JPH0779350A - Picture data compression processing method and picture data re-configuration method - Google Patents

Abstract

PURPOSE:To provide a picture data compression processing method in which picture data are compressed at a high speed by a high compression ratio without deteriorating picture quality of an original picture. CONSTITUTION:Coefficient picture data 3 for each of plural frequency bands are obtained by applying wavelet transformation 2 to original picture data 1 representing an original picture by using a function with a longer filter toward a lower frequency band as a basic wavelet function to obtain coefficient picture data 3 for each of plural frequency bands. Thus, aliasing caused when the picture data 1 are sampled is reduced. Then quantization 4 is applied to the coefficient picture data 3 obtained by the wavelet transformation 2 with a lower bit number toward a higher frequency band and coding 5 is applied to the picture data 3 subjected to quantization 4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of compressing image data, and more particularly to a method of compressing image data for reducing the data amount of an original image by using wavelet transform and a method of recompressing compressed image data. It is about how to configure.

[0002]

2. Description of the Related Art Since an image signal carrying a halftone image such as a TV signal has an enormous amount of information, a wide band transmission line is required for its transmission. Therefore, conventionally, attention has been paid to the fact that such an image signal has large redundancy, and various attempts have been made to compress the image data by suppressing this redundancy. Further, recently, for example, recording of a halftone image on an optical disk or a magnetic disk has been widely performed, and in this case, image data compression is widely applied for the purpose of efficiently recording an image signal on a recording medium. ing.

As one of such image data compression methods, conventionally, when image data is stored or transmitted, the image data is compressed by predictive coding to reduce the amount of data. Stored, transmitted, etc. above,
At the time of image reproduction, a method is adopted in which the compressed image data (compressed image data) is subjected to decoding processing and decompressed, and a visible image is reproduced based on the decompressed image data (decompressed image data). Has been done.

As one of image data compression methods,
A method utilizing vector quantization is known. In this method, two-dimensional image data is divided into blocks of K samples, and in each of the above blocks in a codebook composed of a plurality of different vectors created by defining K vector elements in advance. The vector corresponding to the set of image data and the minimum distortion is selected, and the information indicating the selected vector is encoded in association with each block.

Since the image data in the blocks as described above have a high correlation with each other, one of the vectors in which a relatively small number of image data in each block are prepared is prepared.
It is possible to give a fairly accurate representation by using one. Therefore, transmission or recording of image data can be performed by transmitting or storing a code indicating this vector instead of actual data, so that data compression is realized. For example, the amount of image data for 64 pixels in a halftone image of a 256-level (= 8 bit) density scale is 8 × 64 = 512 bits, and these 64 pixels are regarded as one block, and each image data in the block has 64 elements. Is represented by a vector consisting of
Assuming that you create a codebook prepared as above, 1
The data amount per block is the data amount for vector identification, that is, 8 bits, and the data amount is 8 / (8 ×
64) = 1/64 can be compressed.

After the image data is compressed and recorded or transmitted as described above, the vector element of the vector indicated by the vector identification information is reconstructed data for each block,
The original image is reproduced by using the reconstructed data.

Further, as one of the methods for improving the compression rate in the case of performing the data compression by the above-mentioned predictive coding,
It is conceivable to reduce the bit resolution (density resolution) of the image data together with the predictive coding process, that is, perform the quantization process of coarsely quantizing the image data.

Therefore, the applicant of the present application has proposed an image data compression method by interpolation coding, which is a combination of the method by predictive coding and the method by quantization described above (Japanese Patent Application Laid-Open No. 62-247676). In this method, image data is divided into main data sampled at appropriate intervals and interpolation data other than the main data, and the interpolation data is interpolation prediction coding processing based on the main data, that is, interpolation data is the main data. The image data is compressed by performing interpolative prediction based on the above, and performing variable length coding (conversion into a signal whose code length changes depending on the value) such as Huffman coding with respect to the prediction error.

Further, when compressing image data, it is naturally desirable that the compression rate is high. However, it is technically difficult to expect a large improvement in the compression rate in the above-mentioned interpolation coding, and therefore, in order to achieve a higher compression rate, the image data number reduction processing for reducing the spatial resolution is combined with the above-mentioned interpolation coding. It is possible to match.

