JPH10301550A - Image display method, image output system and image processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To display an image while gradually increasing resolution by using a wavelet conversion. SOLUTION: The data converted by a wavelet conversion part 43 and compressed by a data compression part 44 are transmitted through a communication channel 60, and the data are expanded by an expansion part 51 to be inputted to a wavelet inverse conversion part 52. The image data obtained in the process to successively inverse-convert wavelet from a low-order frequency component toward a high-order frequency component are written in a frame buffer 54 by a memory control part 53 to be display on a display 55. First of all, a rough outline of an original image is displayed, and thereafter, minute contents are displayed gradually.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image processing,
In particular, the present invention relates to a technique for displaying an image from a low resolution to a gradually increased resolution.

[0002]

2. Description of the Related Art Currently, applications for displaying images on a display are increasing. Examples include a WWW (World Wide Web) browser, Adobe (registered trademark) Photoshop (registered trademark), and the like.

[0003] The resolution of the display and the printer is generally about 72 dpi for the former, while it is 300 dpi or more for the latter. If you have an image with a resolution that can be output to a printer, the size of the image is large, and when you display it on a display,
It is necessary to reduce the resolution by some means.

[0004] However, although the speed of the processing device has been improved and the capacity of the memory has been increased, such processing requires a considerable amount of time and memory, so that the user has to wait for the display to be completed. . In particular, when image data is acquired through a low-speed line such as WWW, the larger the data, the longer the waiting time.

[0005] Rather than making the display wait until a complete image is created, only a rough outline is displayed first, and then gradually detailed contents are displayed. Discomfort seems to be reduced. Also, if images from high resolution to low resolution can be generated at the same time, it is possible to support output devices having different resolutions.

At present, some WWW browsers realize such a function. It sends and displays information obtained by thinning the original image data in the row direction, and then,
This is achieved by sending decimated information. However, in this method, the image data needs to be data of a size to be displayed on the screen, and when outputting to a high-resolution output device, the image is converted to a high resolution by some method, or Image must be prepared separately.

[0007]

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and a main object of the present invention is to provide a method and system for displaying an image gradually and clearly using a wavelet transform. Provided is an image display method and system suitable for use for efficiently checking the contents of an image, and a system for generating and outputting image data suitable for an image output device having various resolutions from one image. It is in. Other objects of the present invention will become apparent from the following description.

[0008]

In order to achieve the above object, an image display method provided by the present invention converts image data converted using a wavelet transform from a low-dimensional frequency component to a high-dimensional frequency component. The inverse transform is performed toward the components, and images obtained in the process of the inverse transform are sequentially displayed (claim 1). Another feature is that the inverse conversion process of the image data is terminated when an image of a predetermined resolution is displayed (claim 2), and the horizontal and vertical sizes of the original image before conversion are set. Is not a power of two,
The original image is expanded to an image having a power of 2 in the horizontal and vertical directions, then transformed using wavelet transform, and the image display obtained in the inverse transformation process of the transformed image data is displayed. In this case, only the range of the original image is cut out and displayed (claim 3).

Further, the image output system provided by the present invention is a wavelet inverse transform for performing a wavelet inverse transform from a low-dimensional frequency component to a high-dimensional frequency component on the wavelet-transformed image data of the original image. Means, means for writing data of an image obtained in the process of inverse wavelet transform by the inverse wavelet transform means to a frame buffer, and image output means for displaying or printing the data written to the frame buffer. It is a feature (claim 4). Another feature is that the apparatus further comprises means for terminating the inverse wavelet transform process when image data of a predetermined resolution is obtained in the inverse wavelet transform process by the inverse wavelet transform means. 5) If the image data input to the inverse wavelet transform means is image data that has been subjected to wavelet transform after expanding the original image to an image having a power of 2 in the horizontal and vertical directions, The means for writing image data to the frame buffer cuts only data corresponding to the range of the original image before expansion of the image obtained in the process of the inverse wavelet transform by the inverse wavelet transform means. (Claim 6).

