Image detection apparatus and image detection method capable of detecting roundish shape
First Claim
1. An image detection apparatus comprising:
- a filter operation unit for conducting an operation of calculating a synthetic produce between inputted image data and a plurality of filters, respectively, wherein said filters differ in direction and have respective orientations;
a directional balance operation unit including an angle calculation unit for obtaining an angle made between a vector whose components are a plurality of outputs of said filter operation unit for each direction and a reference vector whose components are equal to each other; and
a roundish shape detection unit for detecting roundish shapes within the inputted image data based on an output of said directional balance calculation unit, wherein said directional balance operation unit includes a standard deviation operation unit for obtaining a standard deviation of a plurality of outputs of said filter operation unit for each different direction of said filters;
wherein said roundish shape detection unit detects said roundish shapes based on an output of said angle calculation unit and an output of said standard deviation operation unit;
wherein said roundish shaped detection unit includes a feature calculation unit for calculating a feature of the inputted image data for each size of said plurality of filters based on the output of said angle calculation unit, the output of said standard deviation operation unit and a pixel intensity value of the inputted image data; and
wherein said feature calculation unit calculates a first feature based on the output of said angle calculation unit and said pixel intensity value in accordance with the following mathematical expression;
1 Assignment
0 Petitions
Accused Products
Abstract
An image detection apparatus includes a filter operation unit for conducting an operation of calculating a synthetic product between inputted image data and a plurality of filters, wherein respectively, the filters differ in direction. And having respective orientations a directional balance operation unit includes an angle calculation unit for obtaining an angle made between a vector whose components are a plurality of outputs of the filter operation unit for each directions and a reference vector whose components are equal to each other. A rounded convex region detection unit detects rounded convex regions within the inputted image based on an output of the directional balance calculation unit.
77 Citations
24 Claims
-
1. An image detection apparatus comprising:
-
a filter operation unit for conducting an operation of calculating a synthetic produce between inputted image data and a plurality of filters, respectively, wherein said filters differ in direction and have respective orientations;
a directional balance operation unit including an angle calculation unit for obtaining an angle made between a vector whose components are a plurality of outputs of said filter operation unit for each direction and a reference vector whose components are equal to each other; and
a roundish shape detection unit for detecting roundish shapes within the inputted image data based on an output of said directional balance calculation unit, wherein said directional balance operation unit includes a standard deviation operation unit for obtaining a standard deviation of a plurality of outputs of said filter operation unit for each different direction of said filters;
wherein said roundish shape detection unit detects said roundish shapes based on an output of said angle calculation unit and an output of said standard deviation operation unit;
wherein said roundish shaped detection unit includes a feature calculation unit for calculating a feature of the inputted image data for each size of said plurality of filters based on the output of said angle calculation unit, the output of said standard deviation operation unit and a pixel intensity value of the inputted image data; and
wherein said feature calculation unit calculates a first feature based on the output of said angle calculation unit and said pixel intensity value in accordance with the following mathematical expression;
- View Dependent Claims (2)
-
-
3. An image detection apparatus comprising:
-
a filter operation unit for conducting an operation of calculating a synthetic product between inputted image data and a plurality of filters, respectively, wherein said filters differ in direction and have respective orientations;
a directional balance operation unit including an angle calculation unit for obtaining an angle made between a vector whose components are a plurality of outputs of said filter operation unit for each direction and a reference vector whose components are equal to each other;