Therefore, the applicant of the present application has proposed an image data compression method capable of achieving a higher compression rate while maintaining higher image quality by combining the above-described interpolation coding and the image data number reduction process (special feature). Kaihei 2-280462).

On the other hand, a method called wavelet transform has been proposed as a method for processing the above-mentioned image data.

Here, the wavelet transform will be described.

The wavelet transform has been recently developed as a method of frequency analysis and has been applied to stereo pattern matching, data compression, etc. (OLIVIER RIOUL and MARTIN VETTERLI; Wavelets a
nd Signal Processing, IEEESP MAGAZINE, P.14-38, OCTOB
ER 1991, Stephane Mallat; Zero-Crossings of a Wavel
et Transform, IEEE TRANSACTIONS ON INFORMATION THEO
RY, VOL.37, NO.4, P.1019-1033, JULY 1991).

This wavelet transform uses a function h as shown in FIG. 8 as a basis function.

[0015]

[Equation 1]

Since the signal is converted into a frequency signal for each of a plurality of frequency bands in the equation, the problem of false vibration unlike the Fourier transform does not occur. That is, if the filtering process is performed by changing the period and the contraction rate of the function h and moving the original signal, it is possible to create a frequency signal adapted to a desired frequency from a fine frequency to a coarse frequency. For example, as shown in FIG. 9, the signal Sorg
Is wavelet transformed and inverse wavelet transformed for each frequency band, and the signal Sorg is Fourier transformed as shown in FIG. 10 and the signal is inverse Fourier transformed for each frequency band. Compared with the conversion, it is possible to obtain the frequency signal in the frequency band corresponding to the vibration of the original signal Sorg. That is, in the Fourier transform, the frequency band 7 corresponding to the part B of the original signal Sorg
While the vibration is generated in the portion B'of, the wavelet transform does not generate the vibration in the portion A'of the frequency band W7 corresponding to the portion A of the original signal Sorg like the original signal. Become.

Further, using this wavelet transform,
A method for compressing the above-mentioned image data has been proposed (Marc Antonini et al., Image Coding Using Wavelet.
Transform, IEEE TRANSACTIONS ON IMAGE PROCESSING
, VOL.1, NO.2, p205-220, APRIL 1992).

According to this method, original image data representing an image is converted into
By sequentially performing wavelet transform while sampling at a predetermined interval with a predetermined basic wavelet function, the original image data is converted into a plurality of image data with different combinations of frequency bands from the high frequency band to the low frequency band in both main and sub directions. Image data in the high frequency band, which is decomposed and carries a lot of noise components with respect to these image data, has a small number of bits or has a number of bits of 0,
The original image data is compressed by allocating a large number of bits to the image data of the low frequency band carrying the information of the main subject and performing the above-mentioned vector quantization. According to this method, the compression rate of the original image data can be improved, and the inverse wavelet transform is sequentially performed while complementing the part of the data which is thinned out at a predetermined sampling interval with respect to the compressed image data. By applying
The original image can be completely restored.

In the method of compressing image data using the wavelet transform described above, the original image data is divided into image data from the high frequency band to the low frequency band by repeating the wavelet transform and sampling at predetermined intervals. Therefore, aliasing occurs when sampling is performed. Here, aliasing refers to a phenomenon in which a high frequency component of original data is mixed with a low frequency component. Therefore, by sampling the image data, the high frequency components of the sampled image data are mixed with the low frequency components. For example,
When the sampling interval described above is sampled for each pixel of image data, the Nyquist frequency (when sampling a signal having a limited frequency band at a constant interval, the maximum sampling interval that can uniquely describe the original signal waveform Reciprocal of value)
The high frequency component of is mixed with the low frequency component, and the Nyquist frequency of the image data becomes 1/2 of the Nyquist frequency of the original image data. Due to such aliasing, when the wavelet-transformed image data is subjected to the inverse wavelet transform, the sampled data cannot be completely restored as the original image,
There is a problem that artifacts (false images) occur in the sampled image data portion.

This problem can be solved by appropriately selecting the basic wavelet function used when applying the wavelet transform to the original image data. That is, by appropriately selecting the basic wavelet function, when the wavelet-transformed image data is subjected to the inverse wavelet transform, the influence of aliasing can be mutually compensated in the low frequency band and the high frequency band. Artifacts do not occur due to aliasing caused by sampling the data.