An image processing system provided by the present invention comprises an image compression system and an image output system. The image compression system performs a wavelet transform on original image data, and a wavelet transform means. Compression means for compressing the image data converted by the means, wherein the image output system expands the data received by the compression means of the image compression system received via a communication channel or the like. Means, wavelet inverse transform means for performing inverse wavelet transform from low-dimensional frequency components to high-dimensional frequency components on the data expanded by the expansion means, and a wavelet inverse transform by the wavelet inverse transform means. Image data obtained in the process Means for writing, characterized by comprising the image output unit for displaying or printing the data written to the frame buffer (claim 7). Another feature is that the image output system terminates the wavelet inverse transform process when data of an image of a predetermined resolution is obtained in the wavelet inverse transform process by the wavelet inverse transform means. The image compression system may further include image expansion means for expanding the original image into an image having a power-of-two horizontal and vertical size, and the wavelet transform means includes: The means for performing a wavelet transform on the data of the image expanded by the image expansion means, and writing the data of the image of the image output system to the frame buffer,
Cutting out only data corresponding to the range of the original image before expansion from the image obtained in the process of the inverse wavelet conversion by the inverse wavelet conversion means and storing it in the frame buffer (claim 9); The means is to create an extended image by repeatedly using the data of the original image.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, when creating feature quantities of an image, each feature quantity of a 1/4 image, then each feature quantity of a 1/16 image, By gradually creating the feature amount of the subdivided image and displaying it in reverse order, it will be displayed from the rough outline gradually to the finer part, and the original image data will be made high resolution In addition, when the resolution to be displayed is reached, the display is discontinued to generate an image of an arbitrary resolution.

Wavelet transform is used to create image feature values. There are many literatures on the wavelet transform such as "Wavelet Beginner's Guide" (Susumu Sakakibara, Tokyo Denki University Press). The wavelet transform is a transform used in signal processing, image compression, and the like, and its outline is as follows.

An image is a two-dimensional signal and can be expressed as f (x, y). In the wavelet transform, f (x, y) is represented by the following function.

[0014]

(Equation 1)

The expansion coefficient of this function is decomposed as follows.

[0016]

(Equation 2)

In the subscripts LL, HH, LH, and HL of the above equation, L represents a low frequency component, and H represents a high frequency component. A similar equation can be obtained when restoring the original image. An image is obtained by displaying a value corresponding to the value of this function at each pixel (x, y) of the image. In the case of color, f (x, y) corresponds to each of RGB.

When the image data is decomposed by the decomposition algorithm of equations (1) to (5), the data is divided into four data sets of LL component, HH component, LH component, and HL component. And the entire size is the same as the original image. However, the main part of the image is actually included in the LL component, and an image similar to the original image can be obtained even if only the LL component is displayed. This is used to perform 1/4 compression.

When the decomposition (formula (2)) for extracting the LL component from the original image data is repeatedly applied, a data sequence in which the data amount is reduced to に each time is applied.

[0020]

[Outside 1]

Can be obtained. This is called Labracien Pyramid. Such a decomposition is called a pyramid algorithm in the field of image compression.

In actual image compression, data values are quantized or coded in addition to being simply decomposed into LL components. This is basically the case. By displaying the values of the Laplacian pyramid sequentially from the top (in order from low-dimensional frequency components to high-dimensional frequency components), a clearer image can be obtained gradually. In addition, since the Laplacian pyramid can be considered to calculate the feature amount of the area obtained by dividing each matrix of the image into two, if the size to be displayed is smaller than the size of the original image, it is displayed with that size. By stopping the operation, an image having a required resolution can be obtained.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is an explanatory diagram of a three-stage wavelet transform. In FIG. 1, (a) is an original image. By the first stage wavelet transform on this original image,
As shown in (b), four of LL1, HL1, LH1, and HH2
Is decomposed into two frequency components. At the time of conversion, since the sub-sampling of 1/2 in both the horizontal and vertical directions is performed, the size of the image of each component is reduced to 1/4 of the original image (the size in each of the horizontal and vertical directions is reduced to 1/4). 2). LL
One component is a "blurred" version of the original image, but most of the information in the original image is included here. Therefore, in the second stage wavelet transform, the LL1 component is treated as an original image. Therefore, the LL1 component is decomposed into four components LL2, HL2, LH2, and HH2 as shown in (c) by the second stage wavelet transform. The size of the image of each of these components is 1/4 of the image of the LL1 component, and is therefore 1/16 of the original image. Similarly, by the third stage wavelet transform,
As shown in (d), the LL2 component is LL3, HL3, L3.
It is decomposed into four components, H3 and HH3. The size of the image of each of these components is 1/4 of the LL2 image, that is, 1 of the original image.
/ 64.

