Method and means for color detection and modification

Claim 3

A method of color detection and modification in accordance with claim 1, wherein said luminance factor signal has a value of zero for values of luminance lying at or between the greatest and least values of luminance for which samples were obtained, and has a value greater than zero for all other values of luminance.

Claim 4

A method of color detection and modification in accordance with claim 2, wherein the luminance factor signal varies linearly with luminance over those values of luminance which lie outside the range between the greatest and least values of luminance for which samples were obtained, the luminance factor signal being a continuous function over the entire luminance range.

Claim 5

A method of color correction in accordance with claim 1, wherein said extended curve defines values of the two other parameters which tend towards a pair of predetermined values for the two extreme values in the entire range of obtained luminance signal values, and which define that pair of predetermined values at the said extreme luminance signal values.

Claim 6

A method in accordance with claim 1, 2, 3 or 4, wherein said predetermined function of said signals is also dependent on a predetermined spread signal, such that the weighting signal has a high level when said differences are low, and decreases smoothly to zero as said differences become higher; and wherein said spread signal controls the rate at which the weighting signal decreases to zero, such that for high values of the spread signal, the weighting signal decreases more slowly to zero with increasing values of the difference signals.

Claim 7

A method in accordance with claim 1, 2, 3 or 4, wherein said luminance parameter represents pure luminance, the two other parameters being the Commission Internationale de L'Eclairage (C.I.E.) x and y components.

Claim 8

A method in accordance with claim 1, 2, 3 or 4 wherein said predetermined function of said signals is also dependent on a predetermined spead signal, such that the weighting signal has a high level when said differences are low, and decreases smoothly to zero as said differences become higher, said spread signal being calculated as a function of luminance.

Claim 9

A method in accordance with claim 1, 2, 3 or 4, wherein said luminance parameter represents pure luminance, the other two parameters being the C.I.E. u and v color components.

Claim 10

A method in accordance with claim 1, 2, 3 or 4, wherein in said step of combining the color correction values with the weighting signal, each color correction value is multiplied by the weighting signal.

Claim 11

A method in accordance with claim 1, 2, 3 or 4, wherein said sample curve defining variation of hue and saturation throughout the luminance range of the sample is a continuous curve comprising linear portions representing a linear interpolation between said sample points.

Claim 12

A method in accordance with claim 1, wherein said extended curve is extended from said sample curve by linear extrapolation at each end of the luminance range of the samples, the extended curve defining a pair of predetermined values for the two extreme values in the entire range of obtained luminance signal values.

Claim 13

A method of color correction in accordance with claim 1, wherein said weighting signal varies with said difference values and said luminance factor signal in such a manner that a sufficient departure of any one of the difference values from zero reduces to zero the difference between the modified color correction values and the predetermined set of color correction values.

Claim 14

A method in accordance with claim 12, wherein said weighting signal is equal to the product of a first predetermined function of the luminance deviation signal, a second predetermined function of a first one of the chromaticity deviation signals, and a third predetermined function of a second one of the chromaticity deviation signals, each of said functions assuming a predetermined constant value for zero difference values, and declining smoothly and monotonically from said predetermined value to zero for other difference values.

Claim 15

A method in accordance with claim 12, wherein said weighting signal is equal to exp[-S(pl 2 +δx 2 +δy 2 )], where S is a predetermined constant, pl is the luminance factor value, and δx and δy are the chromaticity deviation signal values.

Claim 16

A method in accordance with claim 12, wherein said weighting signal is equal to unity minus the product of a predetermined constant value and the largest of the absolute values of said chromaticity and luminance factor signals if this value is positive, the error factor signal otherwise being zero.

Claim 17

A method in accordance with claim 13, wherein said weighting signal varies in accordance with said chromaticity deviation signals and said luminance factor signal in a manner defined by the following expression: ##EQU3## where w is the weighting signal and A,B are predetermined constants, A≧1, B>0.

Claim 18

Means for detecting and modifying color values in portions of a color image, said means comprising: a. input means for receiving input color signals which represent a colored image, and for converting said image into a plurality of pixel portions, each pixel having a discrete location; b. digital conversion means for converting the input color signal at each pixel into three digital coordinate values, said first coordinate value representing an approximation of luminance, said second and third coordinate values representing chromaticity; c. sampling means to selectively sample one or more of said pixels representing a color value for which detection and modification is desired; d. means for storing said second and third chromaticity coordinate values as a function of said luminance coordinate values for each pixel sampled by said sampling means, said means defining a three-dimensional curve in color space when said coordinate values are stored in a memory means, said luminous coordinate defining the longitudinal path of said curve while said second and third chromaticity values define the LOCl of said curve at each level of luminance sampled; e. function generator means for defining a three-dimensional color sub-space around said color curve, said function means defining a weighted value for each pixel having color coordinate values that fall within said three-dimensional color sub-spaces; f. comparator means for sequentially sampling each pixel in said image and comparing the color coordinate values of said pixel with the color coordinate values of the color sub-space defined by said function generator means; and g. means responsive to said comparator means to combine a second color signal with said input signal for each pixel having color coordinate values falling within said defined color sub-space.