[0021]

However, in the method of Antonini et al. Described above, the image data in the high frequency band of the image data obtained by the wavelet transform is quantized with 0 or less bits. Therefore, the image data for which the aliasing necessary for the inverse wavelet transform of the image data should be compensated becomes 0 or becomes small. Therefore, although the method of Antonini et al. Described above can improve the compression rate of image data, aliasing cannot be properly compensated when reconstructing compressed image data, and the reconstructed image cannot be corrected. The above-mentioned artifact is generated.

Such a problem is caused by increasing the filter length of the basic wavelet function at the time of performing the wavelet transform so as to increase the degree of freedom of the frequency transform response and select a function that can reduce aliasing. Although it can be solved, if the filter length of the function is made long, the calculation time for performing the wavelet transform becomes long, so that high-speed compression processing cannot be performed.

In view of the above circumstances, the present invention is capable of compressing image data with a high compression rate, and is free from the problem of artifacts generated in a reconstructed image and can be processed at higher speed. An object of the present invention is to provide a compression processing method and an image data reconstruction method.

[0024]

According to an image data compression processing method of the present invention, original image data representing an image is sampled at a predetermined interval and sequentially subjected to a wavelet transform to reduce the original image data from a high frequency band to a low frequency band. Sequential decomposition into a plurality of coefficient image data representing different frequency bands up to the frequency band, the plurality of coefficient image data at least for the coefficient image data of the highest frequency band by a smaller number of bits than the coefficient image data of other frequency bands In the image data compression processing method of quantizing and encoding the quantized coefficient image data to remove noise in high frequency components of the original image data to perform compression processing on the original image data, Of the coefficient image data, the coefficient image data in the lower frequency band reduces aliasing. That a long function of the filter length is characterized in applying the wavelet transform as the basic wavelet function.

The image data reconstructing method according to the present invention is for reconstructing original image data compressed by the image data compressing method according to the present invention, and decoding the encoded coefficient image data. The image data compression according to claim 1, wherein the decoded coefficient image data is subjected to inverse wavelet transform using a function having a long filter length that reduces aliasing as coefficient image data in a lower frequency band is used as a basic wavelet function. The original image data compressed by the processing method is reconstructed.

[0026]

The image data compression processing method according to the present invention compresses image data by performing wavelet transformation. In the image data compression processing method of Antonini et al. Described above, when performing wavelet transformation while sampling original image data. The coefficient image data in the low frequency band is characterized in that the filter length of the basic wavelet function used when performing the wavelet transform is made longer than that in the coefficient image data in the high frequency band. That is, since the image data in the high frequency band includes many components such as noise and the number of data to be processed is large, even if aliasing occurs, it does not affect the image quality of the original image so much. This is a high-speed compression process by performing a wavelet transform using a short basic wavelet function. On the other hand, the image data in the low frequency band carries important information and the number of data to be processed is small, so the wavelet transform is performed by the basic wavelet function with a long filter length to reduce aliasing. is there.

Therefore, even when the compressed reduced image data is reconstructed, an image due to aliasing does not occur in the image of the frequency band carrying important information.
Further, it is possible to perform wavelet transform at high speed on image data in a high frequency band, which has a large number of data to be processed. Therefore, as a whole, an image without artifacts can be reconstructed, the image data can be compressed at high speed, and the image data can be compressed with a high compression rate.

[0028]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing the basic concept of an embodiment of an image data compression processing method according to the present invention.

As shown in FIG. 1, the image data compression processing method according to the present invention basically uses a function having a longer filter length in a lower frequency band according to the method of Antonini et al. Wavelet transform 2 is applied as a wavelet function to obtain coefficient image data 3 for each of a plurality of frequency bands. Then wavelet transform 2
Quantization 4 is performed on the coefficient image data 3 obtained by the above with a lower number of bits as the frequency band is higher, and encoding 5 is performed on each quantized image data 3.

The details of the embodiment according to the present invention will be described below.

The present embodiment is a radiation image information recording / reproducing system using a stimulable phosphor sheet recorded in JP-A-55-12492 and JP-A-56-11395. It is intended to read a radiation image of a human body recorded on a body sheet as digital image data by laser beam scanning. As shown in FIG. 2, the radiation image is read by moving the sheet 10 in the sub-scanning direction (vertical direction) while scanning the stimulable phosphor sheet 10 with a laser beam in the main scanning direction (horizontal direction). Then, the sheet 10 is two-dimensionally scanned.

