JPH07131656A - Image encoding and decoding method - Google Patents

Abstract

PURPOSE:To shorten image compression processing time and to simplify the selection of a compressed image having an image quality required for read, diagnosis and preservation by encoding the plural kinds of image data just for the capacity of respective image buffers while repeatedly setting the plural kinds of encode parameters. CONSTITUTION:Image data for one part or the entire part of original image data are copied from a source image data file 10 to an image area 20, the copied original image data are copied just for the capacity of a transfer buffer 21, and an encoding circuit 31 performs over two kinds of image encoding (namely image compression). In this case, since an image encoder switches the kinds of image encoding for the unit of total capacity of image buffers 30a and 30b, the image data are just once transferred through the original image data file 10, image area 20, transfer buffer 21 and image buffers 30a and 30b, and two or more kinds of image encoding can be performed for the unit of the total capacity references of the respective image buffers 30a and 30b.

Description

Detailed Description of the Invention

[0001]

[Industrial field of use] X-ray imaging and X-ray computed tomograp
hy: X-ray tomography), MRI (magnetic resonance im)
aging: Nuclear magnetic resonance imaging)
There are various types of image inspection, and a large amount of image information is generated.

At present, these image information are printed on a film for interpretation / diagnosis and stored, but recently, these image information are digitized, and the interpretation / diagnosis / storage is performed by using a computer. Has become.

However, since the amount of image information data is enormous, due to the limit of the capacity of the current computer system,
To reduce the load on the network when transferring images between computers, speed up the time to read image data from storage devices (magnetic disk, magneto-optical disk),
In addition, in order to reduce the cost of these devices, it is practically indispensable to handle the image data in a compressed state and restore it at the display or printing stage.

Image data compression methods include "reversible encoding" in which image quality is not deteriorated at all and "irreversible encoding" in which image quality is deteriorated. "Lossless encoding" reduces the original image data to 1 /
In contrast to the compression of only 2 to 1/3, the “lossy encoding” can increase the compression rate with an image quality that does not pose a problem for doctors' interpretation and diagnosis.

Therefore, it is possible to shorten the time required for a plurality of types of image coding (compression) processing, to easily select a coded image having an image quality necessary for image interpretation, diagnosis, and storage, and further to hold the amount of image data to be stored. There is a need for an image coding method and a decoding method that can reduce the number of times and shorten the image coding / restoration processing time.

[0006]

2. Description of the Related Art FIG. 8 is a diagram for explaining a conventional image coding apparatus, FIG. 8 (a) shows a configuration example of the image coding apparatus, and FIG. 8 (b) shows a coding process. The unit is shown. First, in FIG. 8A, conventionally, when compressing a plurality of images of different types (hereinafter, referred to as encoding), it is necessary to perform the following procedure for the number of times.

(1) Encoding parameters are set in the encoding circuit 31 to instruct image encoding. (2) "Original image data file 10" of file device 1 →
In the "Image area 20" transfer, the original image data file 10
Copy part or all of the above to "Image area 20".

(3) "Image area 20" → "Transfer buffer 2"
In "1" transfer, the data of "Image area 20" is copied for the capacity of "Transfer buffer 21". (4) When transferring from "Transfer buffer 21" → "Image buffer 30a", the amount of "Transfer buffer 21a"
21 ”data is copied.

(5) In the image buffer switching process, the "image buffer 30a" and the "image buffer 30b" are switched:
The "image buffer 30a" is connected to the encoding circuit 31 side, and the "image buffer 30b" is connected to the computer device 2 side (specifically, the transfer buffer 21).

The encoding circuit 31 includes image buffers 30a and 30b.
The data in the "image buffer 30a" is encoded and the encoded data is written in the "transfer buffer 22" by using the switching of the above as a trigger.

(6) "Transfer buffer 22" → "compression code area 23" In the transfer operation, the encoded data of "transfer buffer 22" is copied to "compression code area 23". (7) If necessary, repeat the above until all the images are encoded.
The operations from (1) to (6) are repeated.

(8) In the "compression code area 23" → "compression coded data file 11" transfer, the above compression coded data is copied to the compression coded data file 11 of the file device 1.

