JP2005123660A - Image processing apparatus and method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To decrease a change caused in the impression of an image or the like before and after the image processing. <P>SOLUTION: A color space inverse conversion section 104 discriminates to which color space input image data belong. A conversion control section 108 controls an image processing section 102 and the color space inverse conversion section 104 in response to the discrimination. That is, when receiving image data of a linear color space, the conversion control section 108 allows a color space conversion section 100 and the image processing section 102 to merely pass through the image data. Or when receiving image data of a nonlinear color space, the conversion control section 108 controls the color space inverse conversion section 104 to linearly convert components of the image data, which are linearly convertible independently of other components among the components of the image data and also controls the color space inverse conversion section 104 to inversely convert the color space of the convertible components into the original nonlinear color space. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus and method for processing image data.

For example, with the spread of ICC (International Color Consortium) profiles, color image processing belonging to a plurality of types of color spaces is required.
Further, for example, when the image resolution changes before and after the image processing apparatus, resolution conversion is performed to match the input image with the resolution of the output apparatus.
In order to realize the resolution conversion, for example, a bilinear interpolation algorithm is used.
In the bilinear interpolation algorithm, the input pixel value PI (x, y) is multiplied by the area ratio of a plurality of input pixels including the pixel range of the output pixel, and the output pixel value PO (X, Y) is calculated.

However, the pixel value and the color light intensity have a non-linear relationship in a uniform perceptual color space where the human eyes perceive the color difference almost evenly.
When resolution conversion processing is performed on an image belonging to such a non-linear color space, the linearly interpolated output pixel value does not become the original value, so that the impression of the image differs before and after the processing.
In order to prevent a change in image impression before and after processing, for example, Patent Document 1 discloses a method in which an input image is subjected to color space conversion to a linear color space, image processing is performed, and the image is then returned to a non-linear color space.

JP 2000-101859 A

However, if the input image belongs to a complex non-linear color space such as CIELAB, the conversion process to the linear color space becomes complicated, so that the conversion process takes time.
Further, since the conversion process is complicated, the number of effective gradations of the image is reduced, and so-called pseudo contours are likely to occur.

[Image processing device]
In order to achieve the above object, a first image processing apparatus according to the present invention includes first image data belonging to a linear color space or second image data belonging to one or more nonlinear color spaces. Is a color space determination means for determining whether the image data belongs to the linear color space or the nonlinear color space, and passes the first image data according to a result of the determination, or any one of the nonlinear colors The second image data belonging to the color space is converted into first image data belonging to the linear color space, and the first image data passed through or the first image data obtained as a result of the conversion. Image processing means for performing predetermined image processing on one image data, and first image data obtained as a result of the conversion and subjected to the predetermined image processing according to the determination result, Non-linear Inversely converted into the second image data belonging to one of the space, or the is plain, and an inverse transform means for plain the first image data to which the predetermined image processing has been performed.

  Preferably, the converting means passes the second image data except when it is determined that the second image data belongs to any of the predetermined image data, and the image processing means includes: The predetermined processing is performed on the passed second image data, and the inverse conversion unit is configured except when it is determined that the second image data belongs to any of the predetermined image data. The second image data that has been subjected to the predetermined processing is passed through.

  Preferably, the image data belonging to one or more of the non-linear color spaces includes a plurality of channels, and the color space determination means includes a plurality of images included in the image data determined to belong to the second image data. Determining whether any of the channels is independent or non-independent of the other channels, and the converting means converts the independent channels included in the second image data into a linear color space, and First image data, the image processing means performs the predetermined processing on the first image data obtained as a result of the conversion, and the inverse conversion means is obtained as a result of the conversion. The independent channel included in the first image data subjected to the predetermined image processing is inversely converted into one of the nonlinear color spaces.

  Preferably, the non-linear color space includes a uniform perceptual color space and / or a relative lightness color space.

  The second image processing apparatus according to the present invention may be configured such that the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces is the linear color space. And a color space determining means for determining which one of the nonlinear color spaces belongs, and according to a result of the determination, the first image data is passed or the first color image belonging to any one of the nonlinear color spaces Conversion means for converting the second image data into first image data belonging to a linear color space.

