JP2714926B2 - Image processing method and image processing apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to original images and processing
Store image families consisting of images and play back images
The present invention relates to an image reproducing method to be performed. [0002] BACKGROUND OF THE INVENTION Conventionally, in image processing technology, storage devices
Image data (called the original image) stored in the
Image processing by the image processing device
When processing is performed, the processed image (called the processed image) data
Frequently see the case of storing in the storage device again
Be killed. For example, a wide range from low to high density
For images with dynamic range,
Is an image with only a low density area or a high density area
Images are needed for certain research or diagnostic purposes.
If necessary, refer to the processed image later (re-
Raw) if you want to be able to
The processed image is usually stored in a storage device. In such a case, in the prior art, each processed image
To store each image data in a storage device
According to this method, FIG.
One original image data D as shown in FIG.₀
And a plurality of processed image data D₁ , D_Two , ... D_n Mixed with
And one original image data
One image family will be formed. Say here
Image family is one original image data D
₀ And a plurality of processed image data D₁ , ... D_n Formed by
Is defined as a set of image data to be copied. However, based on such a conventional storage method,
The method of reproducing the processed image information has the following disadvantages.
You. For example, storing original image data
N is the storage capacity required for
Assume that the image size does not change with the image processing performed.
You. ), Obtained by performing image processing on the original image
If the number of images is m, this family is stored.
The total storage capacity M required to M = (1 + m) N Becomes Therefore, even when m is 1, the original image data
Required twice the capacity of
Accordingly, M increases linearly. Thus, the traditional memory
The method uses a large storage capacity such as an optical disk device in the storage device
Is an inefficient storage method even if a device with
I can. [0006] SUMMARY OF THE INVENTION The present invention relates to the above-described conventional memory.
It was made in view of the problems of the method,
Description of image family consisting of null image and its processed image
With the aim of minimizing the amount of storage required
In order to achieve the above object, the present invention provides a method for reading an image information recording medium.
Image information obtained by removingInformationApply image processing
An image processing method for obtaining a processed image and outputting and displaying the processed image
Reading the image information recording medium.
Image information obtainedInformationPerform compression processing on the compressed image
Information, and use the compressed image information as original image information.
The image information recording medium
Image information obtained by readingInformationGiven to
Information indicating the type of image processing, and the image information recording medium
Image information obtained by reading the bodyInformationAlms
The information indicating the processing content of the image processing
Storing the processed image in the storage unit in association with the information;
Before displaying in the storage means,
Information indicating the type of image processing and processing of the image processing
Image processing based on information indicating the content is performed on the original image
EmotionInformationAnd output the resulting processed image.
It is characterized by displaying the force. [0007] DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail. In the image reproducing method according to the present invention,
Storage capacity required for each of the
And N, a family with m processed images
The total storage capacity M required by M = (ε + N) + mε = N + (1 + m) ε Since ε takes a very small value compared to N,
Neglect the increase in M due to the increase in the number of physical images
Can be Also, it is required for each image in the family.
Let L be the average storage capacity And one image becomes larger as the number m of processed images increases.
The average storage capacity L required per hit decreases rapidly. According to the conventional storage method, L ≒ N
The storage method according to the image reproducing method of the present invention is a conventional storage method.
The average amount of information required per image is about (m
+1). That is, m processes
In the family where the image exists,
Each of (m + 1) images is compressed to about (m + 1) times
This has the same effect as being stored in a compressed form. Was
For example, when the value of m is 9, each image in the family
Approximately 1/10 compression of image data
Value. FIG. 5 shows the structure of a header in the present storage method.
And the header part is processed as shown in FIG.
The target file pointer part and the image processing information part
It is included in the components. File to be processed
The pointer section stores the processed image file in the storage device.
Section that stores information indicating where it is stored
Minutes, for example, the description of the image file
Address information in the storage device is stored. image
The processing information pointed to by the processing target pointer part is the processing target
Types of image processing to be performed on
Image processing information such as
I have. For example, the image processing to be performed is gradation conversion processing.
, Information indicating that the process is a gradation conversion process.
Report (called the type of processing) and the gradation conversion table used for the processing.
Bull or the information that can refer to it
Information (called processing details) and information that supplements the processing details
It is remembered. FIG. 6 shows various types of image processing information formats.
