JP2640549B2 - Image data recording method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data recording method for recording image data obtained by performing main scanning and sub scanning on an image on a storage medium such as an optical disk. It is.

(Prior Art) In various fields, images are stored, and the images are taken out and observed as necessary. For example, in medical institutions such as hospitals, many medical images are used for medical treatment or research. Most of these medical images are X-ray transmission images, but recently, CT images and MR
Images and the like are also being widely used.

By the way, such medical images need to be interpolated in order to know changes in the patient's illness, and may be required to be kept for a predetermined period by law. That is, the number of pieces is increasing every day.

Conventionally, this medical image has been stored in the form of a hard copy as it is, so securing the storage space, management work, and search work have become a heavy burden on each hospital and the like.

However, in recent years, a so-called image filing apparatus has been used, which retrievably records (files) an image such as a medical image on a recording medium in the form of image data. If a medical image is recorded on a recording medium using this image filing apparatus, space saving and labor saving can be realized while storing the image, and the search operation of the image can be facilitated and speeded up.

The image filing apparatus is configured to record image data on, for example, an optical disk or the like having an enormous storage capacity. However, even an image data for one image requires a considerable recording capacity. Therefore, it is usually stored in the form of compressed image data that is data compressed.

Various methods have been proposed for the data compression processing method, and among these methods, there is a method called interpolation coding. In this method, a large number of pixels constituting an image are taken out, for example, every other pixel to form main data, and for the remaining pixels, an interpolation value interpolated using the main data and an actual value of the pixel are calculated. Interpolation data composed of differences from image data is formed, and these main data,
In this method, the interpolation data is recorded after data compression.

In this image filing apparatus, search data corresponding to each image for specifying the image is input together with image data representing the image itself, and a database is constructed based on the search data. It is performed based on.

Here, in order to obtain image data representing the original image, the main data that has been subjected to the compression processing is decompressed for the main data, and the main data before the compression processing is restored by performing decompression processing corresponding to the inverse operation of the compression processing. For the interpolation data, the compressed interpolation data is subjected to decompression processing corresponding to the inverse operation of the compression processing, and the original data is restored by using the main data, and is combined with the restored main data. Image data representing the original image is restored.

(Problem to be Solved by the Invention) In the image filing apparatus configured as described above, when reading out the searched compressed image data by performing the image search, for example, a defect is found in a part of the recording medium such as the optical disk. As a result, compressed data (main data or interpolation data) may not be read normally, and compressed data partially destroyed may be read.

In this case, if the compressed data is, for example, image data that has been subjected to compression processing by the interpolation coding method, as described above, the interpolation coding separates image data corresponding to one image into main data and interpolation data. Because
Read compressed data (main data or interpolation data)
If a part of the data is destroyed, there is a problem that the correspondence between the main data and the interpolated data is lost, and not only the part where the data is destroyed but also the image data after that part may not be restored. .

The present invention has been made in view of the above circumstances. For example, when reading compressed data recorded separately in the main data and interpolation data and the like, even if a part of the read compressed data is broken, the main data and the interpolation data It is an object of the present invention to provide an image data recording method capable of recognizing a correspondence between data items.

(Means for Solving the Problems) An image data recording method according to the present invention comprises: a plurality of image data obtained by performing main scanning and sub-scanning on an image, the plurality of image data corresponding to a plurality of pixels constituting the image; Is classified into each block of one or more main scans, and for each block, primary compressed data generated from the image data, and secondary compressed data generated from the primary compressed data and the image data. In an image data recording method for generating compressed data and recording the primary compressed data and the secondary compressed data on a recording medium, it is preferable that at least one end of the corresponding primary compressed data and secondary compressed data correspond to each other. A control code indicating the number of pieces of primary compressed data and a number of pieces of secondary compressed data over the entire surface of the image. Is shall.