[Image processing method]
In addition, according to the first image processing method of the present invention, the first image data belonging to the linear color space or the second image data belonging to one or more nonlinear color spaces is the linear color space. And the non-linear color space, the first image data is passed through according to the determination result, or the second image data belonging to any one of the non-linear color spaces is determined. The image data is converted into first image data belonging to a linear color space, and predetermined image processing is performed on the processed first image data or the first image data obtained as a result of the conversion. The first image data obtained as a result of the conversion and subjected to the predetermined image processing is inversely converted into second image data belonging to one of the nonlinear color spaces according to the result of the determination. Or the element Suggested that, to plain the first image data to which the predetermined image processing has been performed.

  Further, the second image processing method according to the present invention is such that the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces is the linear color space. And the non-linear color space, the first image data is passed through according to the determination result, or the second image data belonging to any one of the non-linear color spaces is determined. The first image data belonging to the linear color space is converted.

[program]
In addition, the first program according to the present invention is configured such that the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces is the linear color space and the A color space determination step for determining which one of the nonlinear color spaces belongs, and the second image data that passes through the first image data or belongs to one of the nonlinear color spaces according to a result of the determination. A conversion step of converting image data to first image data belonging to a linear color space, and the first image data obtained through the conversion or the first image data obtained as a result of the conversion, An image processing step for performing predetermined image processing, and first image data obtained as a result of the conversion and subjected to the predetermined image processing in accordance with the result of the determination, are stored in the nonlinear color space. Or the inverse transformation belonging second image data Re, or wherein the plain, the predetermined image processing to perform the inverse transform steps in computer plain the first image data was made.

  Further, the second program according to the present invention is configured such that the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces is the linear color space and the A color space determination step for determining which one of the nonlinear color spaces belongs, and the second image data that passes through the first image data or belongs to one of the nonlinear color spaces according to a result of the determination. The computer is caused to execute a conversion step of converting the image data into first image data belonging to the linear color space.

According to the image processing apparatus and the method of the present invention, it is possible to reduce a change that occurs in the impression of an image before and after processing.
Further, according to the image processing apparatus and the method thereof according to the present invention, the processing amount of the image processing can be reduced.

[Outline of the present invention]
First, in order to help understanding of the present invention, an outline thereof will be described.
FIG. 1 is a diagram illustrating a method for enlarging the resolution of an input image by 1.5 times.
In the description here, the input image data PI ₀₀ and PI ₁₀ shown in FIG. 1 partitioned by dotted lines is enlarged 1.5 times to generate output image data PO ₀₀ , PO ₁₀ and PO ₂₀ partitioned by solid lines. The case is taken as a specific example.

In order to enlarge the input image data PI _{X, Y} to output image data PO _{x, y} , bilinear interpolation is performed as shown in Equation 1 below.
PO _{x, y} = r _{X, Y} PI _{X, Y} + r _{X + 1, Y} PI _{X + 1, Y} + r _{X, Y + 1} PI _{X, Y + 1} + r _{X + 1, Y + 1} PI _{X + 1, Y + 1} (Formula 1)
In Equation 1, X and Y indicate the coordinates of the input image, and x and y indicate the coordinates of the output image.

Specifically, in order to obtain the output image data PO ₁₀ shown in FIG.
_{PO 10 = 0.5 × PI 00 +} 0.5 × PI 10 ( Equation 2)
Examples of the method for specifying the color space of the input image data include a method of adding an ICC profile to the image data, or a method of attaching the specification of CIELAB / YCbCr to a tag of Tiff format image data. Can do.

For example, when gamma correction is performed on the channels R, G, and B of the image data in the linear RGB color space, the non-linear channels R ′, G ′, and B ′ that have been gamma-corrected are expressed as shown in Equation 3 below. Can be obtained.
R ′ = R ^{1 / γ} , G ′ = G ^{1 / γ} , B ′ = B ^{1 / γ} (Formula 3)
The nonlinear channels R ′, G ′, and B ′ thus obtained can be independently converted into linear color space image data, and the nonlinear channels R ′, G ′, and B ′ can be converted into γ. The original linear channels R, G, and B can be restored by riding.