FIG. 2 is a diagram showing such a three-stage wavelet transform as a filter sequence. In FIG. 2, an original image 10 is applied with a horizontal low-pass filter 11 and a horizontal high-pass filter 12. The processing of these two filters and each of the other filters includes サ ブ subsampling. A vertical low-pass filter 15 and a vertical high-pass filter 16 are applied to the output of the horizontal high-pass filter 12. The outputs of these two filters are the LH1 and HH1 components, respectively. A vertical low-pass filter 13 and a vertical high-pass filter 14 are applied to the output of the horizontal low-pass filter 11. The output of the vertical high-pass filter 14 is the HL1 component. Vertical low-pass filter 13
Is the LL1 component. This is the first stage of the wavelet transform.

A horizontal low-pass filter 21 and a horizontal high-pass filter 22 are applied to the LL1 component.
Vertical output low-pass filter 2 for filter 21
3 and a vertical high-pass filter 24 are applied, and a vertical low-pass filter 25 and a vertical high-pass filter 26 are applied to the output of the filter 22. This is the second stage of the wavelet transform, and the filter 2
The outputs of 3, 24, 25 and 26 are LL2 and HL, respectively.
2, LH2 and HH2.

A horizontal low-pass filter 31 and a horizontal high-pass filter 32 are applied to the LL2 component, and a vertical low-pass filter 3 is applied to the output of the filter 31.
3 and a vertical high-pass filter 34 are applied, and a vertical low-pass filter 35 and a vertical high-pass filter 36 are applied to the output of the filter 32. Up to this point, the wavelet transform of the third stage is performed.
The outputs of 34, 35, and 36 are LL3, HL3, L
It is a component of H3 and HH3.

In the following description, it is assumed that the above-described three-stage wavelet transform is performed.

FIG. 3 shows a schematic configuration of an example of an image processing system according to the present invention. In FIG. 3, reference numeral 40 denotes an image compression system, and 50 denotes an image output system. The image compression system 40 includes an original image memory 41 for storing the original image, an image expanding unit 42 for expanding the original image as necessary (described later), a wavelet transform unit 43 for performing the wavelet transform as described above, and a wavelet transform result. On the other hand, a data compression unit 44 that performs quantization, encoding, and the like to compress the data amount, a control unit 45 that controls the above units and controls the transfer of compressed data to an image output system, and a user (not shown). It consists of parts for making various settings.

The data compressed by the data compression unit 44 is sent to the image output system 50 via the communication channel 60. Here, the above-described three-stage wavelet transform is performed by the wavelet transform unit 43, and FIG.
LL3, HL3, LH3, HH3, HL shown in (d)
The compressed data of each component of 2, 2, LH2, HH2, HL1, LH1, and HH1 is sent to the image output system 60 in this order.

The image output system 50 uses the communication channel 6
0, a decompression unit 51 for decompressing the compressed data received via
A wavelet inverse transform unit 52 that performs a wavelet inverse transform on the data expanded by the expander 51, a memory management unit 52 that manages the writing of the image data after the wavelet inverse transform into the frame buffer 54, and the frame buffer 54. A display 55 and a printer 56 for displaying and printing data on the screen and a printer 56; a control unit 57 for controlling these units and receiving data from the image compression system 40; and a user, not shown, for making various settings. Consisting of parts. The memory management unit 52 basically includes a data sequence input from the inverse wavelet transform unit 52.

[0032]

[Outside 2]

Frame buffer 54 corresponding to k and l
The value is written to the address of the image, and an operation of enlarging or cutting the image is also performed as necessary (described later).