The storage areas for each data have the following relationship. "Original image data file 10" ≥ "Image area 20" ≥
"Transfer buffer 21" ≥ "Image buffer 30a, or 30
b ”“ compressed coded data file 11 ”≧“ compressed code area 2 ”
3 ”≧“ transfer buffer 22 ”As described above, conventionally, when two or more types of image encoding are performed, the above operation is repeated as many times as the number of types of the encoding. , "Original image data file 10" → "Image area 20" → "Transfer buffer 21"
-> It was necessary to perform the image transfer of "image buffer 30a or 30b" as many times as the number of the types.

At this time, as shown in FIG. 8 (b), the coding process of the same coding type, specifically, the coding process of each region corresponding to the capacity of the image buffers 30a, 30b, , Image data was transferred and encoded.

Further, in a device (hereinafter, abbreviated as a display device) 4 for displaying an image in which the conventional encoded data is restored as shown in FIG. 8A, a plurality of encoded images generated from the same original image are used. Even when displaying images, they were treated as separate unrelated images.

In particular, in a medical image, it is common that the image to be displayed can be used as an image interpretation target after being subjected to image processing such as gradation processing, enlargement and filtering processing. When comparing the image qualities of a plurality of compressed images, it is necessary to perform the same image processing on each image, but a series of these operations is performed in real time for each image having a different compression rate.

Further, in the medical field, although there is no influence on image interpretation / diagnosis as long as only the image quality of a diseased part is good, the conventional method will be described with reference to FIGS. 8 (a) and 8 (b). As described above, in order to make the image quality of a diseased part equal to or higher than the standard for image interpretation, the image quality of a part that is not necessary for image interpretation must be matched to it, and the unnecessary part is also necessary for the image interpretation / diagnosis. It was encoded with the same encoding type as the part.

[0018]

Therefore, 1) When only one image encoding mechanism (encoding circuit) 31 is provided, the same original image is, for example, "reversible encoding" and "lossy encoding". In order to perform the two types of image coding, the original image data must be transferred twice to the image coding mechanism (coding circuit). The same problem also occurs when a plurality of "lossless encoding" with different image quality conditions, for example, different compression ratios are performed. Therefore, there is a problem that it takes a lot of time to transfer the image and the compressed code data.

2) When comparing the original image (or "lossless coded image") and the "lossy coded" image, it is necessary to compare parts unnecessary for image interpretation / diagnosis with images of the same compression method. There is no problem, because there is no effective comparison means only for the necessary parts and the image interpretation doctor cannot accurately determine the image quality, or it is not possible to reduce the data amount by lowering the image quality only for the parts unnecessary for the image interpretation and diagnosis. there were.

Further, the coding / decoding parameters at the time of coding / restoring an image are
Coding circuit 3 from the central processing unit (CPU) 24 of computer device 2
Since it was set to 1, there was a problem that it took time to encode and restore the image.

In view of the above-mentioned conventional drawbacks, the present invention shortens a plurality of types of image compression processing time, and easily selects a compressed image having an image quality necessary for image interpretation, diagnosis, and storage, and further, an image to be stored. An object of the present invention is to provide an image coding method capable of reducing the data amount and shortening the parameter setting time of the image compression / decompression processing.

[0022]

1 to 7 are views showing an embodiment of the present invention, FIG. 1 shows a configuration example of an image coding apparatus of the present invention, and FIG. 3 to 5 show configuration examples of a display device of the present invention, FIGS. 3 to 5 show configuration examples of encoded data according to the present invention, and FIGS. 6 and 7 show encoding parameters for an encoding circuit. The method of setting is schematically shown. The above problems can be solved by the image coding method configured as follows.

