JP2001112023A - Image compression method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image compression method that can optimally compress image data. SOLUTION: First image data are received (step S11). Then a prescribed function is used to obtain a parameter of a luminance signal and a parameter of color difference signals are obtained with cross-reference (step S12). Furthermore, an image compression algorithm to which the luminance parameter and the color difference parameter are set is used to compress the image data received in the step S11 (step S13).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image compression method for compressing image data.

[0002]

2. Description of the Related Art Conventionally, DCT (Discrete Cosine Transform)
And an image compression algorithm using a wavelet transform. In these algorithms, a user individually sets a parameter for luminance and a parameter for color difference, and compresses image data.

[0003]

As described above, conventionally, the user individually sets the parameters of the luminance and the parameters of the color difference. For example, the data amount of the compressed image data is stored in the storage medium. However, there is a problem that the storage capacity may be exceeded or the image quality may be degraded.

[0004] The present invention has been made to solve the above problems, and has as its object to obtain an image compression method for always optimally compressing image data.

[0005]

According to a first aspect of the present invention, there is provided a method for compressing image data using an image compression algorithm for setting a luminance parameter and a color difference parameter. The luminance parameter and the chrominance parameter are determined in association with each other by a predetermined function.

2. The image compression method according to claim 1, wherein said luminance parameter and said color difference parameter are obtained based on a desired compression ratio of the entire image data.

[0007] According to a third aspect of the present invention, there is provided:
Detecting the storage capacity of the storage location of the compressed image data,
The luminance parameter and the color difference parameter are obtained based on the detected storage capacity.

[0008] The problem solving means according to claim 4 of the present invention is:
If the data amount of the image data after compression has not reached the target value, the luminance parameter and the color difference parameter are updated, and the image data is compressed again.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a flowchart showing an image compression method according to the first embodiment of the present invention.

An apparatus that performs the image compression method is, for example, an apparatus that handles image data such as a personal computer or a digital camera (hereinafter, referred to as an “image processing apparatus”).

First, image data is input to an image processing apparatus (step S11). FIG. 2 shows an example of the image data.
The image data DA in FIG. 2 includes m × n pixel data, and one pixel data is a total of 24 data of red data (8 bits), green data (8 bits), and blue data (8 bits). Consists of bits.

Next, parameters of luminance and parameters of chrominance required for the image compression algorithm are set (step S12).

The luminance parameter is, for example, a compression ratio of luminance data, and the chrominance parameter is, for example, a compression ratio of chrominance data. The compression ratio is the ratio between the data before compression and the data after compression, and the higher the compression ratio, the smaller the compressed data.

The parameter of the luminance and the parameter of the color difference are obtained in association with each other by a predetermined function. For example, as shown in FIG. 4 described later, the user sets the compression ratio X of the entire image data, and calculates the luminance from the compression ratio X of the entire image data using a predetermined function (for example, expressions 1 to 7 described later). And the color difference parameter may be obtained. Alternatively, the user sets only the luminance parameter, and determines a predetermined function (for example, Equations 4 to 7 described later) from the luminance parameter.
May be used to determine the color difference parameter. Or,
The user sets only the color difference parameter, and uses a predetermined function from the color difference parameter (for example, Equations 4 to
7) to determine the luminance parameter.

Next, the image data is compressed by the image compression algorithm in which the luminance parameter and the color difference parameter are set in step S12 (step S1).
3).

Next, the compressed data (hereinafter, referred to as compressed data) compressed in step S13 is stored in a recording medium such as a memory, a hard disk, or a floppy disk (step S14).

As described above, in the first embodiment, the parameters of the luminance and the parameters of the color difference are set in association with each other by the predetermined function. As described above, since the image processing apparatus can appropriately set the luminance parameter and the color difference parameter on behalf of the user, the image data can always be optimally compressed. Therefore, it is possible to prevent the data amount of the compressed image data from exceeding the storage capacity of the storage medium at the storage destination and deteriorating the image quality, which are cited as the conventional problems.

In the above description, the luminance parameter and the color difference parameter are tentatively expressed as the compression ratio. But,
For example, since a compression ratio can be calculated from compressed data of luminance and color difference and resolution, a parameter of luminance and a parameter of color difference may be expressed as compressed data or resolution.

Embodiment 2 FIG. 3 is a flowchart showing an image compression method according to the second embodiment of the present invention.