Here, the “primary compressed data” and “secondary compressed data” typically mean the main data and the interpolated data, respectively, obtained by the above-described interpolation coding process, but are not limited to these, and are not limited to the original data. The compressed data obtained based on only the image data is referred to as primary compressed data, and the compressed data obtained based on both the original image data and the primary compressed data is referred to as secondary compressed data.

In addition, the “at least one end side” refers to at least one of the head and tail of each piece of primary compressed data or each piece of secondary compressed data, and is not limited to either. Further, for example, both of the head and the end of each compressed data for double check or the like may be used.

Also, the above control code is added to the head of each compressed data,
Other control codes indicating the end may be added to the end of each compressed data.In addition to these control codes, the start of the image data corresponding to one image may be added. Alternatively, another control code indicating the end may be added to the end of the image data corresponding to one image.

(Operation) In the image data recording method of the present invention, since a control code indicating that they correspond to each other is added to at least one end of the corresponding primary compressed data and secondary compressed data, a part of certain compressed data is provided. When the data is destroyed, and even when the data is destroyed across a plurality of compressed data, the mutual correspondence between the primary compressed data and the secondary compressed data can be recognized. Even if the image is destroyed, the image can be restored except for the small area where the data in the image is destroyed.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing an example of an image reading / filing / reproducing system employing an example of the image data recording method of the present invention.

The image filing device 50 includes a system control device 51,
An optical disc device 52 that records data on the optical disc 67 and reads data recorded on the optical disc;
Display unit such as keyboard 61A and CRT display
And a console 61 composed of 61B. Further, the image processing device 30 is connected to the image filing device 50. The image processing device 30 is a device that receives image data S1 from the radiation image information recording and reading device 10 as an example of a source of image data, performs predetermined image processing on the data S1, and outputs the data to the image output device 70. It is.

The radiation image information recording and reading apparatus 10 is disclosed in, for example,
No. -29834, No. 61-94035 etc., circulating and transporting the stimulable phosphor sheet 11 along the circulation path 12,
By irradiating the stimulable phosphor sheet 11 stopped at a position facing the imaging table 13 with radiation 15 emitted from a radiation source 14, a transmitted radiation image of a subject (patient) 16 is accumulated and recorded on the sheet 11. . In the image reading section, the stimulable phosphor sheet 11 on which the radiation image is recorded is main-scanned by the laser light 18 deflected by the optical deflector 19 emitted from the laser light source 17, and the sheet 11 is scanned by the main scanning. The sub-scan is performed by moving in a direction substantially perpendicular to the direction, and the sheet 11 is two-dimensionally scanned. The sheet 11 thus irradiated with the laser light 18 as excitation light
Emits stimulating light that carries the radiation image information,
The stimulated emission light is photoelectrically detected by a photodetector 21 such as a photomultiplier via a light guide 20. The analog output of the photodetector 21 is amplified and A / D converted, and is output from the radiation image information recording and reading device 10 as digital image data S1 carrying a radiation image of the subject 16. The stimulable phosphor sheet 11 for which the image reading has been completed is then sent to the erasing section 2, where it is irradiated with erasing light and reproduced again in a state where radiation image information can be recorded.

An ID terminal 25 is connected to the radiation image information recording / reading apparatus 10, and information such as the patient's name, gender, and date of birth written on the ID card 26 of the patient 16 (hereinafter referred to as the The information (hereinafter referred to as imaging information) such as various conditions relating to radiation imaging, that is, image number, imaging date, imaging region, imaging size, reading sensitivity, and the like is input. The patient information S2 and the imaging information S3 are transferred to the image processing device 30 together with the image data S1. Here, this patient information S2, imaging information
An example of a database for image search is constituted by the search data using S3 and other additional information as search data.

The image processing device 30 can perform, for example, 20 or more gradation processes and 10 or more frequency processes on the digital image data S1. All of these image processing conditions are tabulated, and the optimum one is automatically selected according to the photographing conditions set in the ID terminal 25 described above. The image data S1 ′ that has been subjected to the image processing under the optimum conditions in the image processing device 30 is transferred to the image output device.