(1) Part or all of the original image data is read from the original image data file 10 of the file device 1, copied to the image area 20 of the main memory device 2a of the computer device 2, and the main memory device 2 The image data for the capacity of the transfer buffer 21 is copied from the image area 20 to the transfer buffer 21, the predetermined encoding parameters are set in the encoding circuit 31 of the image encoding mechanism 3, and the encoding circuit 31 Image data of a predetermined capacity is copied from the transfer buffer 21 to the image buffer 30a or 30b, and the image buffer 30a or 30b is switched to the image buffer 30a or 30b.
30a, or 30b is connected to the encoding circuit 31, the image buffer 30b, or 30a is connected to the computer device 2, the switching of the image buffer 30a, or 30b is used as a trigger.
A method of encoding the image data of the image buffer 30a or 30b and transferring the encoded data to the computer device 2, wherein the image data from the file device 1 to the image buffer 30a or 30b is transmitted once. For the transfer of, the setting of a plurality of types of encoding parameters from the computer apparatus 2 is repeated, and the capacity of the image buffers 30a, 30b
It is configured to encode the plurality of types of image data.

(2) The image coding method described in the above item (1), wherein the original image data file is used for image coding.
The original image ID is added to each of the 10 original image data,
A plurality of types of encoded data encoded by the encoding circuit 31 are stored as a database in the encoded data file 11, and the encoding type and the original image ID are designated,
When a plurality of types of images of the original image ID are restored and predetermined image processing is instructed on the restored image of the instructed original image ID of the instructed encoding type, the plurality of encoding types After performing a predetermined image processing on all the restored images corresponding to, display the image of the encoding type specified above, depending on the image quality deterioration for the displayed image,
The coded data of unnecessary images is configured to be selectively deleted.

(3) The image coding described in the item (2) is configured to be performed for each predetermined area of the original image data. (4) The image coding method according to the above item (1) or (2), wherein a unique number is set to each of the coding circuits 31 to set the coding parameters. The n types of patterned quantization tables, the m types of patterned Huffman tables, the number of the quantization table used in the immediately preceding encoding process, and the previous quantization table ID storing the Huffman table number When,
The Huffman table ID and the Huffman table ID included in the encoding condition notified from the central processing unit (CPU) 24 of the computer apparatus 2 for encoding the original image data from the previous time. , The previous quantization table ID and the previous Huffman table ID
Are compared by the central processing unit (CPU) 24 of the computer device 2, and if they match, the matched table value is controlled not to be transferred to the encoding circuit 31.

(5) In the image coding method described in the above item (1) or (2), a decoding parameter is set in the above-mentioned coding circuit (31) to code the coded data. To restore, n kinds of patterned quantization tables each with a unique number, m kinds of patterned Huffman tables, the number of the quantization table used in the previous restoration process, and Huffman The previous quantization table ID () for storing the table number and the previous Huffman table ID () are provided, and are stored in the header portion of the encoded data when the encoded circuit 31 restores the encoded data. The quantization table ID, the Huffman table ID, the previous quantization table ID, and the previous Huffman table ID are compared by the central processing unit (CPU) 24 of the computer apparatus 2, and if they match, one The table value thus obtained is controlled so as not to be transferred to the encoding circuit 31.

[0027]

That is, in the image coding method according to the present invention, a part or all of the original image data is copied from the original image data file 10 to the image area 20, and the copied original image data is copied. In the image encoding device shown in FIG. 1, when the image data is copied by the capacity of the transfer buffer 21 and two or more types of image encoding (that is, image compression) is performed by the encoding circuit 31, the image buffers 30a and 30b of the image encoding device shown in FIG. By switching the image coding type in units of total capacity, the original image data file 10 → image area 20 → transfer buffer 21-
The image data is transferred to the image buffers 30a and 30b once, and the two or more types of image encoding are performed in a unit of the total capacity of the image buffers 30a and 30b.

Further, each encoded data in the compressed encoded data file 11 is given an original image ID indicating that it is the same original image data, and one original image ID and one encoding type , The corresponding encoded data is restored, and when the restored image is subjected to predetermined image processing, the same image processing is performed on all restored images of other coding types. Then, the restored original image of the encoding type designated at the beginning is displayed as an image, and the same original image I is displayed.
By evaluating the image quality of the image of D, the image quality comparison is facilitated, and an unnecessary image is designated, and the corresponding encoded data is deleted.

As described above, in the image coding according to the present invention, the coding is performed in the unit of the total capacity of the image buffers 30a and 30b shown in FIG. That is, it is divided into a number of lines depending on the total capacity of the image buffers 30a and 30b and encoded. In the restored image of the divided encoded (compressed) data,
By combining and displaying images that have undergone prescribed image processing, such as in a montage photograph, the coding rate of areas that are not related to image interpretation can be increased, and the image quality required for such image interpretation and high encoding (compression) can be achieved. ) It is to make it possible to obtain the rate.