Next, wavelet transform is performed on the original image data.

FIG. 3 is a diagram showing details of the wavelet transform for the original image data Sorg.

In the present embodiment, the orthogonal wavelet transform in which each coefficient of the wavelet transform is orthogonal is performed, which is described in the above-mentioned document of Marc Antonini et al. Furthermore, in this embodiment,
It is assumed that the coefficient image data in the highest frequency band among the coefficient image data obtained by the wavelet transform is quantized with the number of bits set to 0.

First, as shown in FIG. 3, the original image data S
Filtering processing is performed by the function g _N and the function h _N obtained from the basic wavelet function in the main scanning direction of org. That is, the filtering processing for each column of pixels arranged in the main scanning direction by such functions g _N and h _{N is} performed while shifting each pixel in the sub scanning direction to obtain the original image data S.
The wavelet transform coefficient signal of the org in the main scanning direction is obtained.

Here, the functions g _N and h _N are uniquely obtained from the basic wavelet function, and the filter length is made longer as the coefficient image data becomes data in the low frequency band by the wavelet transform. This function h _N is shown in Table 1 below. In Table 1, N = 2 (function with the shortest filter length) to N = 10
(Function with the longest filter length) is shown.
Here, g _N are those obtained from h _N, the relationship between g _N and h _N denote the following equation (2).

[0038]

[Table 1]

G _N = (− 1) ⁿ h _N (2) The original image data Sorg described above is filtered by the functions g ₂ and h _{2 having} the shortest filter length, and the wavelet of the original image data is obtained. It is assumed that the conversion coefficient signals Wg0 and Wh0 are obtained. Where the function h
₂ is shown in FIG. 4 together with its frequency enhancement characteristics.

In this way, when the wavelet transform coefficient signals Wg0 and Wh0 are obtained by the functions g ₂ and h ₂ ,
With respect to the wavelet transform coefficient signals Wg0 and Wh0, pixels in the main scanning direction are sampled every other pixel, and the number of pixels in the main scanning direction is halved. Due to this sampling, aliasing occurs in which high frequency components having a Nyquist frequency of 0.5 or more are mixed with low frequency components of 0.5 or less among the Nyquist frequencies of the image data filtered by the function h ₂ . Then, filtering processing is performed by the functions g ₂ and h _{2 in} the sub-scanning directions of the wavelet transform coefficient signals Wg0 and Wh0, respectively, from which the pixels have been thinned, to obtain the wavelet transform coefficient signals WW ₀ , WV ₀ , VW ₀ and VV ₀ . obtain.

Next, the wavelet transform coefficient signal W
With respect to W ₀ , WV ₀ , VW _0, and VV ₀ , the pixels in the sub-scanning direction are sampled every other pixel, and the number of pixels in the sub-scanning direction is halved. This sampling causes aliasing as described above. As a result, each wavelet transform coefficient signal VV ₀ , WV ₀ ,
The number of pixels of VW ₀ and WW ₀ is 1/4 of the number of pixels of the original image data Sorg. Next, filtering processing is performed in the main scanning direction of the wavelet transform coefficient signal VV ₀ by the functions g ₃ and h _{3 having} a longer filter length than the functions g ₂ and h ₂ . The function h ₃ has a filter length of 6 as shown in Table 1, and the function g ₃ is obtained from the function h ₃ .

That is, the filtering process for each column of pixels arranged in the main scanning direction by the functions g ₃ and h _{3 is} performed while shifting each pixel in the sub scanning direction to obtain the wavelet transform coefficient of the wavelet transform coefficient signal VV _{0 in} the main scan direction. The signals Wg1 and Wh1 are obtained.

Here, the wavelet transform coefficient signal VV ₀
Since the number of pixels is half of the original image data by sampling in both main and sub directions, the frequency band of the image is half that of the original image data. Therefore, the wavelet transform coefficient signal VV ₀ is converted into the functions g ₃ , h
By performing the filtering process in ₃ , the wavelet transform coefficient signal VV of the frequency components of the original image data
Wavelet transform coefficient signals Wg1 and Wh1 that represent lower frequency components than the frequency component represented by ₀ are obtained.

