Abstract: A computer system processes encoded digital data representing an image at image portions. The image at each portion has color coordinates S1, S2, S3 in a first color coordinate system (CCS), wherein S1?{square root over (T12+T22+T32)}, S2=T2/S1, S3=T3/S1, wherein T1, T2, T3 are color coordinates in a predefined 70%-orthonormal linear CCS. For each image portion whose color coordinate S1 is in a predefined range, the encoded digital data comprise color coordinates s1, s2, s3 such that: s1=kB·(ln(?S1)+?) rounded to an integer, where ? and ? are predefined constants, s2=kef·S2/S1 rounded to an integer, s3=kef·S3/S1 rounded to an integer, wherein kef is about 3—kB, or kef is equal to the smallest power of 2 which is greater than or equal to about 3·kB. The processing comprises using the encoded digital data to perform editing, and/or displaying, and/or decoding the encoded data.

Abstract: Color is edited using a color representation including digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B, f=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the other coordinates e and f or C and H. Other features are also provided. Brightness coding methods are provided to reduced the size of image data for storage and/or network transmission. The coding methods include logarithmic coding. Some embodiments use logarithmic or linear coding depending on the brightness at a particular pixel.

Abstract: When editing an image (160.2) with a computer system, a command may be issued to display a reference image (160.0 or 160.1) to allow a human user to visually compare the current image (160.2) with the reference image. In response, some embodiments display the entire reference image in the position of the current image. In some embodiments, if the current image was rotated, trimmed, or otherwise modified in respect to its geometry, the reference image is also rotated, trimmed, and/or otherwise modified in respect to its geometry when displayed for comparison. If another image (“third image”) (610) was incorporated into the current image during editing, then the reference image may or may not be combined with the third image when displayed for comparison with the current image. Some embodiments allow the user to specify whether or not the reference image should be combined with the third image.

Abstract: A computer system (810) receives a desired upper bound ?Smax for a relative quantization step ?S??S??/?S?? to be used when quantizing any color in some range of colors. Here S? and S? are adjacent colors in the set of colors to be made available for the quantized image, and ?·? is a norm in a 70%-orthonormal linear color coordinate system, the norm being the square root of the sum of squares of the tristimulus values. The computer system determines (510) suitable quantization steps for the brightness coordinate (B) and the chromatic coordinates (e,ƒ) in anon-linear color coordinate system, and quantizes (520) the brightness and chromatic coordinates accordingly.

Abstract: White balancing is performed using a mapping that maps the visible gamut's boundary into itself. A predefined color Win is mapped into a color Wout which is a white color or a color perceived as white under some viewing conditions. If Sin is some other color, a corresponding color Pin is determined on the visible gamut's boundary. Pin can be on the intersection of the visible gamut's boundary with a plane containing Sin and Win in a linear color coordinate system. Sin is mapped into a color Sout obtained from Wout and a value of the mapping on Pin.

Abstract: Color is edited using a color representation including digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B; f=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the other coordinates e and f or C and H. Other features are also provided. Brightness coding methods are provided to reduced the size of image data for storage and/or network transmission. The coding methods include logarithmic coding. Some embodiments use logarithmic or linear coding depending on the brightness at a particular pixel.

Abstract: In gamut mapping, as the brightness of the original color (S) increases further and further beyond the source gamut (110) boundary, the corresponding color (S?) in the target gamut (120) becomes closer and closer to a white color (W). Image distortion is reduced as a result.

Abstract: In a digital image having pixels pi (i=1, 2, . . . ) and a brightness B(pi) at each pixel pi, the global dynamic range of an image is changed by replacing the brightness B(pi) with the brightness B*(pi)=B(pi)×(B0/Bavg(pi))?, where B0 and ? are predefined constants, and Bavg(pi) is some average brightness value, e.g. the mean brightness over a region R(pi) containing the pixel pi. The local dynamic range is not changed significantly in some embodiments. The brightness of an image at a pixel can be defined so that multiplication of the brightness by a positive value k corresponds to multiplication of tristimulus values by k. In some embodiments, a color coordinate system is used with the brightness being one of the coordinates.

Abstract: Color is edited using a color representation including digital values B (brightness), e and ƒ such that B=?{square root over (D2+E2+F2)}, e=E/B, ƒ=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the other coordinates e and ƒ or C and H. Other image editing techniques and other features are also provided.

Abstract: Color is edited using a color representation including digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B, f=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Color-to-monochromatic image conversion, or changing the color of a monochromatic image, can be accomplished by setting e and f to constant values, and possibly (not necessarily) leaving B unchanged. Alternatively, the conversion can be accomplished by setting C and H to constant values. Other features are also provided.