Next, when the operation mode in which gradually clear images are sequentially displayed and the original image is finally displayed unless the end instruction is given by the user is specified by the user, the wavelet of the image output system 50 is used. The processing of the inverse conversion unit 52 and the memory management unit 53 will be described.

FIG. 4 shows a schematic flow of the processing in this case. As described above, LL3, HL3,. . . , Compressed data of each component data is received, decompressed by the decompression unit 51 and input to the inverse wavelet transform unit 52 in order. As shown in FIG. 4, each time the component data is input, the inverse wavelet transform unit 52 performs the inverse wavelet transform, and the memory management unit 53 writes the data after the inverse transform into the frame buffer 54 (step 102). Is repeated. Note that the user can specify that the intermediate image obtained at each stage of the wavelet inverse transform process be displayed in the same size as the original image. In this case, in step 102, the memory management unit 53 The intermediate image data is written into the frame buffer 54 after being enlarged by a necessary magnification (step 1).
02). When the control unit 57 determines that the final image, that is, the original image is displayed, or when it is determined that a termination instruction is given by the user (step 100), a series of processing is terminated. Although not shown in FIG. 4, when the user wants to print out the finally displayed image, if the user gives a print instruction, the image data in the frame buffer 54 under the control of the control unit 57, that is, the display The data of the middle image is printed out by the printer 56.

The processing of the inverse wavelet transform unit 52 and the memory management unit 53 until the original image is displayed will be described step by step.

<a> Upon receiving the LL3 data from the decompression unit 51, the inverse wavelet transform unit 52 outputs the data as it is to the memory management unit 53 as inverse transform data. The memory management unit 53 writes the LL3 data into the frame buffer 54 as it is. At this time, an image “blurred” over the entire size of 1/64 of the original image is displayed on the display 55. If an image display of the same size as the original image is designated, the memory management unit 54
The data obtained by enlarging the three data in the horizontal direction and the vertical direction by eight times, that is, data obtained by enlarging one pixel to 8 × 8 pixels, is written to the frame buffer 54.

<B> Upon receiving the next HL3 data from the decompression unit 51, the inverse wavelet transform unit 52
3 data, inputted LL3 data, and not inputted LH3
Wavelet inverse transform is performed using each data of HH3 (this is not inputted and all are treated as 0), and LL2
It outputs considerable data (this is incomplete data of LL2 and is represented by LL2a). The memory management unit 53
This LL2a data is written to the frame buffer 54 as it is. At this time, an image that is slightly clearer than the image displayed in step a is displayed. However, the size of the image is 1/16 of the original image. Therefore, when the image display of the same size as the original image is designated, the memory management unit 53 enlarges the LL2a data four times in the horizontal and vertical directions, that is, enlarges one pixel to 4 × 4 pixels. And then write to the frame buffer 54.

<C> Upon receiving the next LH3 data from the decompression unit 51, the inverse wavelet transform unit 52 uses the input LL3, HL3 and the uninput HH3 (all treated as 0) to perform the inverse wavelet inverse. Conversion is performed, and data equivalent to LL2 (this is incomplete data of LL2 and is represented as LL2b) is output. The memory management unit 53 writes the LL2b as it is in the frame buffer 54 (the displayed image is 1/16 the size of the original image), or is designated to display an image of the same size as the original image. If the LL2b data is horizontal,
The image is enlarged four times in the vertical direction, that is, one pixel is enlarged to 4 × 4 pixels and written into the frame buffer 54. In this way, a slightly clearer image is displayed than in step b.

<D> Upon receiving the HH3 data from the decompression unit 51, the inverse wavelet transform unit 52 performs inverse wavelet transform using the input LL3, HL3, and LH3, and restores the LL2 data. The memory management unit 53 writes the LL2 data as it is in the frame buffer 54, or when the display of the same size as the original image is designated, enlarges the LL2 data four times in the horizontal and vertical directions. One pixel is enlarged to 4 × 4 pixels and written to the frame buffer 54. In this way, a slightly clearer image is displayed than in step c.