a roundish shape detection unit for detecting roundish shapes within the inputted image data based on an output of said directional balance calculation unit; and
a false-positive region deletion unit for deleting a roundish shape which has been determined as a false-positive region from the roundish shapes detected by said roundish shape detection unit, wherein said false-positive region deletion unit comprises;
(i) area calculation means for calculating a maximum area of an inscribed circle inscribed to each of a plurality of detected convex rounded regions; and
(ii) a deletion region detection unit for detecting a to-be-deleted roundish shape based on an output of said directional balance operation unit if an output of said area calculation means is smaller than a predetermined value; and
wherein said roundish shape detection unit detects said roundish shapes in accordance with the following expression;
-
-
4. An image detection apparatus for detecting roundish shapes from an inputted image, said apparatus comprising:
-
a gradient vector calculation unit for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of d pixel on a half-line with a pixel of interest defined as a starting point of the half-line, said gradient vector having a magnitude corresponding to a differential value of the pixel intensity on a pixel position and a direction corresponding to a direction perpendicular to a plane contacting said three-dimensional curved surface;
an orthogonal projection calculation unit for calculating an orthogonal projection of said gradient vector obtained by the gradient vector calculation unit to the half-line;
a maximum length detection unit for detecting a maximum length of the orthogonal projection on each of said half-lines calculated by the orthogonal projection calculation unit;
a similarity calculation unit for calculating a similarity between said maximum length of the orthogonal projection to the half-line and that to a different half-line detected by the maximum length detection unit;
a correlation value calculation unit for obtaining a correlation value by adding the similarity between said maximum length of the orthogonal projection to one half-line and that to each of different half-lines calculated by the similarity calculation unit; and
a maximum correlation value calculation Unit for calculating said correlation value obtained by the correlation value calculation unit for all of the half-lines, fining a maximum correlation value of the correlation values and calculating a maximum value of said maximum correlation value when said pixel of interest is moved thoroughly within said roundish shapes, wherein said roundish shapes are detected based on an output from said maximum correlation value calculation unit; and
wherein said correlation value calculation unit adds correlation values of a certain half-line obtained in relation to other (d−
1) half-lines, respectively, in accordance with the following mathematical expression;
where (x0, y0)=a certain point within an input candidate malignant tumor region(xi, yi) (i=1, . . . , d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi)=a pixel intensity value on a pixel (xi, yi), A(α
i0, α
j0)=sin(β
ij0)β
ij0=|α
i0−
α
j0| mod π
, andα
i=2π
i/d(i=1, . . . ,d)
and the maximum correlation value calculation unit calculates values with respect to all the half-lines using the mathematical
and selects the maximum correlation value in accordance with the following mathematical expression;
where (x0, y0)=a certain point within the input candidate malignant tumor region. - View Dependent Claims (5, 6, 7, 8)
said half-line passes one of points at which a circumference of a circle centering around said noted point is equally divided into arcs. -
6. An image detection apparatus according to claim 5, wherein
in said similarity calculation unit, similarity between said maximum lengths is calculated using -
( p i0 , p j0 ) = min ( p i0 p j0 , p j0 p i0 )
where pi0 and pj0 are orthogonal projections obtained for two half-lines having a point (x0, y0) defined as a starting point and gradients different from each other, respectively.
-
-
7. An image detection apparatus according to claim 5, wherein
in said similarity calculation unit, similarity between said maximum lengths is calculated using -
( p i0 , p j0 ) = min ( p i0 p j0 , p j0 p i0 ) min ( p i0 , p j0 )
where pi0 and pj0 are orthogonal projections obtained for two half-lines having a point (x0, y0) defined as a starting point and gradients different from each other, respectively.
