Method for classifying an object in a moving picture

Claim 1

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; obtaining Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of selected sampling points in said object image, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels.

Claim 2

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; obtaining Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of selected sampling points in said object image, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels.

Claim 8

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; obtaining Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of selected sampling points in said object image, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of rotation positions with each said sampling point being a center of rotation, wherein said object is a human body, and a central wavelength 2π/u 0 of said Gabor wavelet is about two times a width of said human body, and wherein said Gabor wavelet expansion coefficients are obtained based on Eq. (19), ( T m , l  s )  ( x 0 , y 0 ) = α m  ∑ y = - q / 2 + 1 q / 2  ∑ x = - p / 2 + 1 p / 2  s  ( x , y )  ψ l  ( α m  x - x 0 , α m  y - y 0 ) , ( 19 ) where ψ l ( x, y )=ψ( {dot over (x)}, {dot over (y)} ),  (20) [ x . y . ] = [ cos   ϕ l sin   ϕ l - sin   ϕ l cos   ϕ l ]  [ x y ] , ( 21 ) and ψ  ( x , y ) = 1 4   π 2  σ 0 2   - 1 4  π   σ 0 2  ( x 2 + y 2 )  [  j  u 0  x -  - π   σ 0 2  u 0 2 ] , ( 22 ) j: (−1) ½ (imaginary unit), α: scale constant, m: scaling parameter for giving a scale transformation level, p: number of pixels in the x-direction of object image, q: number of pixels in the y-direction of object image, (x 0 , y 0 ): coordinates of a sampling point, ψ l : rotation angle of Gabor wavelet around a sampling point (x 0 , y 0 ), l: rotation step, u 0 : central frequency of Gabor wavelet of Eq. (22), and σ 0 2 : variance of Gaussian window function.

Claim 9

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; obtaining Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of selected sampling points in said object image, wherein said Gabor wavelet expansion coefficients are obtained at a plurality of rotation positions with each said sampling point being a center of rotation, wherein said object is a human body, and a central wavelength 2π/u 0 of said Gabor wavelet is about two times a width of said human body, and wherein said Gabor wavelet expansion coefficients are obtained based on Eq. (19), ( T m , l  s )  ( x 0 , y 0 ) = α m  ∑ y = - q / 2 + 1 q / 2  ∑ x = - p / 2 + 1 p / 2  s  ( x , y )  ψ l  ( α m  x - x 0 , α m  y - y 0 ) , ( 19 ) where ψ l ( x, y )=ψ( {dot over (x)}, {dot over (y)} ),  (20) [ x . y . ] = [ cos   ϕ l sin   ϕ l - sin   ϕ l cos   ϕ l ]  [ x y ] , ( 21 ) and ψ  ( x , y ) = 1 4   π 2  σ 0 2   - 1 4  π   σ 0 2  ( x 2 + y 2 )  [  j  u 0  x -  - π   σ 0 2  u 0 2 ] , ( 22 ) j: (−1) ½ (imaginary unit), α: scale constant, m: scaling parameter for giving a scale transformation level, p: number of pixels in the x-direction of object image, q: number of pixels in the y-direction of object image, (x 0 , y 0 ): coordinates of a sampling point, ψ l : rotation angle of Gabor wavelet around a sampling point (x 0 , y 0 ), l: rotation step, u 0 : central frequency of Gabor wavelet of Eq. (22), and σ 0 2 : variance of Gaussian window function.

Claim 10

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; extracting an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; obtaining Gabor wavelet expansion coefficients of said extracted object image; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said Gabor wavelet expansion coefficients for said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients for said reference motions and unknown motion are obtained at a plurality of selected sampling points in said object image with a coordinate origin being set to approximately a center of said object, and obtained at a plurality of predetermined rotation positions with each said sampling point being a center of rotation, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels.

Claim 11

A method for classifying an object in a moving picture, comprising: preparing a template that includes Gabor wavelet expansion coefficients of an image of said object in a plurality of frames of a video image sequence representing each of a plurality of different reference motions of said object; extracting an image of said object in a plurality of frames of a video image sequence representing an unknown motion of said object; obtaining Gabor wavelet expansion coefficients of said extracted object image; calculating matching factors between said unknown motion and said reference motions based on said Gabor wavelet expansion coefficients for said unknown motion and said Gabor wavelet expansion coefficients for said reference motions in said template; and classifying said unknown motion based on said matching factors, wherein said Gabor wavelet expansion coefficients for said reference motions and unknown motion are obtained at a plurality of selected sampling points in said object image with a coordinate origin being set to approximately a center of said object, and obtained at a plurality of predetermined rotation positions with each said sampling point being a center of rotation, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels.