<E> Upon receiving the next HL2 data from the decompression unit 51, the inverse wavelet transform unit 52 adds the LL2 data obtained in step d, the uninput LH data,
2, HH2 data (all are treated as 0 here)
Is used to perform inverse wavelet transform to obtain data corresponding to LL1 (because this is incomplete data of LL1, LL1a
Is expressed). The memory management unit 52 uses this LL1
a is written to the frame buffer 54 as it is (an image having a size of 1/4 of the original image is displayed), or when the image display of the same size as the original image is designated, the LL1a data is Then, the image is enlarged twice in the vertical direction, that is, one pixel is enlarged to 2 × 2 pixels and then written to the frame buffer 54. Thus, a clearer image is displayed.

<F> When receiving the next LH2 data from the decompression unit 51, the inverse wavelet transform unit 52 additionally receives the LL2 data obtained in step d and the input H
Wavelet inverse transformation is performed using L2 data and non-input HH2 data (all treated as 0), and data corresponding to LL1 (LL1b because it is incomplete data of LL1)
Is expressed). The memory management unit 53 uses the LL1
b data is directly written into the frame buffer 54,
Alternatively, when an image display having the same size as the original image is designated, the LL1b data is enlarged twice in the horizontal and vertical directions, that is, one pixel is enlarged to 2 × 2 pixels and then written to the frame buffer 54. . Thus, a clearer image is displayed than in step e.

<G> Upon receiving the next HH2 data from the decompression unit 51, the inverse wavelet transform unit 52 additionally receives the LL2 data obtained in step d, the input H
Inverse wavelet transform is performed using each data of L2 and LH2 to restore LL1 data. Memory management unit 53
Transmits the LL1 data as it is to the frame buffer 54.
Or if an image display of the same size as the original image is designated, the LL1 data is doubled in the horizontal and vertical directions and then written to the frame buffer 54. In this way, an image is displayed that is less sharp than the original image but allows the contents of the original image to be checked to a considerable extent.

<H> Upon receiving the next HL1 data from the decompression unit 51, the inverse wavelet transform unit 52 additionally receives the HL1 data obtained in step g, the uninput LH data
Inverse wavelet transform is performed using each data of HH1 and HH1 (all data is treated as 0), and data corresponding to the original image is represented (original image a because it is incomplete data of the original image). The memory management unit 53 stores the original image a data in the frame buffer 5
Write to 4. Thus, a clearer image is displayed than in step g.

<I> Upon receiving the next LH1 data from the decompression unit 51, the inverse wavelet transform unit 52 additionally receives the LL1 data obtained in step g, the input H
Wavelet inverse transform is performed using L1 data and uninput HH1 data (all are treated as 0), and data corresponding to the original image (indicated as original image b since it is incomplete data of the original image) is output. The memory management unit 53 writes the original image b data into the frame buffer 54. As a result, an image very close to the original image is displayed.

<J> Upon receiving the last HH1 data from the decompression unit 51, the inverse wavelet transform unit 52 uses the LL1 data obtained in step g and the input HL1 and LH data in addition to this. The original image data is restored by performing the inverse wavelet transform. The original image data is written into the memory management unit 54, so that a complete original image is displayed.

As described above, the data of the image of the higher resolution is gradually generated from the data of the image of the lowest resolution of the same original image on the frame buffer 54 in order, and the display 55
Will be displayed. For the purpose of confirming the content of the original image, if the user can confirm the content at the time when the low-resolution image is displayed, the user may input an end instruction and terminate the processing there, and No need to wait for the display
The time required for confirmation can be reduced. If the user wants to print out an image with the printer 56, the user waits until image data of a resolution suitable for the resolution of the printer 56 is generated in the frame buffer 54, and then inputs a print instruction. Can be printed.

Next, the processing of the inverse wavelet transform unit 52 and the memory management unit 53 of the image output system 50 when the user specifies a mode for restoring and displaying an image of a preset size will be described. .

FIG. 5 is a flowchart of the setting procedure in this case. In step 201, the horizontal and vertical sizes x and y of the image to be displayed are specified. The size of the image specified here is the size of the image of the output data of the inverse wavelet transform unit 52 (in other words, x, y
Is the resolution of the image you want to display.) If the original image has a size of 2048 × 2028 pixels and it is desired to display an image corresponding to the LL2 component, for example, x = y = 512 may be specified.