Further, at the time of encoding or restoring the image, one or both of the "quantization table" and the "Huffman table" identifier (ID) necessary for encoding and restoring the image are used immediately before. When the "quantization table ID" and the "Huffman table ID" match, the values of the "quantization table" and the "Huffman table" are not set in the encoding circuit.

Therefore, it is possible to shorten the time required for a plurality of types of image compression processing, to easily select the coded data of the image quality required for image interpretation, diagnosis, and storage, and to eliminate the need for coded data. By deleting the, the amount of image data held by the image encoding device can be reduced.
Further, by omitting the transfer of the quantization table and the Huffman table, there is an effect that the time of the image compression / decompression processing is speeded up.

[0032]

Embodiments of the present invention will be described in detail below with reference to the drawings. 1 to 7 described above are views showing an embodiment of the present invention.

In the present invention, in order to perform two or more types of image compression (encoding), the original image data of a predetermined size is transferred only once, and only the encoding parameters are set thereafter.
A means for performing a plurality of types of image compression (encoding) for the size, and an original image ID for identifying an original image when the image is encoded
Is added to store a plurality of types of encoded data in the database, the encoding type and the original image ID are designated, the plurality of types of images of the original image ID are restored, and the designated image processing is performed. Means for displaying the displayed image and deleting the encoded data of the unnecessary image from the database according to the deterioration of the image quality of the displayed image, or during the encoding / restoring process,
When the setting IDs of the quantization table and the Huffman table are the same as the IDs used immediately before, the encoding circuit implements the present invention by means of omitting the setting of the values of the quantization table and the Huffman table. It is a necessary means for
The same reference numerals indicate the same objects throughout the drawings.

The image coding method of the present invention will be described below with reference to FIGS. First, referring to FIG. 1, a plurality of image encoding processes of different types are omitted from redundant image transfer,
A method of performing high-speed image coding will be described. {Embodiment corresponding to claim 1} In the present invention, when two or more types of image encoding (image compression) are performed, "image buffer
By switching the encoding (compression) type in units of total capacity of "30a" and "Image buffer 30b", "Original image data file 10" → "Image area 20" → "Transfer buffer 21" → "Image buffer 21" 30a, or 30b "image transfer is performed only once, enabling high-speed image encoding (compression). The details will be described below.

(1) "Original image data file 10" → "Image area 20" transfer: Part of the original image data file 10,
Or copy the whole image into "Image area 20". (2) "Image area 20" → "Transfer buffer 21" transfer:
Copy the image data of "Image area 20" for the capacity of "Transfer buffer 21". Image encoding of the first type: (3) The parameters of the first type are set in the encoding circuit and the "encoding circuit 31" is activated.

(4) Transfer from "transfer buffer 21" to "image buffer 30a": Copy a predetermined amount of data from "transfer buffer 21" in "image buffer 30a". (5) Switching the image buffer: Switching between "Image buffer 30a" and "Image buffer 30b": "Image buffer 30a"
Is connected to the encoding circuit 31 side, and the "image buffer 30b" is connected to the computer device 2 side, specifically, the "transfer buffer 21".

The encoding circuit 31 encodes the data in the "image buffer 30a" by using the switching of the image buffer as a trigger and writes the encoded data into the "transfer buffer 22".

(6) Transfer from "transfer buffer 21" to "image buffer 30b": Copy a predetermined amount of image data from "transfer buffer 21" to "image buffer 30b".

(7) Switching of image buffer: "Image buffer 30b" and "image buffer 30a" are switched, "image buffer 30b" is connected to the encoding circuit 31 side, and "image buffer 30a" is connected to the computer device 2 side. Connect to.

The encoding circuit 31 encodes the data in the "image buffer 30b" by using the switching of the image buffer as a trigger and writes the encoded data in the "transfer buffer 22".