In this way, when the wavelet transform coefficient signals Wg1 and Wh1 are obtained, for the wavelet transform coefficient signals Wg1 and Wh1, pixels in the main scanning direction are sampled every other pixel, and the number of pixels in the main scanning direction is further increased. 1 /
Set to 2. Then wavelet transform coefficient signals Wg1, W
The wavelet transform coefficient signal WW is obtained by performing the filtering process in the sub-scanning direction of each h1 using the functions g ₃ and h _3.
Obtain ₁ , WV ₁ , VW ₁ and VV ₁ .

Next, the wavelet transform coefficient signal W
With respect to W ₁ , WV ₁ , VW ₁ , and VV ₁ , the pixels in the sub-scanning direction are sampled every other pixel, and the number of pixels in the sub-scanning direction is halved. As a result, the number of pixels of each wavelet transform coefficient signal VV ₁ , WV ₁ , VW ₁ , WW ₁ becomes 1/16 of the number of pixels of the original image data Sorg.

Thereafter, in the same manner as described above, the function g ₄ , whose filter length is longer than the functions g ₃ and h _{3 in} the main scanning direction of the wavelet transform coefficient signal VV ₁ from which pixels are thinned,
The filtering process is performed by h _4, the pixels in the main scanning direction of the obtained wavelet transform coefficient signal are sampled, and the wavelet transform coefficient signal obtained by thinning out the pixels is filtered by the functions g ₄ and h _{4 in the} sub scanning direction. And wavelet transform coefficient signals WW ₂ , WV ₂ , VW ₂ and VV ₂ are obtained.

By repeating such wavelet transform N times, wavelet transform coefficient signals WW _{0 to} W
_{W N, WV 0 ~WV N,} VW 0 ~VW N, and VV _N
To get Here, the wavelet transform factor signals obtained by the wavelet transform of the N-th WW _N, WV _N, V
W _N, VV _N, as compared with the original image data by sampling the number of pixels in the main sub both directions (1/2) because it is a ^N, each wavelet transform factor signal has a low enough frequency bands N is large, It becomes the data representing the low frequency component of the frequency component of the original image data.

Therefore, the wavelet transform coefficient signal WW _i (i = 0 to N, the same applies hereinafter) is used as the original image data Sor.
It represents a change in the frequency of g in both main and sub directions, and the larger i is, the lower the frequency of the signal becomes. The wavelet transform coefficient signal WV _i represents the change in the frequency of the image signal Sorg in the main scanning direction, and the larger i is, the lower the frequency signal becomes. The frequency in the main scanning direction is lower than the frequency in the sub scanning direction. Further, the wavelet transform coefficient signal VW _i represents the change in the frequency of the image signal Sorg in the sub-scanning direction. The larger i becomes, the lower the frequency signal becomes, and the frequency in the sub-scanning direction becomes lower than the frequency in the main scanning direction. ing.

FIG. 6 shows a wavelet transform coefficient signal for each of a plurality of frequency bands. Note that, in FIG. 6, for convenience, the state up to the third wavelet transform is shown. In FIG. 6, the wavelet transform coefficient signal WW ₃ is (1/1)
2) It has been reduced to ³ .

Further, as the number of wavelet transforms is increased, the filter lengths of the functions g _N and h _N are increased as shown in Table 1, so that the aliasing generated when sampling the pixels of the wavelet transform coefficient signal is eliminated. Can be reduced. For example, in the function h ₁₀ shown in FIG. 5, the filter length is 20, and the frequency emphasis characteristic is as shown in FIG. 5 (b). That is, in the function h ₁₀ , the enhancement degree at the Nyquist frequency of 0.5 or higher is smaller than that of the function h ₂ shown in FIG. 4 (b). Therefore, sampling of pixels reduces high frequency components mixed with low frequency components having a Nyquist frequency of 0.5 or less, thereby reducing aliasing.

On the other hand, if the filter length of the function is increased, the amount of calculation for filtering increases and the processing time increases. However, due to the above-described sampling, the data amount of the wavelet transform coefficient signal in the low frequency band is high. Since it is smaller than the coefficient signal of No. 2, the delay of the compression process due to the increase of the processing time does not pose a problem.

Next, the wavelet retransform coefficient signals WV _i and VW, which have been subjected to the wavelet transform again,
Quantization is performed on _i (excluding i = 1), WW _i , and VV _i .