On the other hand, for example, in order to obtain the linear channel R, when it is necessary to perform an operation as shown in the following expression 4 on the non-linear channels R ′, G ′, and B ′, Even if the above calculation is performed, the original linear channel R cannot be obtained.
R = e (f _R (R ′), h (G ′), i (B ′)) (Formula 4)

On the other hand, other non-linear channels may be independently converted to the original linear channel.
As described above, the image data channel of the non-linear channel includes an independent channel that can be converted from a linear state to a non-linear state independently, and can be converted back to a linear state independently from a non-linear state, and a linear channel. A state may be converted into a non-linear state independent of other channels, and a non-independent channel that may be converted back from a non-linear state to a linear state independent of other channels may be included.

Of these independent and non-independent channels, it takes a lot of processing time to convert non-independent non-linear channels to linear channels, and non-independent non-linear channels are converted to linear channels. If the resolution conversion process is performed on the image data obtained in this way, a pseudo contour tends to occur.
From this point of view, the present invention converts only the channels that can be converted into linear independently among the channels included in the image data of the non-linear color space, and linearly converts channels that cannot be converted into linear independently. It is configured to perform resolution conversion processing and the like without converting to.

FIG. 2 is a diagram showing a configuration of the image processing apparatus 1 according to the present invention.
FIG. 3 is a diagram illustrating a configuration of the computer 16 that executes the image processing apparatus 1 when the image processing apparatus 1 illustrated in FIG. 2 is realized by software.
As illustrated in FIG. 2, the image processing apparatus 1 includes a color space conversion unit 100, an image processing unit 102, a color space inverse conversion unit 104, a color space determination unit 106, and a conversion control unit 108.

As shown in FIG. 3, the computer 16 includes a computer main body 160 including a CPU 162 and a memory 164, a display / input device 166 including a keyboard / display, an image input / output device 168 for inputting / outputting image data, and a CD. It comprises a recording device 170 such as a device / HDD device.
That is, the computer 16 includes a configuration part as a general computer capable of inputting and outputting image data.

Each component of the image processing apparatus 1 is realized by dedicated hardware, for example.
Alternatively, each component of the image processing apparatus 1 is realized as a computer program, for example, supplied to the computer 16 via the recording medium 172, loaded into the memory 164, and executed.

With these components, the image processing apparatus 1 accepts image data in the linear color space input from the outside, performs image processing such as resolution conversion processing, and the like. Is converted back to a different color space (including non-linear color space) and output.
In addition, the image processing apparatus 1 uses these components to select only channels (independent channels) that can be converted into the linear color space independently from the channels included in the image data of the nonlinear color space input from the outside. Convert to linear image data in the linear color space, leave the other channels (non-independent channels) as they are, and perform image processing such as resolution conversion processing. Reversely convert to a different color space (including linear color space) and output.

In the image processing apparatus 1, the color space determination unit 106 checks, for example, a tag attached to the input image data and determines whether the input image data is image data in a linear color space or any kind of nonlinear color space. Is determined, and the conversion control unit 108 is notified of the determination result.
The image processing unit 102 performs image processing such as resolution conversion processing on the image data in the linear color space input from the color space conversion unit 100 and outputs the image data to the color space inverse conversion unit 104.
In addition, the image processing unit 102 performs image processing on the non-independent channels that are not converted to linear and the linearly converted independent channels that are included in the non-linear image data input from the color space conversion unit 100, and the color space Output to the inverse transform unit 104.

The conversion control unit 108 controls the conversion processing of the color space conversion unit 100 and the color space inverse conversion unit 104 according to the determination result of the color space determination unit 106.
In other words, for example, when the input image data is linear color space image data, the conversion control unit 108 controls the color space conversion unit 100 to pass the input image data and output it to the image processing unit 102. To do.
It should be noted that “through” does not matter whether the processing is performed without processing, or the conversion processing is performed using a lookup table with the same input / output, and the processing is performed as a result.

For example, when the input image data is non-linear color space image data and includes only channels that can be linearly converted independently, the conversion control unit 108 includes the color space conversion unit 100 in the input image data. All the channels to be transmitted are converted into linear and controlled so as to be output to the image processing unit 102.
For example, when the input image data is non-linear color space image data and includes an independent channel and a non-independent channel, the conversion control unit 108 includes the independent space included in the input image data. Control is performed so that only the channel is linearly converted, and the non-independent channel is passed through and output to the image processing unit 102.