The image output device 70 is composed of, for example, an image display CRT display device, and a visible image based on the image data S1 'is displayed on the screen. The visible image displayed on the CRT screen is used for diagnosis of the patient 16. The image output device 70 may be a laser printer device or the like in addition to the image display CRT display device.

Next, recording (filing) of a radiation image in the image filing apparatus 50 will be described. The system controller 51 of the image filing device 50 is composed of a known computer system, and the CPU (central processing unit) 53 includes a memory 54, interfaces 55 and 56, a magnetic disk unit 59, and a control unit 57 for controlling the same. And a floppy disk unit 60. Note that the floppy disk unit 60
Is also controlled by the control unit 57. The keyboard 61A and the display unit 61B are connected to the CPU 53, and the interface 55 is connected to the interface 31 of the image processing device 30. The optical disk device 52 includes an interface 62 connected to an interface 56 of the system controller 51, and an optical disk control unit.
63 and an optical disk unit 64.

The above-described patient information S2, photographing information S3, etc.
From 0, the data is transferred to the system controller 51 and is sequentially recorded on the magnetic disk 65 driven by the magnetic disk unit 59 to construct a database. The patient information S2 and the imaging information S3 and the like are also transferred to the optical disk device 52, and the optical disk driven by the optical disk unit 64 together with the image data S5 transferred together from the image processing device 30.
Recorded and stored in 67. The image data S5 is compressed image data obtained by performing a data compression process in the image processing device 30 on the image data S1 in a state before performing the image processing. The image processing apparatus 30 also outputs an image processing condition S4 when the image data S1 is subjected to image processing to obtain image data S1 ′ to be subjected to image processing, and is recorded on the optical disk 67.

Here, a data compression processing method of the image data S1 in the image processing device 30 will be described.

FIG. 2 is a diagram schematically showing each pixel of a part of the read image. For the sake of simplicity, each pixel is represented by reference characters x _ij , a _ij , b _ij , c _ij (i, j = 1, 2,...) And image data corresponding to each pixel Shall also be expressed.

Each line composed of pixel rows arranged in the horizontal direction in the figure corresponds to each main scan by the laser beam 18 by the optical deflector in FIG. Each pixel on the even-numbered line and each pixel on the odd-numbered line are shifted from each other by half a pixel in the horizontal direction in the drawing. The misregistration is realized by adjusting sampling timing at the time of reading, or by performing an interpolation operation after reading. Here, as shown in the figure, two lines are collectively called a group. In this embodiment, each group corresponds to each block according to the present invention.

Here, the main data is composed of the pixels x _ij (i, j = 1, 2,...) _Taken out of every other pixel of the even-numbered line and taken out. The remaining pixels a _ij , b _ij , c _ij (i, j =
1, 2,...) Constitute interpolation data, and the following processing is performed on the interpolation data.

First, using the main data x _ij (i, j = 1, 2,...), The interpolated value (expected value) of each data a _ij , b _ij , c _ij (i, j = 1, 2,...) a
_ij ′, b _ij ′, c _ij ′ (i, j = 1, 2,...) are obtained based on the following equations.

a _ij '= (x _ij + x _{i, j + 1} ) / 2 (1) b _ij ' = (2 · x _ij + x _{i, j + 1} + 2 · x _{i + 1, j} + x
_{i + 1, j + 1} ) / 6 (2) c _ij ′ = (x _ij + 2 · x _{i, j + 1} + _{i + 1, j} + 2 · x
_{i + 1, j + 1} ) / 6 (3) Next, these interpolated values a _ij ′, b _ij ′, c _ij ′ (i, j = 1, 2,...)
…) And actual image data a _ij , b _ij , c _ij (i, j = 1,2, ……)
Image data Δa _ij , Δb
_ij , Δc _ij (i, j = 1, 2,...) are obtained.