(8) “Transfer buffer 22” → “compressed code area 23” transfer: Coded data of “transfer buffer 22” is copied to “compressed code area 23”. (9) "Compressed code area 23" → "Compressed coded data file 11" Transfer: "Image buffer 30a" and "Image buffer
Copy the compressed coded data of the image of 30b "into the compressed coded data file 11 of the file device 1.

If the "compression code area 23" is provided for each coding type and the capacity is sufficiently large,
Finally, it is also possible to collectively transfer the "compression code area 23" → "compression coded data file 11". Second type encoding: (10) The second type encoding parameter is set in the “encoding circuit 31”, and image encoding (image compression) is instructed.

(11) Switching of image buffers: In the present invention, since the images transferred for the first type of encoding process still remain in the "image buffer 30a" and the "image buffer 30b". The new "Transfer Buffer 21"
Switching from "image buffer 30a" to "image buffer 30b" without transferring from the
Connect to the 31 side and connect the "image buffer 30b" to the computer device 2
Connect to the side. The encoding circuit 31 uses the switching of the image buffer as a trigger to encode the data in the "image buffer 30a" and write the encoded data in the "transfer buffer 22".

(12) Switching of image buffer: Switching between "image buffer 30b" and "image buffer 30a", connecting "image buffer 30b" to the encoding circuit 31 side, and connecting "image buffer 30a" to computer device 2 side Connect to.

The encoding circuit 31 encodes the data in the "image buffer 30b" by using the switching of the image buffer as a trigger: The encoded data is written in the "transfer buffer 22".

(13) “Transfer buffer 22” → “compressed code area 23” transfer: The encoded data of the “transfer buffer 22” is copied to the “compressed code area 23”. (14) "Compressed code area 23" → "Compressed coded data file 11" Transfer: The compressed coded data of the images of "Image buffer 30a" and "Image buffer 30b" are converted to "Compressed coded data file 11". Make a copy.

As described above, the "compression code area 2
3 ”is provided for each type of encoding (compression), and if the capacity is sufficiently large, at the end, transfer the above“ compression code area 23 ”→“ compression coded data file 11 ”. It can be carried out.

Third type to nth type compression: (15) The nth type parameter is set in the encoding circuit 31 to instruct image encoding.

(16) The procedure of (11) to (14) is repeated, and in the above embodiment, by providing the "image buffers 30a and 30b" corresponding to the "encoding circuit 31", the "transfer Image buffer 30b or 30a from buffer 21
To the image buffer 30a, or 30b
It is obvious that the image data encoding circuit 31 can perform the encoding process in parallel.

Next, an embodiment corresponding to claim 2 will be described with reference to FIG.
Will be described below. The image coding method described here performs a plurality of image coding (compression) with different degrees of image quality deterioration, and at the time of display, the image quality is
The image is deleted by the operator except for the compression coded data of the image evaluated by the operator.

First, image information (or in encoded data,
Or, when the information indicating the same original image and original image ID are given to a higher-level database), one of the encoded data is selected, and the restored image is subjected to image processing. , Restore all other encoded data,
By performing the same image processing on the restored images, it is possible to compare the image quality of the same original image IDs, which facilitates the image quality comparison. It also provides a method for automatically deleting unnecessary data after comparison.

FIG. 2 schematically shows a display device connected to the image coding device according to the present invention. (1) At the time of image encoding, information indicating the same original image (for example, a number issued in the order in which the original images are stored, hereinafter referred to as an original image ID) is stored in the encoded data or in a higher-level database. Etc.

(2) For example, when the operator instructs the image display together with the original image ID by the encoding type, the operator has the instructed original image ID and stores the encoded data of the instructed encoding type. The images are restored, and restored images 1 to n corresponding to other coding types are generated, and only the image designated by the operator is displayed.

(3) When the operator instructs a predetermined image processing on the displayed image, the display device shown in FIG. 2 executes the instructed image processing with the same original image ID.
Apply to images with. That is, the same image processing is performed on all the restored images for the other encoded data that have been restored.

(4) When the operator evaluates the image quality of the displayed image and gives an instruction to delete unnecessary images, the display device deletes the compression code (encoded data) i of the specified image. . Hereafter, until the image of the image quality suitable for diagnosis is obtained,
The operation of (4) above is repeated.