First, as in the first embodiment, step S1
1, S12 and S13 are performed.

Next, it is determined whether or not the data amount of the compressed data compressed in step S13 has reached a target value (step S21). The target value is, for example, the storage capacity of the storage medium at the storage destination, and is set in advance in the second embodiment.

If the result of the determination in step S21 is that the data amount of the compressed data has not reached the target value, the operation proceeds to step S22. In step S22, the luminance parameter and the chrominance parameter are updated. This update uses a predetermined function for updating. For example, the luminance parameter and the color difference parameter are updated by changing a predetermined value using a linear function. Then, step S13 is performed again.

As described above, the feedback is repeated until the data amount of the compressed data reaches the target value. If the data amount of the compressed data has reached the target value as a result of repeating the feedback, the process proceeds to step S14.
Even if the feedback is repeated a preset number of times, if the data amount of the compressed data does not reach the target value, there is a high possibility that an infinite loop will occur, so the processing is forcibly terminated.

Next, the compressed data is stored in the recording medium (step S14). Here, since the compressed data has reached the target value, all of the compressed data can be stored in the recording medium.

As described above, if the data amount after compression has not reached the target value, the luminance parameter and the color difference parameter are updated, and the image data is compressed again. This makes it possible to obtain compressed image data of a data amount that has reached the target value.

Embodiment 3 FIG. The third embodiment relates to the setting of the luminance parameter and the chrominance parameter in step S12 of FIGS.

For example, a luminance parameter and a color difference parameter are obtained as shown in FIG. First, for example, a user sets a desired compression ratio X of the entire image data (step S121). Next, based on the compression ratio X of the entire image data set in step S121, the luminance parameter (compression ratio) and the chrominance parameter (compression ratio) are determined by associating them with a predetermined function (steps S122 and S1).
23). This makes it possible to optimally set the luminance parameter and the color difference parameter from the user-desired compression ratio X of the entire image data.

Alternatively, the luminance parameter and the color difference parameter may be obtained as shown in FIG. Detecting the storage capacity of the storage medium where the compressed data is stored (step S
121). For example, a storage capacity ROM may be provided in the recording medium, and the storage capacity may be detected by reading the storage capacity from the ROM, or the storage capacity may be detected by detecting the maximum value of the address of the recording medium. May be. Next, based on the detected storage capacity, the luminance parameter (compression rate) and the color difference parameter (compression rate) are determined in association with each other by a predetermined function (step S12).
2, S123). This makes it possible to optimally set the luminance parameter and the color difference parameter from the detected storage capacity. Thus, for example, even if the storage capacity varies, all the compressed image data can be reliably stored in the storage capacity, and the image data is compressed while maintaining the best image quality according to the storage capacity. can do.

In particular, when the algorithm of FIG. 3 and the algorithm of FIG. 4 are combined, the compression ratio X is adjusted so that the amount of compressed data is closest to the storage capacity of the recording medium and does not exceed the storage capacity. The compressed data can be reduced in units of 1 byte while adjusting.

Embodiment 4 Embodiment 4 describes an image compression algorithm.

FIG. 6 is a flowchart showing the image compression algorithm used in step S13 of FIG.

First, a wavelet transform is performed on the image data (step S131).

The wavelet transform has recently attracted attention in place of DCT (discrete cosine transform). The wavelet transform is one of orthogonal transforms for transforming image data into frequency components, similarly to DCT. The wavelet transform divides the frequency components of the image into high and low frequencies, and then
Split the image vertically and then split the image horizontally. It is said that the wavelet transform has less noticeable block distortion and smaller noise than DCT. An image compression algorithm using wavelet transform or DCT does not compress image data in RGB format, but converts image data in RGB format into YCbCr (Y
(Same as UV) format and compress the YCbCr format image data. Of YCbCr, Y indicates luminance (degree of lightness and darkness), Cb indicates a color difference between intermediate colors from green to blue (blueness), and Cr indicates a color difference between green to red (degree of redness). The data amount of the luminance Y is YD, the data amount of the color difference Cb is CbD, and the data amount of the color difference Cr is CrD.
And The format of YD: CbD: CrD is, for example, 4: 2: 0, 4: 2: 2, 4: 4: 4, 4: 1: 1.
Is set as follows. In the case of 4: 2: 0, 4: 2: 2
Although the data amount is smaller than in the case of (1), the image quality of the restored image deteriorates. In the case of 4: 1: 1, 4:
The image quality is worse than in the case of 4: 4.