Δa _ij = a _ij −a _ij ′ (4) Δb _ij = b _ij −b _ij ′ (5) Δc _ij = c _ij −c _ij ′ (6) These differential image data Δa _ij , Δb _ij , Δc _ij (i, j = 1,
2,...) Are converted into Huffman codes according to the Huffman code table shown in Table 1. The Huffman code table shown in Table 1 is configured with shorter codes as the value approaches zero. This is the interpolation value obtained by the above equations (1), (2) and (3) based on the main data x _ij (i, j = 1, 2,...).
What are the actual image data a _ij , b _ij , c _ij (i, j = 1, 2,...) of a _ij ′, b _ij ′, c _ij ′ (i, j = 1, 2,...) In many cases, approximations are made. Therefore, by converting the difference to a Huffman code, it is possible to further reduce the data amount. Thus, the difference image data Δa _ij , Δb _ij , Δc _ij (i, j = 1,2,...) In each group that have been Huffman-encoded.
Form an interpolation data sequence.

The above main data x _ij (i, j = 1,2,...)
For each group, image data x _{i, j} corresponding to the immediately preceding pixel and image data x _{i, j + 1} of the pixel according to the following equation (7)
And the difference image data Δx
_{i, j + 1} (i, j = 1,2,...) is obtained.

Δx _{i, j + 1} = x _{i, j + 1} −x _ij (7) The difference image data Δx _{i, j + 1} (i, j = 1,2,...) Is also Huffman-coded. This is because there is often a strong correlation between pixels adjacent to each other on the image, and therefore, the difference image data Δx _{i, j + 1} obtained by the equation (7)
Is often close to zero, and thus the data amount is reduced by Huffman coding. In this way, the main data x _ij ((i, j = 1, 2,...)) Is _{divided into the first} main data x _{i, 1} and the second main data x _{i
For i, j + 1} ((i, j = 1,2,...), a main data string composed of Huffman-coded difference image data Δx _{i, j + 1} (i, j = 1,2,...) In this embodiment, the main data sequence and the interpolated data sequence corresponding to each group one by one finally obtained are the primary compressed data and the secondary compressed data according to the present invention, respectively. Is equivalent to

After the main data string and the interpolated data string are obtained as described above, a sign is added to the head or end of each data string.

3A and 3B are diagrams showing the main data and the interpolated data array, respectively, generated in this manner. FIG. 4 is a diagram showing the bit array of codes added to the beginning and end of each data sequence. FIG.

As shown in FIG. 3A, a code SOP (see FIG. 4 (c)) indicating that it is the head of one image is added to the head of group 1 in the main data. The head of each group after the main data group 2 is denoted by a symbol SOL indicating that it is the head of each group. This SOL is
As shown in FIG. 4 (a), in addition to 17 consecutive "0" bits indicating that each group is a delimiter, a group unique to each group for distinguishing each group from other groups. Numbered. At the end of the last group is a code EOP indicating the end of one image.
(See FIG. 4 (d)).

As shown in FIG. 3B, a code SOP indicating the head of the image and a code EOP indicating the end of the image are respectively provided at the head of the group 1 and the end of the last group of the interpolation data. Is attached. The end of each group except the last group is denoted by a symbol EOL (see FIG. 4 (b)) indicating the end of each group. Similarly to the SOL described above, this EOL has 17 consecutive “0” bits indicating that it is a delimiter between groups, and a unique group number is assigned to each group to distinguish each group from other groups. Have been.

Therefore, the code SOP added to the main data and interpolation data,
With SOL, EOL, and EOP, the main data string and the interpolation data string belonging to the same group can be associated with each other.

As described above, after the data is divided into the main data and the interpolated data and the data compression is performed, and each group is assigned a code, the main data (see FIG. 3A) and the interpolated data (see FIG. 3B) The information is recorded on the optical disk 67 in a state of being divided into two.

FIG. 5 is a diagram schematically showing the recording format of the optical disc 67. The recording of the main data and the interpolation data (collectively referred to as compressed image data S5), the patient information S2, the imaging information S3, and the like obtained as described above on the optical disk 67 will be described in detail with reference to FIG. .