An example is shown, in which (a) the image processing is gradation conversion.
Image processing information format stored in header of case
This is one example. In this example, the image processing information is (1) Information indicating that the type of processing is gradation conversion (2) Information amount per pixel of target image (K bits) (3) Pixel values 0 to (2) of the target image^k Gradation for -1)
Each pixel value after conversion is 2 in k bits.^k Individual floor
Key conversion table information It consists of three pieces of information. The image processing to be performed is a digital filter.
Of frequency processing by the convolution method using
If the processing is a convolution using a digital filter
That indicates frequency processing by the
Class) and the digital filter matrix to be used for processing
Itself or information that can refer to it
(Processing content) and information supplementing the processing content are in the header
Is stored in FIG. 6B shows a case where the image processing is a digital filter.
Frequency processing by the convolution method using
The format of the image processing information stored in the header.
This is an example of a set. In this case, the image processing information is (1) Convolution using digital filter
Information indicating that frequency processing is performed by the (2) Store the digital filter matrix in the header
Amount of information required for each element (k bits) (3) Number of lines (n) and color of filter matrix
Number (m) (4) Each element of the n × m digital filter matrix
Each value taken by (1,1) to (n, m) is k bits
Digital filter matrix information of nxm Consists of The image processing to be performed uses orthogonal transformation.
In the case of frequency processing, image processing information stored in the header
FIG. 6C shows an example of the format shown in FIG. In this example, the image processing information is (1) that the processing is frequency processing using orthogonal transform
Information (2) The type of orthogonal transformation performed on the image data (for example,
Fourier transform, cosine transform, Hadamard transform, gradient transform
Information) (3) Multiplying the transform coefficient obtained by the orthogonal transform
Amount of information required to store each weighting factor in the header
(K bits) (4) The format of each weighting factor stored in the header (
For example, how many integers and decimals are fixed
Indicates whether it is stored in the (5) Number of lines in weight matrix composed of weight coefficients
(N) and number of columns (m) (6) Each element (1, 1) to n × m weight matrix
The value taken by (n, m) follows the format shown in (4) above.
Weight matrix where n × m items are arranged in k bits
Information Consists of FIG. 6 (d) shows the information shown in FIG. 6 (c).
This is a modification. For example, the weight multiplied by the
The weight matrix of the only coefficient is prepared in advance and stored.
When stored in a device or the like, information indicating the processing content and
Information indicating the weight matrix stored as
Remember. For example, if you have multiple weight matrices,
If there is, enter the number assigned to each matrix in advance.
You may remember. Note that the format shown in FIG.
Is applied to more than frequency processing using orthogonal transform.
be able to. As described above, the image processing information of the header
The storage format of the information stored in the
Any format that can reproduce the same process
You may. Also processed in a format that can reproduce the process
If the image processing can store the contents,
Any image processing is not limited to processing and frequency processing.
No. For example, projective transformation and its special case
It may be a geometric (coordinate) transformation such as a geometric transformation. Was
For example, the two-dimensional affine transformation is the equilibrium movement, rotation,
In the conversion that enlarges and reduces, the coordinate system of the image before conversion is x-
The coordinate system of the converted image is defined as a uv coordinate system,
Then [0018] (Equation 1) It is expressed as (X, y) is expressed using (u, v)
When, [0019] (Equation 2) Becomes According to this equation, the grid point (u, v) corresponding to the transformed
The point (x, y) before the transformation is obtained.
x and y are real values, so (x, y) is a non-grid point
Become. Therefore, the density value of the non-grid point (x, y) is calculated by interpolation.
Need to ask. The simplest interpolation method,
Case in the (x-y) coordinate system closest to the non-lattice point (x, y)
Nearest neighbor method using child point density value, more accurate method
At four grid points around the non-grid point (x, y)
Linear interpolation method using linear density
Cubic convolution known for degree interpolation
There is a law. Therefore, for a given image, a two-dimensional affine
In the case of image processing that performs transformation, determine a two-dimensional affine transformation
Conversion coefficients a, b, c, d, t_x , T_y Or secondary
Transform coefficients that determine the original affine inverse transform and the density of non-grid points
The type of interpolation method or interpolation formula for determining the degree
By storing it in the header,
The method can be adapted. Other image processing such as orthogonal transform
Species (Fourier transform, cosine transform, Hadamard transform,
Combo using digital filter)
Filtering that is not based on the solution method (for example,
Such as processing with a median filter)
Single pixel rearrangement (eg 90 ° image rotation, corner
-Dither method, etc.)