-
-
8. An image detection apparatus according to claim 4, further comprising:
a weighting calculation unit provided in an upstream stage portion of said correlation value calculation unit for conducting weighting based on a magnitude of an angle made between different half-lines, in accordance with the following mathematical expression;
-
-
9. An image detection apparatus for detecting roundish shapes from an inputted image, said apparatus comprising:
-
a gradient vector calculation unit for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point, said gradient vector having a magnitude corresponding to a differential value of the pixel intensity on a pixel position and a direction corresponding to a direction perpendicular to a tangent plane on said three-dimensional curved surface;
an orthogonal projection calculation unit for calculating an orthogonal projection of said gradient vector obtained by the gradient vector calculation unit to the half-line;
a maximum position detection unit for detecting a position at which the orthogonal projection of said half-line calculated by the orthogonal projection calculation unit reach a maximum;
a distance similarity calculation unit for calculating a distance between the maximum orthogonal projection position detected by the maximum position detection unit and said pixel of interest and calculating a similarity of distances calculated for respective half-lines;
a correlation value calculation unit for obtaining a correlation value by adding said distance similarity calculated for one half-line and that for each of different halt-lines calculated by the similarity calculation unit; and
a maximum correlation value calculation unit for calculating said correlation values obtained by the correlation value calculation unit for all of the half-lines, fining a maximum correlation value of the calculated correlation values and calculating a maximum value of said maximum correlation values when said pixel of interest is moved thoroughly within said roundish shapes, wherein said roundish shapes are detected based on an output from said maximum correlation value calculation unit; and
wherein said correlation value calculation unit adds correlation values of a certain half-line obtained in relation to other (d−
1) half-lines, respectively, in accordance with the following mathematical expression;
where (x0, y0) = a certain point within an input candidate malignant tumor region (x1, y1) (i=1, . . . , d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi)=a pixel intensity value on a pixel (xi, yi), A(α
i0, α
j0)=sin (β
ij0)β
ij0=|α
j0| mod π
, andα
i=2π
i/d(i=1, . . . ,d)and the maximum correlation value calculation unit calculates values with respect to all the half-lines using the mathematical expression;
and selects the maximum correlation value in accordance with the following mathematical expression;
where (x0, y0)=a certain point within the input candidate malignant tumor region. - View Dependent Claims (10, 11)
in said distance similarity calculation unit, said distance similarity is calculated using where ri0 is a vector having a point (x0, y0) as a starting point and one of points (xi, yi) (i=1, . . . d) at which a circumference of a circle of a radius of r is equally divided into d arcs, as an end point, and rj0 is a vector having a direction different from a direction of the vector ri0).
-
-
11. An image detection apparatus according to claim 9, further comprising:
-
a weighting calculation unit, provided in an upstream stage of said correlation value calculation unit, for conducting weighting based on a magnitude of an angle made between different half lines, in accordance with equation (33) on page 42 of the specification; and
wherein in said correlation value calculation unit, said distance similarity obtained by said similarity calculation unit is multiplied by a weight and resultant values are added for all directions.
-
-
12. An image detection method comprising:
-
a filter operation step of conducting an operation of calculating a synthetic product between inputted image data and a plurality of filters, respectively, wherein said filters differ in direction and have respective orientations;
a directional balance operation step including an angle calculation step of obtaining an angle made between a vector whose components are a plurality of outputs of said filter operation step for each direction and a reference vector whose components are equal to each other; and
a roundish shape detection step of detecting roundish shapes within the inputted image data based on an operation result of said directional balance operation step, wherein said directional balance operation step includes a standard deviation operation step of obtaining a standard deviation of a plurality of outputs obtained in said filter operation step for each different direction of said filters;
wherein in said roundish shape detection step, said roundish shapes are detected based on an operation result of said angle calculation step and an operation result of said standard deviation operation step;
wherein said roundish shaped detection step includes a feature calculation step of calculating a feature of the image data for each size of said plurality of filters based on the operation result of said angle calculation step, the operation result of said standard deviation operation step and a pixel intensity value of the inputted image data; and