Among the wavelet transform coefficient signals, the wavelet transform coefficient signal in the high frequency band is
Since it carries unnecessary information such as noise, and for wavelet transform coefficient signals in the low frequency band, it carries important information such as the main subject, so the coefficient signal in the high frequency band has a lower number of bits. To quantize. That is, as shown in FIG. 6, the wavelet transform coefficient signals WW ₁ , WV ₁ and VW ₁ in the high frequency band are set to 0 bit, and the wavelet transform coefficient signal W
Quantization is performed with 1 bit for W ₂ , ₂ bits for the wavelet transform coefficient signals WV ₂ and VW ₂ , and 8 bits for wavelet transform coefficient signals of more than that.

Here, when quantizing data, the higher the number of bits, the more the data can be compressed in a state closer to the original image, but the compression rate cannot be improved so much. Further, if the number of bits is reduced, the compression rate can be improved, but the error when the compressed data is restored is large, and the noise is large compared with the original image.

Therefore, in the present invention, a small number of bits are assigned to the image data in the high frequency band that carries a large amount of noise components, and a large number of bits are assigned to the image data in the low frequency band that carries the information of the main subject. Therefore, the image quality is maintained by increasing the number of bits in the important part, and the image quality is maintained in the less important part because the image quality does not matter so much, and the compression is performed while maintaining the image quality of the main part of the image as a whole. It is designed to improve the rate.

After each wavelet transform coefficient signal is quantized as described above, compression processing is performed by performing the above-mentioned Huffman coding, predictive coding and the like.

Although the quantization level has been described as being constant for each label, the quantization level may be changed for each frequency band. For example, in the high frequency band, more quantization bits are used. Reduce the number. Further, the number of bits may be set to 0 as the quantization level, and in this case, the code length is 0, so that a high compression rate can be realized.

The original image data Sorg encoded and compressed in this way is stored in a recording medium such as an optical disk, and is stored and transported.

Next, a method of reconstructing compressed data will be described.

First, the compressed original image data is decoded by Huffman coding and predictive coding to obtain the above-mentioned wavelet transform coefficient signals WV _i ,
Obtain VW _i and WW _i .

Next, the wavelet transform coefficient signals WV _i , VW _i , WW obtained by the decoding are performed.
_i, performs inverse wavelet transform for VV _i.

FIG. 7 is a diagram showing details of the inverse wavelet transform.

[0063] As shown in FIG. 7, each wavelet transform factor signals _{VV N, VW N, WV N} , the WW _N performs processing spacing of one pixel among pixels arranged in the sub-scanning direction (× in FIG. 2). Followed by different functions h _N 'is a function h _N described above wavelet transform factor signals VV _N this spaced in the sub-scanning direction, the wavelet transform factor signal VW _N was above in the sub-scanning direction function g _N is Filtering processing is performed using different functions g _N ′. That is, the function g _N ', h _N' wavelet transform factor signal VV _N according conducted while Shifts pixel by pixel filtering processing for each pixel of a row aligned in the sub-scanning direction VW _N in the main scanning direction, the wavelet transform factor signals VV _N, to obtain an inverse wavelet transform factor signal VW _N,
The inverse wavelet transform coefficient signal WhN 'is obtained by doubling this and adding.

Here, the functions g _N , h _N and the function g _N ′,
The relationship with h _N ′ is given by the following equation (3).

H _N ′ [n] = h _N [−n] g _N ′ [n] = g _N [−n] (3) where [−n] represents left / right half rotation about the central axis of the function, , Functions g _N and h _N are those for performing inverse wavelet transformation by means of functions g _N ′ and h _N ′ which are asymmetric with respect to the central axis.

[0066] On the other hand, in parallel with this, the function h _N 'wavelet transform factor signals WV _N in the sub-function g wavelet transform factor signals WW _N in the sub-scanning direction
_N get 'performs filtering processing by, give an inverse wavelet transform factor signals of the wavelet transform factor signals WV _N, WW _N, inverse wavelet transform factor signals WgN by adding to 2 times this'.

Next, the inverse wavelet transform coefficient signal W
With respect to hN 'and WgN', a process of spacing one pixel between pixels arranged in the main scanning direction is performed. The 'function h _N in the main scanning direction' then inverse wavelet transform factor signals WhN,
'The function g _N in the main scanning direction' inverse wavelet transform factor signals WGN filtering processing by the wavelet transform factor signals WhN ', WGN' give the inverse wavelet transform factor signals, inverse by adding this double to A wavelet transform coefficient signal VVN _-1 'is obtained.