Of the present invention by storing the contents of image processing information
Can be applied. Under the above-described image file configuration,
An embodiment of the present invention will be described with reference to FIG. FIG. 1 shows an embodiment of an image reproducing apparatus according to the present invention.
An example is shown in a block diagram, where 1 is, for example, paper,
Image information storage media such as films and stimulable phosphor plates
Read the image information recorded in the
An image input device to output, for example, laser light
Mechanical scanning or electrical scanning of elements such as CCD
To convert the image information in the recording medium into an electric signal,
A device that outputs a digital signal with an A / D converter
is there. However, the output signal from the image input device 1 is digital.
Signal for image information recording media and image input devices.
There is no limitation on the number of mechanisms and devices. [0023] 2 is that the image can be visually caught.
Image output device that outputs or displays
You. For example, hard copy images on paper or film
Or a CRT that displays images
May be the display device. Also, the number of these devices is
A plurality of different types of devices may be arranged in parallel.
You may. 3 is for storing files such as image data.
Storage devices such as magnetic disk devices and optical discs.
Device. 4 indicates that the user operates the image processor 5.
Image processing, image input from image input device 1, image output
Operation table for inputting processing commands such as image output to the apparatus 2
Terminal device having a keyboard, for example. Image processor 5 (indicated by a chain line)
) Analyzes the command input from the operation table 4
And a device for executing the commanded processing, and
Input and output of files such as image files stored in
In the file input / output control unit 51 to be controlled and in the storage device 3
A file for managing files such as stored image files
File management unit 52, image input device 1, and image output device
Data input / output control that controls the input / output of image data to / from
Unit 53 and a data for temporarily storing information such as image data.
Data storage unit 54 and various images such as gradation processing and frequency processing.
Image processing unit 55 having an image processing function, header creation, etc.
The header processing unit 56 having the function of
Processing system that gives processing instructions to each part according to the received instructions
Analyzes commands input from the control unit 58 and the operation table 4
A command analysis unit 57 for giving an instruction to the processing control unit 58;
Format of header information such as image processing information format
Header information format table 5 storing the header
9. Next, the operation of the image reproducing apparatus having the above configuration will be described.
This will be described with reference to the flowchart of FIG. The operation of the present apparatus is classified into the following three. (1) Input → storage operation (2) Processing → storage operation (3) Read → output operation First, the (1) input → storage operation will be described. Input from operation table 4 → input storage command
(F-1) The command analyzer 57 analyzes the contents of the command.
(F-2), and outputs an instruction to the processing control unit 58. That
As a result, the image data is stored in the image storage medium by the image input device 1.
The image information is read (F-3). The processing control unit 58
A header creation command is output to the header processing unit 56, and header processing is performed.
The unit 56 creates a new header part in the data storage unit 54
I do. The file contains the original file name in the header.
In other words, an option indicating that an image data section is provided next to the header section.
The original file identification information is stored (for example,
A special value, such as zero, is written in the
Flag that indicates that it is an original file
To identify the original file
(F-4). In addition, other necessary information,
For example, storing the size of image data and one pixel of image data
Storage capacity information (for example, the number of bits) required for
It is preferable to store in a format common to image processing. Next, the processing control unit 58 controls the data input / output
An image data input command is output to the control unit 53, and the data
The data input / output control unit 53 is sent from the image input device 1.
The image data is stored in the data storage unit 54. Enter file
The output unit 51 stores data according to a command from the processing control unit 58.
One header and image data stored in the unit 54
Storage device as an image file (original file)
3 (F-5), and in the file management unit 52,
Register the stored image file. In the present invention, the image input device 1
After inputting the image data from
Data compression processing before the data is stored in the storage device 3
It is carried out. This data compression is stored in the storage device 3.
May be performed at any time before
However, the present invention is not limited to this. For the compression ratio,
It is particularly limited as long as image quality degradation is not a problem.
Although it is not intended, for example, image data is
Data compressed and stored as original file
In this case, if the value of m is set to 9, the family is configured.
About 1/100 (1/10 × 1
/ 10 = 1/100)
Therefore, there is a huge effect in terms of reducing storage capacity.