For example, when the input image data is non-linear color space image data and includes only non-independent channels, the conversion control unit 108 allows all the channels included in the input image data to pass through. And control to output to the image processing unit 102.
Also, the conversion control unit 108 controls the color space inverse conversion unit 104 to pass through and output the non-linear color space image data processed by the image processing unit 102.

Also, the conversion control unit 108 controls the color space inverse conversion unit 104 so that the image data in the linear color space processed by the image processing unit 102 is different from the original non-linear color space image data or the original. The image data is inversely converted into image data of a non-linear color space and output.
Further, the conversion control unit 108 controls the color space inverse conversion unit 104 to inversely convert only the independent channel included in the image data processed by the image processing unit 102 into the image data of the original nonlinear color space, A non-independent channel is passed through and output.
The color space conversion unit 100 and the image processing unit 102 perform conversion processing / inverse conversion processing on the image data in accordance with the control of the conversion control unit 108 described above.

[Overall operation]
FIG. 4 is a flowchart showing the overall operation of the image processing apparatus 1 shown in FIG.
Hereinafter, the overall operation of the image processing apparatus 1 (FIG. 2) will be described.
As shown in FIG. 4, the color space determination unit 106 of the image processing apparatus 1 determines whether or not the input image data is image data in a linear color space.
In step 100 (S100), the image processing apparatus 1 proceeds to the processing of S102 when the input image data is image data in the linear color space, and proceeds to the processing of S104 otherwise.

In step 102 (S102), the conversion control unit 108 controls the color space conversion unit 100 and the color space inverse conversion unit 104 so as to pass through the image data input thereto.
That is, in the image processing apparatus 1, the color space conversion unit 100 passes the input image data and outputs it to the image processing unit 102.
The image processing unit 102 processes this image data.
The color space inverse transform unit 104 passes the processed image data and sets it as output image data.

In step 104 (S104), the color space determination unit 106 determines whether the input image data is non-linear color space image data and all of the channels included in the image data are independent channels.
That is, the color space inverse transform unit 104 determines whether or not all channels included in the input image data can be linearized independently.
The image processing apparatus 1 proceeds to the process of S106 when all the channels included in the input image data can be linearized independently, and proceeds to the process of S108 otherwise.

In step 106 (S106), the conversion control unit 108 controls the color space conversion unit 100 to convert the input image data into image data in the linear color space, and also controls the color space inverse conversion unit 104, The image data processed by the image processing unit 102 is inversely converted into image data in the original nonlinear color space, for example.
That is, in the image processing apparatus 1, the color space conversion unit 100 converts input image data into image data in a linear color space.
The image processing unit 102 processes the linearly converted image data.
The color space inverse transform unit 104 inversely transforms the processed image data into, for example, image data in the original non-linear color space to obtain output image data.

In step 108 (S108), the color space determination unit 106 determines whether the input image data is image data in a non-linear color space and includes independent channels and non-independent channels.
The image processing apparatus 1 proceeds to the process of S110 when the input image data in the non-linear color space includes an independent channel and a non-independent channel, and otherwise proceeds to the process of S102.

In step 110 (S110), the conversion control unit 108 controls the color space conversion unit 100 to convert the independent channel included in the input image data into a channel of the linear color space, and allows the non-independent channel to pass through. The color space reverse conversion unit 104 is controlled to reversely convert the independent channel included in the image data processed by the image processing unit 102 into, for example, a channel in the original nonlinear color space.
That is, in the image processing apparatus 1, the color space conversion unit 100 converts only an independent channel of input image data into a channel of a linear color space.
The image processing unit 102 processes the linearly converted independent channel and the passed non-independent channel.
The color space inverse transform unit 104 inversely transforms the independent channel of the processed image data into the original non-linear color space, passes the non-independent channel, and sets the output image data.

  The present invention can be used in an image processing apparatus and method for processing image data.

It is a figure which illustrates the method of enlarging the resolution of an input image 1.5 times. It is a figure which shows the structure of the image processing apparatus which concerns on this invention. FIG. 3 is a diagram illustrating a configuration of a computer that executes an image processing apparatus when the image processing apparatus illustrated in FIG. 2 is realized by software. 3 is a flowchart illustrating an overall operation of the image processing apparatus illustrated in FIG. 2.