wherein in said feature calculation step, a first feature is calculated based on the operation result of said angle calculation step and said pixel intensity value in accordance with the following mathematical expression;
- View Dependent Claims (13)
-
-
14. An image detection method comprising:
-
a filter operation step of conducting an operation of calculating d synthetic product between inputted image data and a plurality of filters, respectively, wherein said filters differ in direction and have respective orientations;
a directional balance operation step including an angle calculation step of obtaining an angle made between a vector whose components are a plurality of outputs of said filter operation unit for each direction and a reference vector whose components are equal to each other;
a roundish shape detection step of detecting roundish shapes within the inputted image data based on an operation result of said directional balance calculation step; and
a false-positive region deletion step of deleting a roundish shape which has been determined as a false-positive region from the roundish shapes detected by said roundish shape detection step, wherein said false-positive region deletion step comprises;
(i) an area calculation means step of calculating a maximum area of an inscribed circle inscribed to each of detected convex rounded regions; and
(ii) a deletion region detection unit step of detecting a to-be-deleted roundish shape based on an operation result of said directional balance operation step if a calculation result of said area calculation step is smaller than a predetermined value; and
wherein in said roundish shape detection step, said roundish shapes are detected in accordance with the following expression;
-
-
15. An image detection method for detecting roundish shapes from an inputted image, said method comprising:
-
a gradient vector calculation step for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point, said gradient vector having a magnitude corresponding to a differential value of the pixel intensity on a pixel position and a direction corresponding to a direction perpendicular to a tangent plane on said three-dimensional curved surface;
an orthogonal projection calculation step of calculating an orthogonal projection of said gradient vector to the half-line;
a maximum position detection step of detecting a position at which the orthogonal projection of said half-line reaches a maximum;
a distance similarity calculation step of calculating a similarity between the maximum orthogonal projection position and said pixel of interest and calculating similarity of distances calculated for respective half-lines;
a correlation value calculation step of obtaining a correlation value by adding said distance similarity calculated for one half-line and that for each of different half-lines; and
a maximum correlation value calculation step of calculating said correlation values for all of the half-lines, fining a maximum correlation value of the calculated correlation values and calculating a maximum value of said maximum correlation value when said pixel of interest is moved thoroughly within said roundish shapes, wherein said roundish shapes are detected based on the maximum value obtained in said maximum correlation value calculation step; and
wherein in said correlation value calculation step, correlation values of a certain half-line obtained in relation to other (d−
1) half--lines, respectively, are added in accordance with the following mathematical expression;
where (x0, y0)=a certain point within an input candidate malignant tumor region(xi, yi) (i=1, . . . , d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi)=a pixel intensity value on a pixel (xi, yi), A (α
i0, α
j0)=sin (β
ij0)β
ij0=|α
i0| mod π
, andα
i=2π
i/d(i−
1, . . . ,d)and in said maximum correlation value calculation step, values with respect to all the half-lines are calculated using the mathematical expression;
and the maximum correlation value is selected in accordance with the following mathematical expression;
where (x0, y0)=a certain point within the input candidate malignant tumor region. - View Dependent Claims (16, 17)
in said distance similarity calculation step, said distance similarity is calculated using
where ri0 is a vector having a point (x0, yi) as a starting point and one of points (xi, yi) (i=1, . . . d) at which a circumference of a circle of a radius of r is equally divided into d arcs, as an end point, and rj0 is a vector having a direction different from a direction of the vector ri0).
-
-
17. A method according to claim 15, further comprising:
a weighting calculation step, upstream of said correlation value calculation step, for conducting weighting based on a magnitude of an angle made between different half-lines, in accordance with the following mathematical expression;
-
18. An image detection method for detecting roundish shapes from an inputted image, comprising:
-
a gradient vector calculation step for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point of the half-line, said gradient vector having a magnitude corresponding to a differential value of the pixel intensities on pixel positions and a direction corresponding to a direction perpendicular to a tangent plane on said three-dimensional curved surface;
an orthogonal projection calculation step of calculating an orthogonal projection of said gradient vector to the half-line;
a maximum length detection step of detecting a maximum length of the orthogonal projection on each of said half-lines;
