Chromakey method and associated circuit arrangement

Claim 2

A chromakey method in which the pixel data of a first video picture are mixed with the pixel data of a second video picture in such a way that parts of the first video picture represent the foreground and the second video picture represents the background of an overall picture, and in which all color vectors associated with the background of the first video picture constitute a key region (S) in an n-dimensional color space, where n is an integer equal to or greater than 3, characterized in that the method comprises the following steps: a) dividing the color space into two sub-spaces (A,B) by an (n-1)-dimensional boundary region (G) in such a way that the first sub-space (A) comprise the key region (S) and possible transition colors, while the second sub-space (B) and the boundary region (G) comprise the color vectors of the first video picture which are associated with its foreground; and b) for a color vector p 1 which is located within the first sub-space (A), defining the value of a keying function k(p 1 )(0≦k(p 1 )≦1) as a color vector KC of the key region (S) by way of the requirement that the end point of the vector p 1 -k(p 1 )KC is located in the boundary region (G).

Claim 3

A method as claimed in claim 1, characterized in that the color vector p 3 of a pixel of the overall picture, which with respect to a pixel of the first video picture corresponds to the transition color p 1 and with respect to a pixel of the second video picture corresponds to the color vector p 2 , is defined in accordance with the equation p.sub.3 =p.sub.1 -k(p.sub.1)KC+k(p.sub.1)p.sub.2.

Claim 4

A method as claimed in claim 1, characterized in that the points of the boundary region (G) comply with an equation g(x)=0 so that the vector p 1 -k(p i )KC complies with the equation g(p.sub.1 -k(p.sub.1)KC)=0.

Claim 5

A method as claimed in claim 3, characterized in that the equation g(x)=0 is linear per region.

Claim 6

An arrangement for performing a chromakey mixing in which the pixel data of a first video picture are mixed with the pixel data of a second video picture in such a way that parts of the first video picture represent the foreground and the second video picture represents the background of an overall picture, and in which all color vectors associated with the background of the first video picture constitute a key region (S) in an n-dimensional color space, where n is an integer equal to or greater than 3, characterized in that the arrangement comprises: means for dividing the color space into two sub-spaces (A,B) by an (n-1)-dimensional boundary region (G) in such a way that the first sub-space (A) comprises the key region (S) and possible transition colors, while the second sub-space (B) and the boundary region (G) comprise the color vectors of the first video picture which are associated with its foreground, wherein the points of the boundary region (G) comply with an equation g(x)=0 so that the vector p 1 -k(p 1 )KC complies with the equation g(p 1 -k(p 1 )KC)=0; and means for defining, for a color vector p 1 which is located within the first sub-space (A), the value of a keying function k(p 1 )(0≦k(p 1 )≦1) as a color vector KC of the key region (S) by way of the requirement that the end point of the vector p 1 -k(p 1 )KC is located in the boundary region (G).

Claim 7

An arrangement for performing a chromakey mixing in which the pixel data of a first video picture are mixed with the pixel data of a second video picture in such a way that parts of the first video picture represent the foreground and the second video picture represents the background of an overall picture, and in which all color vectors associated with the background of the first video picture constitute a key region (S) in an n-dimensional color space, where n is an integer equal to or greater than 3, characterized in that the arrangement comprises: means for dividing the color space into two sub-spaces (A,B) by an (n-1)-dimensional boundary region (G) in such a way that the first sub-space (A) comprises the key region (S) and possible transition colors, while the second sub-space (B) and the boundary region (G) comprise the color vectors of the first video picture which are associated with its foreground, wherein the points of the boundary region (G) comply with an equation g(x)=0 so that the vector p 1 -k(p 1 )KC complies with the equation g(p 1 -k(p 1 )KC)=0; and means for defining, for a color vector p 1 which is located within the first sub-space (A), the value of a keying function k(p 1 )(0≦k(p 1 )≦1) as a color vector KC of the key region (S) by way of the requirement that the end point of the vector p 1 -k(p 1 )KC is located in the boundary region (G).

Claim 8

An arrangement as claimed in claim 5, characterized in that said dividing means and said defining means comprise: means for providing parameters fixing a key color KC (|KC|, Φ,Θ), a transition color p 1 (Cb, Cr, Y) and a boundary region (α,β); a transformation unit coupled to said providing means for transforming the transition color p 1 to components r, s and t in a rotated coordinate system in response to the key color parameters Φ and Θ in accordance with the matrix: ##EQU3## and means for calculating the value of k from the components r, s and t in response to the key color parameter |KC| and the boundary region parameters α and β in accordance with k(p.sub.1)=(r-s* cot α-t* cot β)/|KC|.

Claim 9

An arrangement as claimed in claim 6, characterized in that said transformation unit and said calculating means are components of a computer (C).