Claim 19

Means for detecting and modifying color values in portions of a color image, said means comprising: a. input means for receiving input color signals which represent a colored image, and for converting said image into a plurality of pixel portions, each pixel having a discrete location; b. digital conversion means for converting the input color signal at each pixel into three digital coordinate values, said first coordinate value representing an approximation of luminance, said second and third coordinate values representing chromaticity; c. sampling means to selectively sample one or more of said pixels representing a color value for which detection and modification is desired; d. means for storing said second and third chromaticity coordinate values as a function of said luminance coordinate values for each pixel sampled by said sampling means, said means defining a three-dimensional curve in color space when said coordinate values are stored in a memory means, said luminous coordinate defining the longitudinal path of said curve while said second and third chromaticity values define the LOCl of said curve at each level of luminance sampled; e. function generator means for defining a three-dimensional color sub-space around said color curve, said function means defining a weighted value for each pixel having color coordinate values that fall within said three-dimensional color sub-spaces; f. comparator means for sequentially sampling each pixel in said image and comparing the color coordinate values of said pixel with the color coordinate values of the color sub-space defined by said function generator means; and g. means responsive to said comparator means to combine a second color signal with said input signal for each pixel having color coordinate values falling within said defined color sub-space.

Claim 20

Means for detecting and modifying color as claimed in claim 17, wherein the first coordinate value (l) corresponds to luminance, said second (x) and third (y) chromaticity coordinates define values of hue and saturation, where hue is defined as an angle of rotation around a white or neutral point, and saturation is defined as the radius from said point.

Claim 21

Means for detecting and modifying color as claimed in claim 17, wherein said first coordinate value corresponds to luminance (l), as follows: l=aR+bG+cB wherein R represents red G represents green B represents blue and a, b and c represent any constant greater than 0.

Claim 22

Means for detecting and modifying color as claimed in claim 17, wherein said function generator means varies the weighted value as the distance from the color coordinate values to the color curve varies.

Claim 23

Means for detecting and modifying colors as claimed in claim 20, wherein said function generator means varies the weighted value as follows: w=T.sub.6 [T.sub.1 (l)+T.sub.4 (δx)+T.sub.5 (δy)] wherein w equals the weighted value, l represents the luminance coordinate value, δx and δy represent the second and third chromaticity coordinate values at said comparator means; T 1 is a function table; T 4 and T 5 are Log tables, and T 6 is an exponential table.

Claim 24

Means for detecting and modifying color as claimed in claim 17, wherein said means further includes additional function means to extrapolate luminous values above and below the luminous values obtained by said sampling means.

Claim 25

Means for detecting and modifying color as claimed in claim 17, wherein said memory means comprises three lookup tables stored in a random access memory means.

Claim 26

Means for detecting and modifying color as claimed in claim 17, wherein said means responsive to said weighted value further includes color generating means having independently variable values of luminance and chromaticity, said means also including an operator control whereby the operator may independently vary the luminance or chromaticity of the second color signal as it is combined with the first color signal.

Claim 27

Means for detecting and modifying color as claimed in claim 17, wherein said means further includes a mask generating means to identify the pixels detected by the function generator means.

Claim 28

Means for detecting and modifying color as claimed in claim 24, wherein said means further includes a color insertion means and a threshold means whereby said color insertion means interrupts the original color signal and inserts the second color signal whenever the weighted value exceeds a predetermined level.

Claim 29

Means for detecting and modifying color as claimed in claim 17, wherein three memory planes are arrayed with said pixels; said first memory plane storing the digital luminous coordinate values, the second memory plane storing the second color coordinate digital value, and the third memory plane storing the third color coordinate digital value, said memory planes providing a second storage means for both input color signals and combined color signals on a pixel-by-pixel basis.

Claim 30

Means for detecting and modifying color as claimed in claim 27, wherein said means further includes an integrating circuit for deriving the total proportional area of all pixels assigned a weighted value.

Claim 31

Means for detecting and modifying color as claimed in claim 17, wherein said means further defines interpolation means to provide a continuous color curve that defines variations of hue and saturation through the entire luminance range of the image.

Claim 32

Means for detecting and modifying color as claimed in claim 18, wherein sample values of x are averaged, and sample values of y are averaged when said sampling means selects a pixel with a similar luminous l value.

Claim 33

Means for detecting and modifying color as claimed in claim 18, wherein the x,y coordinates of the color curve are determined by least square approximations when plural values of x and y are obtained.