In the figure, one scale on the vertical axis indicates one track of the optical disk, and one scale on the horizontal axis indicates one sector. In the compressed image data S5, a group of main data shown in FIG. 3A and a group of interpolation data shown in FIG. 3B are arranged in this order in the image data recording area 80 set sufficiently wide on the optical disc 67, in this order. Be recorded. Before and after the area 81 where one piece of compressed image data S5 is recorded, headers 81A for recording the patient information 2 and the imaging information S3 corresponding to the compressed image data S5,
And blocks 81B and 81C for recording the image processing condition S4 in the image processing device 30.

When one piece of the compressed image data S5 is recorded in the image data recording area 80 of the optical disk 67 as described above, the image directory 83 (83A, 83B) corresponding to the compressed image data S5 is stored in the area 82 for the image directory. , 83C ...) are recorded. Basically, the image directory 83 records the head address of the header 81A of the recorded compressed image data S5, the sector length of the outer compressed image data S5, and characteristic information on the compressed image data S5.

In addition to the areas 80 and 82 described above, the optical disc 67 has replacement directories 89A, 89B and 89C for replacing the recorded contents of the image directory 83 with the changed contents when the recorded contents of the image directory 83 are changed. The area 8 forming ...
For example, when new search data is stored, an area 85 for forming a directory corresponding to the new data is provided. In the first track of the optical disk 67, a block 86 for recording the serial number of each disk and the identification codes of the A and B sides.
A number of directory entry blocks 88A, 88B, 88C,... Are provided along with a block 87 indicating that the optical disk has reached a state where no more data can be recorded (quantity). First directory entry block
88A indicates that a group of image directories 83A, 83B, 83C... Is formed, and the head address and sector length of the image directory group formed in the area 82 are recorded. 2
The third directory entry block 88B records the start address and sector length of the replacement directory group (89A, 89B, 89C ...), and the third and subsequent directory entry blocks 88C ... The space is provided for recording the address and the sector length.

As described above, the compressed image data is stored on the optical disc 67.
S5 and patient information S2, imaging information S3, image processing conditions S4, etc. associated with the compressed image data S5 are sequentially recorded for each image. Since the compressed image data S5 is data after data compression, the number of images that can be recorded on the optical disc 67 can be increased as compared with the case where recording is performed without performing data compression. About 1000 images can be recorded. On the other hand, the magnetic disk 65 has a smaller recording capacity than the optical disk 67, but since only a database including search data such as patient information S2 and imaging information S3 is recorded here, a database for, for example, about one million images is recorded. It is possible to accumulate.

Next, search for an image and re-output of a reproduced image will be described.

The operator inputs a desired search key by operating the keyboard 61A while observing the display unit 61B of the console 61. As a search key that can be input from the keyboard 61A, in principle, all information of the patient information S2 and the imaging information S3 can be used. The system controller 51 searches for one or more images corresponding to the input search key based on the database constructed on the magnetic disk 65.
This is displayed as a list on the display unit 61B. For example, when the patient name of the patient information S2 is input as a search key, the display device 61B displays the image numbers of all the images related to the designated patient, and the image information indicating the patient information S2 other than the name and the imaging information S3. Is displayed. The operator selects a desired image by looking at the displayed image list. When such a selection is made, the system control device 51
By driving 52, the reserved image is read from the optical disk 67. At the time of image reading, the first directory entry block 88A is used as a pointer to give an instruction to read an image directory group (area 82), and image directories 83A, 83B, 83C... Are read. One image directory 83 in which the reserved image number is recorded serves as a pointer, and one header 81A indicated by the image directory 83 is designated, and the header 81A and the corresponding compressed image data S5, block 81B, 8
The recorded contents of 1C are read.