a similarity calculation step of calculating a similarity between said maximum length of the orthogonal projection to the half-line and that to a different half-line detected in the maximum length detection step;
a maximum position detection step of detecting a position at which the orthogonal projection of said half-line reaches a maximum;
a distance similarity calculation step of calculating a distance between the maximum orthogonal projection position and said pixel of interest and calculating a similarity of distances calculated for respective half-lines;
a correlation value calculation step of obtaining a correlation value by adding a product of said maximum orthogonal projection length similarity and said distance similarity calculated for one half-line and that or each of different half-lines calculated in the similarity calculation step and adding a resultant product; and
a maximum correlation value calculation step of calculating said correlation value for all of the half-lines, fining a maximum correlation value of the correlation values and calculating a maximum value of said maximum correlation value when said pixel of interest is moved thoroughly within said roundish shapes, wherein said roundish shapes are detected based on the maximum value obtained in said maximum correlation value calculation step; and
wherein in said correlation value calculation step, correlation values of a certain half-line obtained in relation to other (d−
1) half-lines, respectively, are added in accordance with the following mathematical expression;
where (x0, y0)=a certain point within an input candidate malignant tumor region(xi, yi) (i=1, . . . ,d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi)=a pixel intensity value on a pixel (xi, yi), A(α
i0, α
j0)=sin (β
ij0)β
ij0=|α
i0−
α
j0| mod π
, andα
i=2π
i/d(i=1, . . . ,d)and in said maximum correlation value calculation step, values with respect to all the half-lines are calculated using the mathematical expression;
and the maximum correlation value is selected in accordance with the following mathematical expression;
where (x0, y0)=a certain point within the input candidate malignant tumor region. - View Dependent Claims (19, 20, 21)
in said maximum orthogonal projection length similarity calculation step, similarity between said maximum lengths is calculated using
where pi0 and pj0 are orthogonal projections obtained for two half-lines having a point (x0, y0) defined as a starting point and gradients different from each other, respectively.
-
-
20. A method according to claim 18, wherein
in said maximum orthogonal projection length similarity calculation step, similarity between said maximum lengths is calculated using -
( p i0 , p j0 ) = min ( p i0 p j0 , p j0 p i0 ) min ( p i0 , p j0 )
where pi0 and pj0 are orthogonal projections obtained for two half-lines having a point (x0, y0) defined as a starting point and gradients different from each other, respectively.
-
-
21. A method according to claim 18, wherein
in said distance similarity calculation step, said distance similarity is calculated using -
( r i0 , r j0 ) = min ( r i0 r j0 , r j0 r i0 )
where ri0 is a vector having a point (x0, y0) as a starting point and one of points (xi, yi) (i=1, . . . d) at which a circumference of a circle of a radius of r is equally divided into d arcs, as an end point, and rj0 is a vector having a direction different from a direction of the vector ri0).
-
-
22. An image detection method for detecting roundish shapes from an inputted image, comprising:
-
a gradient vector calculation step for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point of the half-line, said gradient vector having a magnitude corresponding to a differential value of the pixel intensities on pixel positions and a direction corresponding to a direction perpendicular to a tangent plane on said three-dimensional curved surface;
an orthogonal projection calculation step of calculating an orthogonal projection of said gradient vector to the half-line;
a maximum length detection step of detecting a maximum length of the orthogonal projection on each of said half-lines;
a similarity calculation step of calculating a similarity between said maximum length of the orthogonal projection to the half-line and that to a different half-line detected in the maximum length detection step;
a maximum position detection step of detecting a position at which the orthogonal projection of said half-line reaches a maximum;
a distance similarity calculation step of calculating a distance between the maximum orthogonal projection position and said pixel of interest and calculating a similarity of distances calculated for respective half-lines;
a correlation value calculation step of obtaining a correlation value by adding a product of said maximum orthogonal projection length similarity and said distance similarity calculated for one half-line and that for each of different half-lines calculated in the similarity calculation step and adding a resultant product;
a maximum correlation value calculation step of calculating said correlation values for all of the half-lines, fining a maximum correlation value of the calculated correlation values and calculating a maximum value of said maximum correlation value when said pixel of interest is moved thoroughly within said roundish shapes; and