Explanation of symbols

1 ... Image processing device,
100: Color space conversion unit,
102 Image processing unit,
104... Color space inverse transform unit,
106: Color space determination unit,
108 ... conversion control unit,
16: Computer,
160 ... computer main body,
162... CPU
164 ... Memory,
166... Display / input device,
168 ... Image input / output device,
170... Recording device,
172 ... Recording medium

Claims (12)

Color that determines whether first image data belonging to a linear color space or second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space Space determination means;
Depending on the result of the determination, the first image data is passed through, or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. Conversion means;
Image processing means for performing predetermined image processing on the passed first image data or the first image data obtained as a result of the conversion;
According to the result of the determination, the first image data obtained as a result of the conversion and subjected to the predetermined image processing is inversely converted into second image data belonging to any of the nonlinear color spaces, Or an inverse conversion means for passing through the first image data that has been passed through and has undergone the predetermined image processing. The converting means passes through the second image data except when it is determined that the second image data belongs to any of the predetermined image data,
The image processing means performs the predetermined processing on the passed second image data,
The reverse conversion means passes through the second image data subjected to the predetermined processing except when it is determined that the second image data belongs to any of the predetermined image data. The image processing apparatus described. The image data belonging to one or more of the non-linear color space includes a plurality of channels,
The color space determination means determines which of a plurality of channels included in the image data determined to belong to the second image data is independent or independent of other channels;
The converting means converts an independent channel included in the second image data into a linear color space to obtain the first image data,
The image processing means performs the predetermined processing on the first image data obtained as a result of the conversion,
The inversion means reversely converts an independent channel obtained as a result of the conversion and included in the first image data subjected to the predetermined image processing into one of the non-linear color spaces. 2. The image processing apparatus according to 2. The image processing device according to claim 1, wherein the non-linear color space includes a uniform perceptual color space and a relative lightness color space or any one of them. A color for determining whether the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space Space determination means;
Depending on the result of the determination, the first image data is passed through or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. An image processing apparatus having conversion means. Determining whether the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space;
Depending on the result of the determination, the first image data is passed through or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. ,
Perform predetermined image processing on the first image data passed through or the first image data obtained as a result of the conversion,
In accordance with the result of the determination, the first image data obtained as a result of the conversion and subjected to the predetermined image processing is inversely converted to second image data belonging to one of the nonlinear color spaces, Alternatively, an image processing method for passing through the first image data that has been passed through and has undergone the predetermined image processing. Determining whether first image data belonging to a linear color space or second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space;
Depending on the result of the determination, the first image data is passed through or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. Image processing method. A color for determining whether the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space A space determination step;
Depending on the result of the determination, the first image data is passed through or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. A conversion step;
An image processing step of performing predetermined image processing on the first image data passed through or the first image data obtained as a result of the conversion;
In accordance with the result of the determination, the first image data obtained as a result of the conversion and subjected to the predetermined image processing is inversely converted to second image data belonging to one of the nonlinear color spaces, Or a reverse conversion step of passing through the first image data passed through and subjected to the predetermined image processing. The converting step passes through the second image data except when it is determined that the second image data belongs to any of the predetermined image data,
The image processing step performs the predetermined processing on the passed second image data,
9. The reverse conversion step passes through the second image data subjected to the predetermined processing except when it is determined that the second image data belongs to any of the predetermined image data. The program described. The image data belonging to one or more of the non-linear color space includes a plurality of channels,
The color space determining step determines which of a plurality of channels included in the image data determined to belong to the second image data is independent or independent of other channels;
The converting step converts an independent channel included in the second image data into a linear color space to obtain the first image data,
The image processing step performs the predetermined processing on the first image data obtained as a result of the conversion,
The reverse conversion step reversely converts an independent channel obtained as a result of the conversion and included in the first image data subjected to the predetermined image processing into one of the nonlinear color spaces. 9. The program according to 9. The program according to any one of claims 8 to 10, wherein the nonlinear color space includes a uniform perceptual color space and / or a relative lightness color space. A color for determining whether the first image data belonging to a linear color space or the second image data belonging to one or more nonlinear color spaces belongs to the linear color space or the nonlinear color space A space determination step;
Depending on the result of the determination, the first image data is passed through or the second image data belonging to any one of the nonlinear color spaces is converted into first image data belonging to a linear color space. A program that causes a computer to execute the conversion step.

Images