The compressed image data S5 read as described above,
Patient information S2 and imaging information S3 recorded in header 81A,
The data S4 indicating the image processing conditions recorded in the blocks 81B and 81C are transferred from the system control device 51 to the image processing device 30. In the image processing apparatus 30, the head of each group of the main data is first searched out of the transferred compressed image data S5, and the inverse operation of the data compression processing (Equation (7)) for the main data, that is, Huffman coding is performed. After restoring based on the extracted data, data decompression is performed by performing an operation in accordance with the equation x _{i, j + 1} = x _ij + Δxi _{, j + 1} (8), and the main data before compression processing is restored.

Next, by associating the main data string with the interpolated data string by referring to the above-mentioned codes SOP, SOL, EOL, EOP,
Based on the main data, interpolated values a _ij ′, b _ij ′, c _ij ′ (i, j = 1, 2,...) Are obtained in accordance with the above-described equations (1) to (3), and Huffman coding is performed. After the restoration based on the interpolation data, the inverse operation of the above equations (4) to (6), that is, the equation a _ij = Δa _ij + a _ij ′ (9) b _ij = Δb _ij + b _ij ′ (10) c _ij = Δc _ij + c _ij ′ (11) The interpolation data is restored by performing the operation according to (11). By combining the main data and the interpolated data thus restored, the original image data S1
Is played.

In the image processing device 30, after performing image processing such as gradation processing and frequency processing on the image data S1 in accordance with the image processing conditions indicated by the data S4, the image data S1 ′ after the image processing is performed.
To the image output device 70. In the image output device 70, the CR
A visible image based on the image data S1 'after the image processing is displayed on the T screen.

Here, if a part of the read compressed image data S5 has been destroyed due to a defect or the like of the optical disk 67 on which the compressed image data S5 has been recorded, the data has been destroyed until now in the main data sequence or the interpolation data sequence. It is not known whether the data has been destroyed across the columns, so once the data was destroyed, normal restoration could not be performed until the end of the image. However, in this embodiment, FIGS. 3A and 3B As shown in the figure, SOP, SOL, EOL, EO
Because P is recorded on the optical disk 67, if data is destroyed, it is possible to know how many columns the data has been destroyed. Can be restored, and therefore image processing after data decompression can be executed without any support.

In the above embodiment, the case of filing a medical radiation image has been described.However, the present invention is not limited to a medical image when recording a medical image such as a CRT image or an MR image other than a radiation image. Can be widely applied when recording is performed.

(Effects of the Invention) As described in detail above, the image data recording method of the present invention provides a control code indicating that the primary compressed data and the secondary compressed data correspond to each other at least on one end side. Since the recording is performed, it is possible to recognize the primary compressed data and the secondary compressed data that correspond to each other, so that the subsequent image processing and image display can always be normally performed.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing an example of an image reading / filing / reproducing system employing an example of the image recording method of the present invention. FIG. 2 is a schematic diagram showing each pixel of a part of a read image. 3A and 3B show data arrays of main data and interpolation data, respectively. FIG. 4 shows bit arrays of codes attached to the beginning and end of each data column. FIG. 5 is a diagram schematically showing the recording format of the optical disk. 10: Radiation image information recording / reading device 30: Image processing device 50: Image filing device 51: System control device 52: Optical disk device 59: Magnetic disk unit 60: Floppy disk unit 64: Optical disk unit 65 … Magnetic disk 66… Floppy disk 67… Optical disk 70… Image output device 81… Image data 83 recorded on the optical disk 83… Image directory, S1… Image data S2… Patient information, S3… information

Claims (1)

(57) [Claims] 1. A method according to claim 1, wherein a plurality of image data obtained by performing main scanning and sub-scanning on the image and corresponding to a plurality of pixels constituting the image are respectively stored in one or a plurality of main scanning lines. Classified into blocks, and for each block, primary compressed data generated from the image data;
In an image data recording method for generating the primary compressed data and the secondary compressed data generated from the image data, and recording the primary compressed data and the secondary compressed data on a recording medium, the primary compressed data and the mutually corresponding primary compressed data and At least one end of the secondary compressed data is provided with a control code indicating that it corresponds to each other, and is recorded separately in a large number of primary compressed data and a large number of secondary compressed data over the entire image. Image data recording method.