Next, the data obtained in step S131 is quantized (step S132). The quantization is performed based on the compression ratio set in step S12. The amount of compressed data is determined by the compression ratio. For quantization, for example, TCQ (Trellis Coded Quantization), WSQ
(Wavelet / scalar quantization) or EZW
(Embedded Zerotree Wavelet) or the like may be applied.

Next, the data quantized in step S132 is encoded (step S133). For encoding, for example, a method such as block encoding (Huffman code) or non-block code (arithmetic code) may be applied.

Next, the data encoded in step S133 is arranged in a predetermined format. The data adjusted to the predetermined format becomes compressed data (step S134).

FIG. 7 is a flowchart for obtaining a luminance parameter and a color difference parameter in step S12 in the case of wavelet transform.

The flowchart of FIG. 7 includes steps S121, S122, and S123 corresponding to FIG. In FIG. 7, step S121 includes steps S121a and S1.
21b, and step S122 includes step S122.
a, S122b, and step S123 is step S123.
123a, S123b, and S123c.

First, the compression ratio X of the image of the entire image data is input (step S121a).

Next, the resolution per pixel B (bp)
p (= bit / pixel)) is obtained from the compression ratio X (step S121b). For example, it may be obtained using Expression 1.

B = 24 / X (Equation 1) Alternatively, it may be obtained by using Equation 2.

B = D / S (Equation 2) where D is the amount of compressed data (the number of bytes × 8),
Is the total data amount (bits) of m × n pixels as shown in FIG.

Next, the luminance compression ratio KR is determined using, for example, Equation 3 (step S122a).

KR = α1 × Bα ² (Equation 3) Note that α1 and α2 are preset constants.

Next, a target value KS (byte) of the luminance data amount is obtained by using, for example, Equation 4 (step S122).
b).

KS = β1 × KRβ ² × S / 8 (Equation 4) Here, β1 and β2 are constants set in advance.

Next, the target value CS of the data amount of the color difference data
(Byte) is obtained using, for example, Equation 5 (Step S1)
23a).

CS = (S × 3 / X−KS) / 2 (Equation 5) Note that the target value CS obtained by Equation 5 is the data amount of one of Cb and Cr. Therefore, (S × 3 / X
-KS) divided by 2.

Next, the color difference resolution P (bpp) per pixel of the color difference is obtained using, for example, Equation 6 (step S).
123b).

P = CS × (S / Y) × 8 (Equation 6) CS × (S / Y) is multiplied by 8 because it is a bit unit. Y is YD: CbD: CrD = 4:
4 when 2: 0, 4 when YD: CbD: CrD = 4: 1: 1, 2 when YD: CbD: CrD = 4: 2: 2, 2 when YD: CbD: CrD = 4: 4: 4 When it is 1.

Next, the color difference compression ratio CR is determined using, for example, Equation 7 (step S123c).

CR = γ1 × Pγ ² (Equation 7) Note that γ1 and γ2 are predetermined constants.

In the fourth embodiment, the above equations 1 to 7 are predetermined functions for associating the luminance parameter with the chrominance parameter.

Note that the flowchart of FIG. 7 may correspond to the flowchart of FIG. 5 for detecting the storage capacity of the recording medium instead of FIG. Step S in this case
In 121a, for example, the compression ratio of the entire image data may be obtained as follows. That is, the storage capacity of the storage medium where the compressed data is stored is detected, and the compression ratio of the entire image data is obtained from the ratio between the detected storage capacity and the data amount of the image data input in step S11 of FIG. Good.

Also, instead of the wavelet transform, DC
T or the like may be used. DCT is used in an image compression algorithm applied to JPEG, for example. In an image compression algorithm using DCT, a coefficient indicating a frequency component is generated by orthogonally transforming a block unit (for example, 8 × 8 pixels) of pixels in an image.

In the case where the wavelet transform or DCT is used, the compressed data YD: CbD: Cr
Even if the image data is compressed so that D becomes equal to YD: CbD: CrD of the data before compression, the image obtained by reconstructing the compressed data does not always have the optimum image quality. For example, of the image data input in step S11, the image quality of a person is reduced, or the image quality of a background is reduced.