Claim 34

Digital means for the automatic detection of color, said means comprising: a. means for generating a three-dimensional color curve for a sample color to be detected, said trajectory defining variations of hue and saturation throughout the luminous range of the sample; b. memory means for storing digital coordinate values of said sample color curve, wherein chromaticity coordinate values are stored as a function of said luminance coordinate values; c. function generating means to define a three-dimensional color sub-space around said color curve, the longitudinal dimensions of said sub-space being determined by said luminous coordinate values and the cross-section of said sub-space being determined by the chromaticity coordinate values; d. input means for receiving additional color signals to be evaluated and compared with said sample color, said input means converting said additional color signals to three digital coordinate values, the first coordinate value representing luminance, the second and third coordinate values representing chromaticity; e. comparator means to compare the coordinate values of said incoming additional color values to the coordinate value of said three-dimensional color sub-space defined by said function generating means; and f. threshold means for receiving the output of said comparator means and detecting any incoming color signal having coordinate values that are defined within the coordinate values defined by said function generating means.

Claim 35

Digital means for the automatic detection of color, said means comprising: a. means for generating a three-dimensional color curve for a sample color to be detected, said trajectory defining variations of hue and saturation throughout the luminous range of the sample; b. memory means for storing digital coordinate values of said sample color curve, wherein chromaticity coordinate values are stored as a function of said luminance coordinate values; c. function generating means to define a three-dimensional color sub-space around said color curve, the longitudinal dimensions of said sub-space being determined by said luminous coordinate values and the cross-section of said sub-space being determined by the chromaticity coordinate values; d. input means for receiving additional color signals to be evaluated and compared with said sample color, said input means converting said additional color signals to three digital coordinate values, the first coordinate value representing luminance, the second and third coordinate values representing chromaticity; e. comparator means to compare the coordinate values of said incoming additional color values to the coordinate value of said three-dimensional color sub-space defined by said function generating means; and f. threshold means for receiving the output of said comparator means and detecting any incoming color signal having coordinate values that are defined within the coordinate values defined by said function generating means.

Claim 36

Means for detecting and modifying color as claimed in claim 32, wherein said first coordinate value corresponds to luminance as follows: l=aR+bG+cB wherein R represents red G represents green B represents blue and a, b and c represent any constant greater than 0.

Claim 37

Means for detecting and modifying color as claimed in claim 32, wherein said means further includes additional function means to extrapolate chromaticity values for luminous values above and below the luminous values of said sample.

Claim 38

Means for detecting and modifying color as claimed in claim 27, wherein said means further includes an integrating circuit for deriving the total proportional area of all pixels assigned a weighted value.

Claim 39

Digital means for the automatic detection of color as claimed in claim 32, wherein said memory means comprises first, second and third lookup tables in a random access memory.

Claim 40

Digital means for the automatic detection of color as claimed in claim 32, wherein said function generator means defines the color sub-space as follows: w=T.sub.6 [T.sub.1 (l)+T.sub.4 (δx)+T.sub.5 (δy)] wherein w is a signal applied to the threshold means to detect said color signal when w rises above a predetermined value, (δx)(δy) being chromaticity values at said comparator means, T 1 being a function table, T 4 and T 5 being log tables, and T 6 being an exponential table.

Claim 45

A method of automatic detecting or modifying color as claimed in claim 38 or 39 wherein the first coordinate value (l) corresponds to luminance, and the second (x) and third (y) chromaticity coordinates defined values of hue and saturation, wherein hue is defined as an angular rotation around a white or neutral point and saturation is defined as a radius from said point.

Claim 46

A method of automatically detecting or modifying colors as claimed in claim 38 or 39 wherein the first coordinate value corresponds to luminance (l), as follows: l=aR+bG+cB wherein R represents red G represents green B represents hue and a, b and c represent any constant greater than 0.

Claim 47

A method of automatic detecting or modifying color as claimed in claim 38 or 39 wherein the function generator creates and varies a weighted value assigned the color sub-space as follows: w=T.sub.6 [T.sub.1 (l)-T.sub.4 (δx)+T.sub.5 (δy)] where w equals the weighted value, represents luminance coordinate value δx and δy represent the second and third chromaticity coordinate values wherein said coordinate values are compared; T 1 being a function table; T 4 and T 5 being Log table, and T 6 being an exponential table.

Claim 48

A method of automatically detecting and modifying color as claimed in claim 39 which further includes the step of generating a mask to identify pixels detected by the function generator means.

Claim 49

A method of automatically detecting and modifying color as claimed in claim 39 which further includes the step of storing each of the modified digital coordinate values on a pixel by pixel basis.

Claim 50

A method of automatically detecting and modifying color as claimed in claim 39 and 45 wherein the digital coordinate values are stored in three memory planes.