a weighting calculation step, upstream of said correlation value calculation step, for conducting weighting based on a magnitude of an angle mode between different half-lines, in accordance with the following mathematical expression;
-
-
23. A computer readable recording medium having stored thereon a program including a command for causing a computer to execute, in detecting roundish shapes from an input image, the following operations:
-
gradient vector calculation processing for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point of the half-line, said gradient vector having a magnitude corresponding to a differential value of the pixel intensity on a pixel position and a direction corresponding to a direction perpendicular to a plane on said three-dimensional curved surface;
orthogonal projection calculation processing for calculating an orthogonal projection of said gradient vector to the half-line;
maximum length detection processing for detecting a maximum length of the orthogonal projection on each of said half-lines;
similarity calculation processing for calculating a similarity between said maximum length of the orthogonal projection to the half-line and that to a different half-line;
correlation value calculation processing for obtaining a correlation value by adding the similarity between said maximum length of the orthogonal projection to one half-line and that to each of different half-lines;
maximum correlation value calculation processing for calculating said correlation value for all of the half-lines, fining a maximum correlation value of the correlation values and calculating a maximum value of said maximum correlation value when said pixel of interest is moved thoroughly within said roundish shapes; and
roundish shape detection processing for detecting said roundish shapes based on said maximum value calculated, wherein in said correlation value calculation processing, correlation values of a certain half-line obtained in relation to other (d−
1) half--lines, respectively, are added in accordance with the following mathematical expression;
where (x0, y0)=a certain point within an input candidate malignant tumor region(xi, yi) (i=1, . . . , d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi) a pixel intensity value on a pixel (xi, yi), A(α
i0, α
j0)=sin (β
ij0)β
ij0=|α
i0−
α
j0| mod π
, andα
i=2π
i/d(i=1, . . . ,d)and in said maximum correlation value calculation processing, values with respect to all the half-lines are calculated using the mathematical expression;
and the maximum correlation value is selected in accordance with the following mathematical expression;
where (x0, y0)=a certain point within the input candidate malignant tumor region.
-
-
24. A computer readable recording medium having stored thereon a program including a command for causing a computer to execute, in detecting roundish shapes from an input image, the following operations:
-
gradient vector calculation processing for, if said image is a three-dimensional curved surface consisting of pixel intensities and pixel positions on a two-dimensional coordinate, calculating a gradient vector of a pixel on a half-line with a pixel of interest defined as a starting point, said gradient vector having a magnitude corresponding to a differential value of the pixel intensity on a pixel position and a direction corresponding to a direction perpendicular to a tangent plane on said three-dimensional curved surface;
orthogonal projection calculation processing for calculating an orthogonal projection of said gradient vector to the half-line;
maximum position detection processing for detecting a position at which the orthogonal projection of said half-line reaches a maximum;
distance similarity calculation processing for calculating a distance between the maximum orthogonal projection position and said pixel of interest and calculating a similarity of distances calculated for respective half-lines;
correlation value calculation processing for obtaining a correlation value by adding said distance similarity calculated for one half-line and that for each of different half-lines;
maximum correlation value calculation processing for calculating said correlation values for all of the half-lines, fining a maximum correlation value of the calculated correlation values and calculating a maximum value of said maximum correlation values when said pixel of interest is moved thoroughly within said roundish shapes; and
roundish shape detection processing for detecting said roundish shapes based on said calculated maximum correlation value, wherein in said correlation value calculation processing, correlation values of a certain half-line obtained in relation to other (d−
1) half-lines, respectively, are added in accordance with the following mathematical expression;
where (x0, y0)=a certain point within an input candidate malignant tumor region(x1, y1) (i=1, . . . , d)=a point at which a circumference of a circle of a radius of r is divided into d arcs ri0=(xi−
x0, yi−
y0)dot (·
) means an inner product operation,ni=a normalized gradient vector, that is, f(xi, yi)=a pixel intensity value on a pixel (xi, yi), A(α
i0, α
j0)=sin (β
j0)β
ij0=|α
i0−
α
j0| mod π
, andα
i=2π
i/d(i=1, . . . ,d)and in said maximum correlation value calculation processing, values with respect to all the half-lines are calculated using the mathematical expression;
and the maximum correlation value is selected in accordance with the following mathematical expression;
where (x0, y0) a certain point within the input candidate malignant tumor region.
-
Specification