FIG. 6 shows a schematic flow of the processing in this case. As described above, LL3, HL3,. . . , Compressed data of each component data is received, decompressed by the decompression unit 51 and input to the inverse wavelet transform unit 52 in order. As shown in FIG. 6, each time the component data is input, the inverse wavelet transform unit 52 performs the inverse wavelet transform, and stores the data after the inverse transform in the memory management unit 5.
3 is written into the frame buffer 54 (step 20).
4) is repeated. The control unit 57 checks whether the size of the image obtained by the inverse wavelet transform of the next input data exceeds the designated image size x, y before the start of the inverse wavelet transform (step 202). ). When it is determined that the size does not exceed the designated size, the wavelet inverse transform unit 52 is operated. When it is determined that the size exceeds the designated size (in other words, even the intermediate image having the designated resolution is displayed. A series of processing ends. Needless to say, if the user instructs, the finally displayed image data can be printed out.

For example, when three-stage wavelet transform data of an original image having a size of 2048 × 2048 pixels is input to the image display system, the inverse wavelet transform unit 5
2 and the memory management unit 53 sequentially execute the processing from step a to step d (however, the image is not enlarged). However, L displayed in step e
The size of the L1 image is 1024 × 1024 pixels, and x
= Y = 512, if the next step e is executed, the size of the image LL1a exceeds the specified size.
At this stage, the control unit 57 ends the processing.

As described above, by specifying the size or resolution of an image, an image up to the size or resolution can be displayed. Therefore, for example, if you want to check the contents of many high-resolution original images in a short time,
By designating the minimum required image size or resolution for confirmation, the time required to confirm the contents of many original images can be reduced. Also, when the resolution of the display 55 or the printer 56 is lower than the resolution of the original image, image data corresponding to the lower resolution can be automatically generated and displayed or printed.

In the wavelet transform, by continuing the transform, the area of the original image data is divided by 1 / in both the horizontal and vertical directions.
Repeat step 2. In order to take advantage of this feature, the size of the original image in the horizontal and vertical directions should be 2 n (n is a positive integer). However, in general, the original image is not always such a size.

According to the present invention, when the image data is subjected to the wavelet transform, the size of the image in the horizontal and vertical directions is set to two.
When the image is restored and displayed, the original size of the original image is known so that an extra portion is not displayed when the image is restored and displayed. Processing when the user specifies such an operation mode will be described with reference to FIGS.

FIG. 7 is a flowchart of a user setting procedure in this case. First, horizontal and vertical sizes x and y (in other words, resolution) of an image to be displayed are set (step 301). Further, since the horizontal and vertical sizes of the original image are expanded to the power of 2 to perform the wavelet transform, the original horizontal and vertical sizes h,
1 is also set (step 302).

FIG. 8 is a schematic flowchart of the processing in the image compression system 40. In the image compression system 40, as shown in FIG. 8, the image expansion unit 42 expands the original image stored in the original image memory 41 so that the size in the horizontal and vertical directions becomes 2 n. To the wavelet transform unit 43, and the wavelet transform unit 43 performs a first-stage wavelet transform on the expanded image (step 310). If the original size of the original image is, for example, 800 × 600 pixels, the image expansion unit expands the original image to an image having a size of 1024 × 1024 pixels. In this case, h = 800 and l = 600 are designated. If it is desired to finally restore and display the LL1 image, for example, specify x = y = 512. Hereinafter, a description will be given assuming that such designation has been made.

The extension of the original image can be performed by filling all the extended portions with 0 or 1 pixels. In this embodiment, however, a method of repeatedly using the original image to fill the extended portion, That is, a method is employed in which data of the extended image is provided to the wavelet transform unit 43 by repeatedly reading data at the same position on the original image memory 41. FIG. 11 is an explanatory diagram of such an image extension. In FIG. 11, reference numeral 501 denotes an original image;
Is the extended image. Original image 501 of extended image 502
Is an extended portion where the original image data is repeatedly used.

Since the data compression utilizes the correlation with the surrounding data, the data compression ratio of the expanded original image can be increased by adopting the latter extension method as compared with the former method. Since an extended original image is not created on the original image memory 41, it is not necessary to increase the size of the original image memory 41.