Claim 13

A method for classifying a pattern of an object, comprising: preparing a template including Gabor wavelet expansion coefficients of a reference pattern image representing a plurality of different reference patterns of said object; obtaining Gabor wavelet expansion coefficients of an unknown pattern image representing an unknown pattern of said object; calculating matching factors between said unknown pattern and said reference patterns based on said Gabor wavelet expansion coefficients for said unknown pattern and said Gabor wavelet expansion coefficients for said reference patterns in said template; and classifying said unknown pattern based on said matching factors, wherein said Gabor wavelet expansion coefficients for said reference patterns and unknown pattern are obtained based on Eq. (15) at a plurality of selected sampling points of said pattern image with a coordinate origin being set to approximately a center of said pattern image, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels, ( T m , l  s )  ( x 0 , y 0 ) = α m  ∑ y = - q / 2 + 1 q / 2  ∑ x = - p / 2 + 1 p / 2  s  ( x , y )  ψ l  ( α m  x - x 0 , α m  y - y 0 ) , ( 15 ) where ψ l ( X, y )=ψ( {dot over (x)}, {dot over (y)} ),  (16) [ x . y . ] = [ cos   ϕ l sin   ϕ l - sin   ϕ l cos   ϕ l ]  [ x y ] , ( 17 ) and ψ  ( x , y ) = 1 4   π 2  σ 0 2   - 1 4   π   σ 0 2  ( x 2 + y 2 )  [  j   u 0  x -  - π   σ 0 2  u 0 2 ] , ( 18 ) j: (−1) ½ (imaginary unit), α: scale constant, m: scaling parameter for giving a scale transformation level, p: number of pixels in the x-direction of an object image, q: number of pixels in the y-direction of an object image, (x 0 , y 0 ): coordinates of a sampling point, φ l : rotation angle of Gabor wavelet around a sampling point (x 0 , y 0 ), l: rotation step u 0 : central frequency of Gabor wavelet of Eq. (18), and σ 0 2 : variance of Gaussian window function.

Claim 14

A method for classifying a pattern of an object, comprising: preparing a template including Gabor wavelet expansion coefficients of a reference pattern image representing a plurality of different reference patterns of said object; obtaining Gabor wavelet expansion coefficients of an unknown pattern image representing an unknown pattern of said object; calculating matching factors between said unknown pattern and said reference patterns based on said Gabor wavelet expansion coefficients for said unknown pattern and said Gabor wavelet expansion coefficients for said reference patterns in said template; and classifying said unknown pattern based on said matching factors, wherein said Gabor wavelet expansion coefficients for said reference patterns and unknown pattern are obtained based on Eq. (15) at a plurality of selected sampling points of said pattern image with a coordinate origin being set to approximately a center of said pattern image, and wherein said Gabor wavelet expansion coefficients are obtained at a plurality of scale transformation levels, and a number of said sampling points is set to a different number for each of said levels, ( T m , l  s )  ( x 0 , y 0 ) = α m  ∑ y = - q / 2 + 1 q / 2  ∑ x = - p / 2 + 1 p / 2  s  ( x , y )  ψ l  ( α m  x - x 0 , α m  y - y 0 ) , ( 15 ) where ψ l ( X, y )=ψ( {dot over (x)}, {dot over (y)} ),  (16) [ x . y . ] = [ cos   ϕ l sin   ϕ l - sin   ϕ l cos   ϕ l ]  [ x y ] , ( 17 ) and ψ  ( x , y ) = 1 4   π 2  σ 0 2   - 1 4   π   σ 0 2  ( x 2 + y 2 )  [  j   u 0  x -  - π   σ 0 2  u 0 2 ] , ( 18 ) j: (−1) ½ (imaginary unit), α: scale constant, m: scaling parameter for giving a scale transformation level, p: number of pixels in the x-direction of an object image, q: number of pixels in the y-direction of an object image, (x 0 , y 0 ): coordinates of a sampling point, φ l : rotation angle of Gabor wavelet around a sampling point (x 0 , y 0 ), l: rotation step u 0 : central frequency of Gabor wavelet of Eq. (18), and σ 0 2 : variance of Gaussian window function.