Therefore, for example, α1, α2, β1, β
The values of 2, 2, γ1 and γ2 for the person and the value for the background are set in the image processing apparatus in advance. For example, a changeover switch is provided in the image processing apparatus. If the user does not want to lower the image quality of the person than the background, the value for the person is selected by the changeover switch. As a result, the image data can be compressed so that the image quality of the person does not deteriorate. Therefore, as shown in FIG. 8, YD: CbD:
Even if CrD and YD: CbD: CrD after compression of the image data are different, the image quality may not deteriorate.

In the above description, the case where the image compression algorithm using the wavelet transform or DCT is applied to step S13 in FIG. 1 may be applied, but may be applied to step S13 in FIG. In this case, in step S22 of FIG. 3, an asymptotic exponential function F (= a ×
X ^b ) using the luminance parameter (compression rate KR = F
(X)) may be updated. X is the compression ratio of the entire image data. The color difference parameter (compression rate CR) is also updated using an exponential function as shown in FIG. Each time the process proceeds to step S22, the compression ratio X of the entire image data is changed, and the luminance parameter and the color difference parameter are updated. Further, if the exponential function is complicated and there is a problem in the amount of calculation or the calculation speed, an approximate expression using several stages of straight lines (linear function) as shown in FIG. 10 may be used instead of the exponential function.

The image compression algorithm may be other than the one using DCT or wavelet transform, and may be any algorithm that requires a luminance parameter and a color difference parameter.

[0060]

According to the first aspect of the present invention, the parameters of the luminance and the parameters of the color difference can be automatically set in association with each other by a predetermined function, so that optimal compressed data can always be obtained. Can be.

According to the second aspect of the present invention, the luminance parameter and the color difference parameter can be set optimally from the desired compression ratio of the entire image data.

According to the third aspect of the present invention, it is possible to optimally set the luminance parameter and the color difference parameter according to the storage capacity. Thus, for example, even if the storage capacity is various, it is possible to compress the image data so that the compressed image data can be reliably stored in the storage capacity, and further, keep the best image quality according to the storage capacity. Can be.

According to the fourth aspect of the present invention, if the data amount after compression has not reached the target value, the luminance parameter and the color difference parameter are updated, and the image data is compressed again. This makes it possible to obtain compressed image data of a target data amount.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an image compression method according to a first embodiment of the present invention.

FIG. 2 is a conceptual diagram illustrating an example of image data.

FIG. 3 is a flowchart illustrating an image compression method according to a second embodiment of the present invention.

FIG. 4 is a flowchart showing a part of the image compression method according to the third embodiment of the present invention.

FIG. 5 is a flowchart showing a part of an image compression method according to Embodiment 3 of the present invention.

FIG. 6 is a flowchart illustrating an image compression algorithm according to the fourth embodiment of the present invention.

FIG. 7 is a flowchart showing a part of the image compression method according to the fourth embodiment of the present invention.

FIG. 8 is a conceptual diagram illustrating an example of a data amount before and after compression of image data according to Embodiment 4 of the present invention.

FIG. 9 is a graph illustrating an example of a function for updating a parameter according to the fourth embodiment of the present invention.

FIG. 10 is a graph illustrating an example of a function for updating a parameter according to the fourth embodiment of the present invention.

[Explanation of symbols]

 DA image data YD, CbD, CrD data amount

Continued on the front page F-term (reference) 5C057 AA01 AA07 EA02 EA07 EL01 EM07 EM09 EM16 GG03 5C059 MA00 MA23 MA24 MC11 ME02 ME11 PP16 SS11 TA60 TC15 TC38 TD06 TD14 UA02 5C078 AA09 BA53 BA57 CA02 CA27 DA00 A01 BA01 BA0

Claims (4)

[Claims] 1. A method for compressing image data using an image compression algorithm requiring a luminance parameter and a chrominance parameter, wherein the luminance parameter and the chrominance parameter are associated by a predetermined function. An image compression method characterized by: 2. The image compression method according to claim 1, wherein the luminance parameter and the color difference parameter are obtained based on a desired compression ratio of the entire image data. 3. The image according to claim 1, wherein a storage capacity of a storage location of the compressed image data is detected, and the luminance parameter and the color difference parameter are obtained based on the detected storage capacity. Compression method. 4. The method according to claim 1, wherein, if the data amount of the image data after compression does not reach a target value, the luminance parameter and the color difference parameter are updated and the image data is compressed again. The image compression method according to any one of the above.