Next, the inverse wavelet transform coefficient signal VVN _-1 'and the wavelet transform coefficient signals VWN _-1 and W
1 between pixels arranged in the sub _- scanning direction for V _N-1 and WW _N- 1
Performs a process for spacing pixels. Thereafter, the inverse wavelet transform coefficient signal VVN _-1 'is added to the function h in the sub-scanning direction.
_The wavelet transform coefficient signal VW _N-1 is filtered by _N-1 'in the sub _- scanning direction by the function g _N-1 '. That is, the function h _N, g short filter length than _N functions _{g N-1 ', h N} -1' wavelet transform factor signal VV _N-1 'according, one row of pixels aligned in the sub-scanning direction of the VW _N-1 Each filtering process is performed by shifting one pixel in the main scanning direction to obtain the wavelet transform coefficient signal VV.
_An inverse wavelet transform coefficient signal WhN-1 'is obtained by doubling and adding the inverse wavelet transform coefficient signals _N-1 ' and VW _N-1 .

On the other hand, in parallel with this, the wavelet transform coefficient signal WV _N-1 in the sub _- scanning direction is given by the function h _N-1 ′, and the wavelet transform coefficient signal WW _N-1 is given in the sub _- scanning direction by the function g _N-1. get 'performs filtering processing by, give an inverse wavelet transform factor signals of the wavelet transform factor signals _{WV N-1, WW N-} 1, the inverse wavelet transform factor signals WGN-1 by adding to 2 times this' .

Next, the inverse wavelet transform coefficient signal W
1 between pixels lined up in the main scanning direction for hN-1 'and WgN-1'
Performs a process for spacing pixels. Then inverse wavelet transform factor signals WhN-1 'to the main scanning direction function h _N-1'
The inverse wavelet transform coefficient signal WgN-1 'is filtered in the main scanning direction by the function g _N-1 ' to obtain inverse wavelet transform coefficient signals WhN-1 'and WgN-1'. To obtain the inverse wavelet transform coefficient signal VV _N-2 ′.

Hereinafter, the inverse wavelet transform coefficient signal VV _i ′ (i = −1 to N) is sequentially created, and finally the inverse wavelet transform coefficient signal VV ₋₁ ′ is obtained. This final inverse wavelet transform coefficient signal VV ₋₁ ′ is the original image data S
Image data representing org.

The wavelet transform coefficient signal VV _-1 ′ thus obtained is sent to an image reproducing device (not shown) for reproduction of a radiation image.

This reproducing device may be a display means such as a CRT or a recording device for performing optical scanning recording on the photosensitive film.

In this way, the original image data Sorg is wavelet-transformed to obtain image data for each of a plurality of frequency bands, and the high-frequency band data of this image data is again wavelet-transformed to obtain a plurality of frequency bands. Of the important part of the data is quantized by increasing the number of bits for the part that carries important information, and the unimportant part is quantized by decreasing the number of bits. It is possible to improve the data compression rate while maintaining the image quality.

Here, each wavelet transform coefficient signal V
_{V N, VW N, WV N} , an interval of one pixel among the pixels for WW _N, the function g _N ', h _N' data for the portion of the pixel spaced by performing an inverse wavelet transform by Although it is formed, this is different from the original image data because it is formed by being restored by extracting the high-frequency component that was mixed with the low-frequency component of the image data by the above-mentioned aliasing. That is, an artifact has occurred. However, for low frequency band coefficient image data,
Since the aliasing is small and this artifact is also small, it is possible to obtain an image with few artifacts for the image data in the low frequency band carrying important information.

Although the functions g _N and h _N for performing the wavelet transform shown in Table 1 are used in the above-mentioned embodiment, the present invention is not limited to this and the wavelet transform coefficient in the low frequency band is used. Any function may be used as long as the signal has a longer filter length.

In addition to this, any function may be used as long as it can perform wavelet transformation, and for example, a function that is not orthogonal but biorthogonal may be used.