Be able to add and actually make a family
Approximately 1/100 data compression for each image data
Obtain playback image information with better image quality than when
Becomes possible. In addition, for example, the image data is
Store the original compressed file as
If the value of m is 1, the family is configured.
For each image data to be formed, about one quarter (1/2 × 1 /
(2 = 1/4)
Can have a tremendous effect on reducing storage capacity.
In addition to the image data that actually constitutes the family,
Compared to a data compression of about 1/4
As a result, it is possible to obtain reproduced image information with good image quality.
You. That is, if the image storage method according to the present invention is used,
Whether the data compression ratio for the original image is
Regardless of the number of images in the family,
It is possible to reduce the storage capacity and obtain high quality reproduced images.
I will be able to. Thus, the input → storage operation is completed. Next, the (2) processing → storage operation will be described.
You. Process → store command input from operation table 4
(F-1) (here, stored in the storage device 3
One for the image data of the original file
Assume that a command to perform image processing has been input) and
The analysis unit 57 analyzes the command contents (F-2), and controls the processing.
An instruction is given to the unit 58. The header processing unit 56 controls processing.
Receiving a command from the storage unit 58,
Create a header section and search the header management section 52,
In the storage device 3 of the original file to be processed,
(For example, address information) indicating the storage location
All are set to the processing target file pointer part in the header.
(F-6, F-7). The header processing unit 56 controls the image to be processed.
The processing contents are stored in the header information format table 59
Refer to the applicable image processing information format, and
Set as image processing information in the image processing information section of
(F-8). All information is set in the header
And the file input / output control unit 51
Command, the header, that is, the processed file is stored in the storage device 3.
(F-9) The file management unit 52 stores the new
Register the stored file that has already been stored.
A processed file is a header that stores image processing information
File). Processing → storage operation in this way
Ends. Next, (3) read-out → output operation will be described.
I will tell. Read from operation table 4 → Output command is input
(F-1) Here, the user specifies as an output target
The original image file or original file
One processed image file that has been processed
Assume that The command analyzer 57 resolves the command content.
(F-2), and gives an instruction to the processing control unit 58. processing
File input / output control unit 51 receiving a command from control unit 58
Is the output file specified by the user.
After searching the storage location by the file management unit 52, the storage device
3 and reads only the header of the file (F-1).
0), and stored in the data storage unit 54. Header processing unit 56
Is the original of the header written in the data storage unit 54
Check the file identification information and confirm that this file is the original
File is determined (F-11). If this file is an original file
Processing control unit 58 corresponds to the file input / output control unit 51
Command to read the image data part of the file (F-1)
3) The image data is stored in the data storage unit 54. De
The data input / output control unit 53 receives a command from the processing control unit 58.
The image data stored in the data storage unit 54
Output to the output device 2 (F-15), and storage → output operation
finish. The file is not the original file
When the file is a processed file
(The pointer to be processed in the header is stored in the storage device 3.
The storage location of the original file is stored)
The processing control unit 58 sends the file input / output control unit 51 an original
Command to read the image data of the file
(F-12, F-13) Header information format table
Stored in the header of the processed file with reference to the file 59
The type of the processing being performed is identified, and the corresponding processing in the image processing unit 55 is identified.
Call a processing function. The called image processing function is
Corresponding image processing in the data information format table 59
Image processing parameters while referring to the information format
Is read from the image processing information section in the header of the processed file.
After establishing the processing conditions, the data storage unit 54
Image processing is performed on the stored image data (F
-14). The data input / output control unit 53 includes a processing control unit 58
Of the image stored in the data storage unit 54
The image data for which the image processing has been completed is output to the image output device 2.
(F-15). Thus, the storage-> output operation ends. The operation shown in the above embodiment is performed by the storage device 3
In the family stored in
File or original file image data
A processing file that has undergone one image processing (ie, FIG. 5
The value of the image processing information number n in FIG.
File pointer is the storage location of the original file
Processing file with a header that stores
Although only the case was described, the image of the present invention
In the playback method, the number of image processing information stored in the header is
It is not limited to one. For image data to be processed, n (n
> 1) When the image processing is performed In the image processing information section
Indicates that n pieces of image processing information are stored as shown in FIG.