Abstract: Color is edited using a color representation including digital values B (brightness), C (chroma) and H (hue) such that B=?{square root over (D2+E2+F2)}, cos C=D/B and tan H=E/F, where DEF is a linear color coordinate system. The hue can be changed without changing the brightness and without changing the saturation if the H coordinate is changed without changing B and C. Also, the brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the C and H coordinates. Other features are also provided.

Abstract: Color is represented using digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B, f=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the other coordinates e and f or C and H. Other features are also provided.

Abstract: Color is edited using a color representation including digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B, f=F/B, where DEF is a linear color coordinate system. Alternatively, color is represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. Brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the other coordinates e and f or C and H. Other features are also provided.

Abstract: Color is edited using a color representation including digital values B (brightness), C (chroma) and H (hue) such that B=?{square root over (D2+E2+F2)}, cos C=D/B and tan H=E/F, where DEF is a linear color coordinate system. The saturation can be changed without changing the brightness and without changing the hue if the C coordinate is changed without changing B and H. Also, the brightness can be changed without a color shift by changing the B coordinate and leaving unchanged the C and H coordinates. Other features are also provided.

Abstract: In a digital image having pixels pi (i=1,2, . . . ) and a brightness B(pi) at each pixel pi, the contrast editing is performed by replacing the brightness B(pi) with B*(pi)=Bavg1??(pi)×Bo?(pi), where ? is a positive constant other than 1, and Bavg is a weighted average of the B values in an image region R(pi) containing the pixel pi. The image can be a color image, with color represented using digital values B (brightness), e and f such that B=?{square root over (D2+E2+F2)}, e=E/B, f=F/B, where DEF is a linear color coordinate system. Alternatively, color can be represented using digital values B, C (chroma) and H (hue), where cos C=D/B and tan H=E/F. The contrast can be edited without a color shift (i.e. without changing the chromaticity coordinates) by changing the B coordinate and leaving unchanged the other coordinates e and f or C and H. Sharpness can be edited using the same techniques with a small region R(pi).

Abstract: A digital image processing system (134) receives a signal S(x,y,t,c) where (x,y) are pixel coordinates, t is time, and c is a color component. The system generates the signal S*(x,y,t,c) as a function of a linear combination of the values S(x?,y?,t?,c) over a set A of points (x?,y?,t?) with plural values (x?,y?) and plural values t?, wherein each value S(x?,y?,t?,c) in the linear combination is multiplied by a coefficient which is a function of the difference S(x,y,t,c)?S(x?,y?,t?,c). In some embodiments, one or more of the values S(x?,y?,t?,c) are replaced with previously calculated values S*(x?,y?,t?,c) to save memory.

Abstract: In the process of interpolation of video and/or audio digital data Sn, an interpolated value is chosen depending on whether or not an edge is detected. The edge detection is performed as follows. Let L=|S2n?S2n?2|, C=|S2n+2?S2n|, and R=|S2n+4?S2n+2|. An edge is detected when R>9(C+L) or L>9(C+R). A predictor P2n+1=K1S2n+K2S2n+2 provides the interpolated value for S2n+1 when an edge is detected, where K1=kL/(kL+kR), K2=kR/(kL+kR), where kL=1/(1+L), kR=1/(1+R) or kL=d2n+1,2n+2/(1+L) and kR=d2n,2n+1/(1+R). A different predictor (e.g. a cubic predictor) is used when no edge is detected. Other embodiments are also provided.

Abstract: Methods and apparatus are provided for performing reversible color conversion on digital image data using integer computer arithmetic. This color conversion provides an approximation of the luminance component to the luminance as perceived by the human visual system with any necessary precision (Kd) and without multiplication and division operations other than shifts. The color conversion can be used in conjunction with a lossless or lossy compression/decompression process. Other embodiments, including non-reversible and non-integer color conversion, are also provided.

Abstract: Methods and apparatus are provided for performing reversible color conversion on digital image data using integer computer arithmetic. This color conversion provides an approximation of the luminance component to the luminance as perceived by the human visual system with any necessary precision (Kd) and without multiplication and division operations other than shifts. The color conversion can be used in conjunction with a lossless or lossy compression/decompression process. Other embodiments, including non-reversible and non-integer color conversion, are also provided.

Abstract: In the process of interpolation of video and/or audio digital data Sn, an interpolated value is chosen depending on whether or not an edge is detected. The edge detection is performed as follows. Let L=|S2n?S2n?2|, C=|S2n+2?S2n|, and R=|S2n+4?S2n+2|. An edge is detected when R>9(C+L) or L>9(C+R). A predictor P2n+1=K1S2n+K2S2n+2 provides the interpolated value for S2n+1 when an edge is detected, where K1=kL/(kL+kR), K2=kR/(kL+kR), where kL=1/(1+L), kR=1/(1+R) or kL=d2n+1,2n+2/(1+L) and kR=d2n,2n+1/(1+R). A different predictor (e.g. a cubic predictor) is used when no edge is detected. Other embodiments are also provided.