As described above, the wavelet transform unit 43 executes the second and third stages of wavelet transform on the result of the first stage wavelet transform of the extended original image (step 311). The conversion result data is subjected to data compression by the data compression unit 44, and the image display system 5
0 (step 312).

FIG. 9 shows a schematic flow of processing of the image display system in this case. As described above, LL3, HL
3,. . . , Compressed data of each component data is received, decompressed by the decompression unit 51 and input to the inverse wavelet transform unit 52 in order. As shown in FIG. 9, each time the component data is input, the inverse wavelet transform unit 52 performs the inverse wavelet transform, and the memory management unit 53 corresponds to the h, l range of the data after the inverse transform. Only the data (that is, the data corresponding to the portion excluding the extension portion of the extension original image) is cut out and the frame buffer 5
4 (step 322) is repeated.

The control unit 57 checks whether the size of the image obtained by the next input inverse wavelet transform exceeds the designated image size x, y before the start of the inverse wavelet transform (step). 321). And
When it is determined that the size does not exceed the designated size, the wavelet inverse transform unit 52 is operated, but when it is determined that the size exceeds the designated size, a series of processes is terminated.

As described above, even if the original image is extended, the image excluding the extended portion is gradually displayed from the smallest image (the image with the lowest resolution) to the larger (high-resolution) image. In addition, since the original image is expanded so as to have a power of 2 in the horizontal and vertical directions, an original image having an arbitrary size can be efficiently processed.

It should be noted that, at the stage of each intermediate image, data is not cut out to remove the extended part, but is cut out only at the time of displaying the finally displayed image. You may. FIG. 10 is a flowchart showing such an example. 10, steps 421 and 422 correspond to steps 321 and 322 in FIG. 10, respectively. However, in step 422, all the image data obtained by the inverse wavelet transform is written into the frame buffer 54. That is, a part corresponding to the extended part of the original image is displayed as the intermediate image to be subjected to the step.

If it is determined in step 421 that the size of the image exceeds the designated x and y, the process proceeds to step 42
Proceed to 4. In this case, if it is desired to finally display the LL1 image, x and y need to be set to values larger than the size of LL2 but smaller than the size of LL1. If it is set as such, when the LL1 data is input to the inverse wavelet transform unit 52, step 4 is executed.
The process proceeds to step 24. The LL1 data is inversely transformed by the inverse wavelet transform unit 52, and the h, l
Are cut off by the memory management unit 53 and written into the frame buffer 54. Thus, the LL1 image from which the extended portion has been removed is displayed on the display 55.
Will be displayed.

[0065]

As is apparent from the above description, the image display method of the present invention allows an image to be displayed with a gradually increasing resolution from a low resolution. It is effective when you want to observe, and you can specify the final resolution of the displayed image, so for example, when browsing a large number of images, display images with the minimum required resolution By doing so, efficient content confirmation is possible.
An image having an arbitrary size that does not have a power of can be displayed in a similar manner.

As is apparent from the above description, the image output system and the image processing system according to the present invention gradually shift the high-resolution image data from the low-resolution image data representing the rough outline of the original image to the frame buffer. In order to display images that gradually become sharper as described above, and to generate an image having a resolution corresponding to the resolution of the image output means desired to be used, from a low-resolution display to a high-resolution printer. Can be displayed or printed at any resolution and without changing the content of the original image, without any power of 2 In addition, since the expansion of the original image of any size is performed by repeatedly using the original image, the correlation of the data of the expanded image is maintained, and the data compression efficiency is increased. It does not require increase in size of the memory for storing original image it is Rukoto has the effect of so.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a three-stage wavelet transform.

FIG. 2 is a diagram showing a filter sequence for a three-stage wavelet transform.

FIG. 3 is a schematic block diagram illustrating an example of an image processing system in which the present invention is implemented.

FIG. 4 is a flowchart illustrating an example of a process of the image output system.

FIG. 5 is a flowchart illustrating an example of a user setting procedure.

FIG. 6 is a flowchart illustrating another example of the processing of the image output system.