It is. In this way, multiple pieces of image processing information are stored in the header
For example, k (1 ≦ k ≦ (n−1), n ≧
2) The (k + 1) th image processing information in the
That indicates where in the header the
If an interface is provided, the storage capacity required for each image processing information
Knows the storage location of each image processing information even if is not constant
be able to. Similarly, the k ′ (2 ≦ k ′ ≦ n) -th image
The (k'-1) th image processing information is included in the processing information.
A pointer is provided to indicate where in the
You may keep it. As described above, the header includes a plurality of images.
Processing information can be included. In other words, the implementation
In the example (2) Processing → storage operation, the user is the processing target
Multiple image processing can be commanded for an image
You. Further, the image reproducing method of the present invention is an object of image processing.
In limiting image files to original files
Absent. That is, the processed file is added to the image file to be processed.
File (header file) may be specified. Hereinafter, description will be made with reference to FIG. The original file (file) is stored in the storage device 3.
Image processing A for the original file
Processed file (file F1) already exists
File F1
Create a new file (File F2) with image processing B
The header structure at the time of the
either will do. (Method 1) As shown in FIG.
The processing information of processing B is stored in the image processing information section in the header of file F2.
Information and set the file F1 in the processing file pointer section.
Sets information that points to a storage location in the storage device
how to. (Method 2) As shown in FIG.
Processing A and processing are added to the image processing information section in the header of file F2.
Set the processing information of the processing B, and in the processing file pointer section
Indicates the storage location of the file F0 in the storage device
How to set information. At this time, two of processing A and processing B
Image processing information in the header by using a pointer
Relative position or absolute position of two pieces of information in
Make clear the relationship. The method is achieved by the above-described arrangement.
Therefore, it is not necessary to store the processed image information.
Of the number of images (m + 1)
When storing image family information, regardless
Can be prevented from increasing in storage capacity. Also, image information
Information obtained by compressing the image as original image information
The original image with the effect of reducing the storage capacity.
The effect of the compression performed on the image is multiplied
The reduction in storage capacity should be greater
Can be. The final configuration provided by the above configuration
Image family like traditional ones
Only compression processing is performed on each image information constituting
Try to achieve at least
Twice the compression rate applied to the original image information
Each image with a compression ratio of at least (more precisely (m + 1) times or more)
This method was implemented because the information had to be compressed.
Each image information is significantly inferior due to the compression process.
It will become. That is, according to the method,
Playback image quality is significantly higher than when using conventional compression technology
Can be improved. Further, the image reproducing method of the present invention
Is no longer performed on all image information processed
However, the effect of reducing the storage capacity can be obtained.
It is stiff. [0048] As described above, according to the present invention,
The storage capacity for storing image family information
Can be reduced, and therefore one image to be stored
Storage cost is greatly reduced, and
It is possible to prevent the image quality from deteriorating. Furthermore, according to the present invention, the information in the device
Read, write, and transfer speeds
Be transformed into

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an embodiment of an image reproducing apparatus for realizing an image reproducing method according to the present invention. FIG. 2 is a flowchart illustrating the operation of the apparatus shown in FIG. FIGS. 3A and 3B are diagrams showing a connection relationship between files in a family of the image reproducing method according to the present invention. FIG. 4A is an explanatory diagram of a conventional image data storing method, and FIG. 4B is an explanatory diagram of an image data storing method according to the present invention. FIG. 5 is a configuration diagram of a header in a storage method according to the present invention. FIG. 6 shows various examples of an image processing information format. [Description of Signs] 1 image input device 2 image output device 3 storage device 4 operation table 5 image processing processor

Claims (1)

(57) [Claims] To obtain a processed image by performing image processing on the image information obtained by reading the image information recording medium, an image processing method for outputting displaying the processed image, obtained by reading the image information recording medium subjected to compression processing for the image information, to obtain compressed image data, the compressed image information as well as stored in the storage means as the original image information, the image information recording image information obtained by reading the medium correspondence information indicating the image information indicating the type of processing, and the processing contents of the image information recording medium for the image information obtained by reading the applied image processing performed for the said original image information Information indicating the type of the image processing stored in the storage means and the image processing when the processed image is output and displayed. Image processing method comprising an image processing based on the information indicating the processing contents that as the applied against the original image information, and outputs display the processed image of the obtained image processed.