Further, as shown in Table 1, FIG. 4 and FIG. 5, the wavelet transform may be performed not only by using a function which is not symmetrical about the central axis but also by using a function which is symmetrical about the central axis. Is. When the wavelet transform is performed using the bilaterally symmetric function as described above, the function for performing the wavelet transform and the function for performing the inverse wavelet transform have the same shape.

Further, in the above-mentioned embodiment, the embodiment in which the original image data representing the radiation image is compressed is described, but the image compression processing method according to the present invention can be applied to a normal image. .

For example, a description will be given of an embodiment in which an image of a 35 mm negative film in which a person or the like is recorded as a main subject is compressed. First, the negative film is read by a digital scanner, image data representing this image is obtained, and this image data is obtained. With respect to the lower frequency band as described above, the wavelet transform is performed by performing the filtering process with the functions g _N and h _{N having} a longer filter length.

Then, as in the above-described embodiment, the high frequency band portion is quantized by a low bit number, the low frequency band portion is quantized by a high bit number, and the image is encoded as necessary. Compress the data.

The original image data can be reconstructed by decoding the compressed image data in the same manner as in the above-mentioned embodiment and further applying the inverse wavelet transform.

By performing the compression process in this way,
While maintaining the image quality of important parts for normal images,
The data compression rate can be improved.

Further, in the above-mentioned embodiment, when quantizing the wavelet transform coefficient signal, the number of bits is set to 0 for the coefficient signal in the high frequency band. However, the present invention is not limited to this, and it is not limited to this. Any number of bits may be used as long as it is lower than the number of bits used when quantizing the coefficient signal of.

[0085]

As described above in detail, in the image data compression processing method according to the present invention, when the coefficient image data for each of a plurality of frequency bands is obtained by the wavelet transform, the coefficient image data in the lower frequency band is aliased. A wavelet is applied as a basic wavelet function with a function having a long filter length that reduces the. Therefore, the coefficient image data in the low frequency band carrying important information is less likely to cause aliasing, and it is possible to reduce the artifacts caused by aliasing when the image data is reconstructed.

Further, as for the image data in the high frequency band where the number of data to be filtered is large, an artifact is generated but it does not carry very important information, so the filtering speed is improved by the frequency having a short filter length. It is a thing. Therefore, the image data in the high frequency band is quantized with a smaller number of bits than the image data in the low frequency band, so that the image data can be obtained as a good image without artifacts while improving the data compression rate. Can be compressed, and the time for performing compression processing can be shortened.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic concept of an image data compression processing method according to the present invention.

FIG. 2 is a diagram showing a method of reading image data used in the present invention.

FIG. 3 is a diagram showing details of a wavelet transform.

FIG. 4 is a diagram showing a basic wavelet function with a short filter length.

FIG. 5 is a diagram showing a basic wavelet function with a long filter length.

FIG. 6 is a diagram showing a wavelet transform coefficient signal.

FIG. 7 is a diagram showing details of inverse wavelet transform.

FIG. 8 is a diagram showing a basic wavelet function used for wavelet transform.

FIG. 9 is a diagram for explaining a wavelet transform.

FIG. 10 is a diagram for explaining a Fourier transform.

[Explanation of symbols]

10 stimulable phosphor sheet h, h ', g, g ' function VV _i for performing a wavelet _{transform, VW i, WV i, WW} i (i = 1~n) wavelet transform factor signals

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location H04N 7/30 // G06T 5/20 9191-5L G06F 15/68 400 A

Claims (2)

[Claims] 1. A plurality of coefficient image data representing different frequency bands from a high frequency band to a low frequency band by sequentially performing wavelet transformation while sampling original image data representing an image at a predetermined interval. Sequentially, the plurality of coefficient image data are quantized by at least the coefficient image data of the highest frequency band with a smaller number of bits than the coefficient image data of other frequency bands, and the quantized coefficient image data is encoded. In the image data compression processing method of removing the noise in the high frequency component of the original image data by applying compression to the original image data, the coefficient image data in the lower frequency band among the plurality of coefficient image data A function with a long filter length that reduces aliasing is used as the basic wavelet function. Image data compression processing method characterized by performing the wavelet transform. 2. A function having a long filter length, which decodes the coded coefficient image data and reduces aliasing for coefficient image data in a lower frequency band, is used as a basic wavelet function. An image data reconstructing method, characterized in that the original image data compressed by the image data compressing method according to claim 1 is reconstructed by performing an inverse wavelet transform.