In};
• at least one instruction that, when executed by a computer, causes said computer to sum the number of pixels in a region corresponding to said selected node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said selected node;
• at least one instruction that, when executed by a computer, causes said computer to store said number of false alarms for said selected node in an electronic storage device;
• at least one instruction that, when executed by a computer, causes said computer to calculate relative entropy, Hq({Gi}i=1q), of said selected node, wherein {Gi}i=1q are children of said selected node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as meaningful when said calculated number of false alarms for said selected node is less than a predetermined value ε;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as not meaningful when said calculated number of false alarms for said selected node is not less than said predetermined value ε;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a node from a set of nodes {I1 . . . In} instruction, said summing the number of pixels in a region corresponding to said node instruction, said calculating a number of false alarms for said selected node instruction, said storing said number of false alarms for said selected node instruction, said calculating relative entropy, Hq({Gi}i=1q), of said selected node wherein {Gi}i=1q are children of said selected node instruction, said labeling said selected node as meaningful when said calculated number of false alarms for said selected node is less than a predetermined value ε instruction, and said labeling said selected node as not meaningful when said calculated number of false alarms for said selected node is not less than said predetermined value ε instruction, until each node in the set of nodes I1 . . . In has been labeled either meaningful or not meaningful;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful node from a set of nodes {I1 . . . In};
• at least one instruction that, when executed by a computer, causes said computer to determine a family of sibling descendants, {Gi}i=1q, of said selected node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node;
• at least one instruction that, when executed by a computer, causes said computer to store said number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node in an electronic storage device;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as not meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is less than or equal to said number of false alarms for said selected node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected node as meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is not less than or equal to said number of false alarms for said selected node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful node from a set of nodes {I1 . . . In} instruction, said determining a family of sibling descendants, {Gi}i=1q, of said selected node instruction, said calculating a number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node instruction, said storing said number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node instruction, said labeling said selected node as not meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is less than or equal to said number of false alarms for said selected node instruction, said labeling said selected node as meaningful when said calculated number of false alarms for said family of sibling descendants, {Gi}i=1q, of said selected node is not less than or equal to said number of false alarms for said selected node instruction, until each node in the set of nodes {I1 . . . In} has been labeled either meaningful or not meaningful;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful node from a set of meaningful nodes {J1 . . . Jn};
• at least one instruction that when executed by a computer, causes said computer to determine individual descendants, G, of said selected meaningful node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said descendants, G, of said selected meaningful node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful node as not a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected node is less than said number of false alarms for said selected meaningful node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful node as a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is not less than said number of false alarms for said selected meaningful node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful node from a set of meaningful nodes {J1 . . . Jn} instruction, said determining individual descendants, G, of said selected meaningful node instruction, said calculating a number of false alarms for said descendants, G, of said selected meaningful node instruction, said labeling said selected meaningful node as not a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is less than said number of false alarms for said selected meaningful node instruction, said labeling said selected meaningful node as a dominant ancestor when said calculated number of false alarms for said descendants, G, of said selected meaningful node is not less than said number of false alarms for said selected meaningful node instruction, until each meaningful node in the set of meaningful nodes {J1 . . . Jn} has been labeled either a dominant ancestor or not a dominant ancestor;
• at least one instruction that, when executed by a computer, causes said computer to select a meaningful and dominant ancestor node from a set of meaningful and dominant ancestor nodes {K1 . . . Kn};
• at least one instruction that, when executed by a computer, causes said computer to determine individual ascendants, G, of said selected meaningful and dominant ancestor node;
• at least one instruction that, when executed by a computer, causes said computer to calculate a number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node;
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful and dominant ancestor node as not maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is less than or equal to said number of false alarms for said selected meaningful and dominant ancestor node;
• or
• wherein at least one instruction that, when executed by a computer, causes said computer to label said selected meaningful and dominant ancestor node as maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is not less than or equal to said number of false alarms for said selected meaningful and dominant node;
• at least one instruction that, when executed by a computer, causes said computer to iterate through said selecting a meaningful and dominant ancestor node from a set of meaningful and dominant ancestor nodes {K1 . . . Kn} instruction, said determining individual ascendants, G, of said selected meaningful and dominant ancestor node instruction, said calculating a number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node instruction, said labeling said selected meaningful and dominant ancestor node as not maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is less than or equal to said number of false alarms for said selected meaningful and dominant ancestor node instruction, said labeling said selected meaningful and dominant ancestor node as maximal meaningful when said calculated number of false alarms for said ascendants, G, of said selected meaningful and dominant ancestor node is not less than or equal to said number of false alarms for said selected meaningful and dominant node instruction, until each meaningful and dominant ancestor node in the set of meaningful and dominant ancestor nodes {K1 . . . Kn} has been labeled either maximal meaningful or not maximal meaningful;
• a readable storage device having computer readable program code stored thereon, wherein the computer readable program code includes computer executable instructions;
• and
• at least one instruction that, when executed by a computer, causes said computer to output in a tangible medium said maximal meaningful nodes.