FIG. 7 is a flowchart illustrating another example of a user setting procedure.

FIG. 8 is a flowchart illustrating an example of processing of the image compression system.

FIG. 9 is a flowchart illustrating another example of the processing of the image output system.

FIG. 10 is a flowchart illustrating another example of the processing of the image output system.

FIG. 11 is an explanatory diagram of expansion of an original image.

[Explanation of symbols]

Reference Signs List 10 Original image 11-16 Filter for first-stage wavelet transform 21-26 Filter for second-stage wavelet transform 31-36 Filter for third-stage wavelet transform 40 Image compression system 41 Original image memory 42 image extension unit 43 wavelet transformation unit 44 data compression unit 45 control unit 50 image output system 51 decompression unit 52 wavelet inverse transformation unit 53 memory management unit 54 frame buffer 55 display 56 printer 57 control unit 60 communication channel 501 original image 502 extended image

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FIG06F 15/403 380A

Claims (10)

[Claims] An image processing method comprising: performing inverse transform of image data transformed by using a wavelet transform from a low-dimensional frequency component to a high-dimensional frequency component; and sequentially displaying an image obtained in the process of the inverse transform. An image display method characterized by the following. 2. The image display method according to claim 1, wherein
An image display method, wherein the process of inversely converting image data is terminated when an image having a resolution specified in advance is displayed. 3. The image display method according to claim 1, wherein the size of the original image before conversion is 2 in the horizontal and vertical directions.
If it is not a power of, the original image is expanded to an image having horizontal and vertical dimensions of a power of 2 and then transformed using the wavelet transform. In the process of inverse transformation of the transformed image data, An image display method characterized in that when displaying an obtained image, only a range of an original image is cut out and displayed. 4. An inverse wavelet transform means for performing an inverse wavelet transform from a low-dimensional frequency component to a high-dimensional frequency component on the wavelet-transformed image data of the original image, and a wavelet inverse transform by said wavelet inverse transform means. An image output system comprising: means for writing data of an image obtained in a conversion process to a frame buffer; and image output means for displaying or printing the data written to the frame buffer. 5. The apparatus according to claim 1, further comprising: means for terminating the wavelet inverse transform process when image data of a predetermined resolution is obtained in the wavelet inverse transform process by the wavelet inverse transform means. 5. The image output system according to claim 4, wherein: 6. The image output system according to claim 4, wherein the image data input to the inverse wavelet transform means expands the original image into an image having a power of 2 in the horizontal and vertical directions. If the image data is subjected to wavelet transform and then, the means for writing the image data to the frame buffer is an original image before expansion of the image obtained in the process of inverse wavelet transform by the inverse wavelet transform means. An image output system which cuts out only data corresponding to the range of (1) and writes it into the frame buffer. 7. An image compression system comprising an image compression system and an image output system, wherein the image compression system performs a wavelet transformation on the data of the original image, and performs a data compression of the image data transformed by the wavelet transformation. The image output system received via a communication channel or the like,
Decompression means for decompressing data compressed by the compression means of the image compression system; and performing inverse wavelet transform on the data decompressed by the decompression means from low-dimensional frequency components to high-dimensional frequency components. Wavelet inverse transform means, means for writing image data obtained in the process of wavelet inverse transform by the wavelet inverse transform means to a frame buffer, and image output means for displaying or printing the data written to the frame buffer. An image processing system comprising: 8. The image processing system according to claim 7, wherein the image output system is configured to output the image data of a predetermined resolution in the process of the inverse wavelet transform by the inverse wavelet transform means. An image processing system further comprising means for terminating a wavelet inverse transform process. 9. The image processing system according to claim 7, wherein said image compression system further comprises image expansion means for expanding the original image into an image having a power of 2 in the horizontal and vertical directions. The wavelet transform means performs a wavelet transform on the image data expanded by the image expanding means, and writes the image data of the image output system into the frame buffer. An image processing system wherein only data corresponding to the range of the original image before expansion is cut out of the image obtained in the process of the inverse wavelet transform according to (1) and stored in the frame buffer. 10. An image processing system according to claim 9, wherein said image expansion means creates an expanded image by repeatedly using data of an original image.