Next, an embodiment corresponding to claim 3 will be described with reference to FIGS. As is apparent from the description of FIG. 1, the image coding process is performed in units of images corresponding to the number of lines depending on the total capacity of the image buffers 30a and 30b.

Therefore, the image coding described with reference to FIG. 2 is performed for each image region of the number of lines depending on the total capacity of the image buffers 30a and 30b. That is, according to the image compression method described in FIG.
The image divided into the number of lines depending on the total capacity of a and 30b is encoded, and the encoded data that is encoded with different encoding (compression) rates is restored for each area, An image generated by performing a predetermined image processing is combined in units of the area like a montage photograph so as to obtain an image quality and a high compression rate necessary for image interpretation. The details will be described below.

First, as a result of the image coding described in FIG.
The image is divided into fixed regions and encoded. Therefore,
A relationship as shown in FIGS. 3A and 3B occurs between the image and the compression code. That is, the original image is "image buffer 30a" and "image buffer 3a".
The number of lines depends on the total capacity of "0b" (region 1,
2, ~). {Refer to FIG. 3 (a)} ・ The compressed code data is generated by the number of types of coded (compressed) data. {Refer to FIG. 3 (b)} The data "delimiter code" indicating a break is added to each compression code data at a position corresponding to the position where the original image is divided.

Image compression normally encodes a previous value difference value, and this "delimiter code" is for clearing the previous value. Clearing the previous value is the same as compressing a new image. In this way, coded data for each area is obtained corresponding to the number of types of coding.

In the image coding method according to the present invention, the JPE, which is an international standard for coding (compressing) natural images, is actually used.
G specifications (according to ISO 10918-1 JPEG standard recommendations) are used.

The "delimiter code" is the R of the JPEG.
Use the ST (restart) marker. The header includes information necessary for performing the decompression process, such as "matrix size of reconstructed image (number of vertical and horizontal pixels)", "compression type", "quantization table", "Huffman table", etc. .

Each encoding generated as described above
The method of restoring the encoded data corresponding to the type of (compression), performing the predetermined image processing and displaying the data is shown below. Image display method: (1) The processes of (1) to (3) described in FIG. 2 are performed to display an image of the image quality necessary for image interpretation.

(2) When the operator designates a closed region of a portion required for image interpretation (or a closed region not required for image interpretation), the coding division region in which all of the region (other than the region) is included is shaded in FIG. Determine as shown in.

The coded divided areas are determined by obtaining the maximum and minimum values in the vertical (vertical) direction of the specified closed area, and determining the maximum and minimum values in the vertical (vertical) direction of each of the divided areas of the coded area. This can be easily done by comparing with the direction.

(3) The portion other than the determined compressed divided area is
Replace with an image with a high compression rate (= poor quality). When the operator gives an instruction, the image quality is raised or lowered and finally decided. Alternatively, if no particular instruction is given, the image may be replaced with an image with the maximum compression rate (= the image with the worst image quality).

(4) Based on the information determined in (3), the generated new encoded data is reconstructed as shown in FIG. 5 (b). (5) The encoded data other than the reconstructed image is deleted from the compressed encoded data file 11 of the file device 1 shown in FIG.

The "compression code area 23" and the "encoding circuit"
Two "transfer buffers 22" between 31 "(in Fig. 1,
The transfer buffers 22a and 22b are provided}, and "compression code area 2
3) and the transfer buffer 22 to transfer the encoded data and the transfer buffer 22 to the encoding circuit 31 to transfer the encoded data in parallel to perform the encoding (compression) processing. , And the decoding (restoration) process is performed in parallel. That is, the "transfer buffer 22a" and the "transfer buffer 22b" are provided on the main storage device 2a of the computer device 2, and at the time of encoding, the "encoding circuit 31" makes one of the "transfer buffer 22a"
While writing the encoded data in "a", the encoded data in "transfer buffer 22b" on the other side is written in "central processing unit (CPU)".
24 "is written in" compression code area 23 ". At the time of restoration, the "Central Processing Unit (CPU) 24" is
While the encoded data of "3" is being written to the "transfer buffer 22a", the "encoding circuit 31" is configured to read the encoded data of the "transfer buffer 22b" on the opposite side, so that Code area 23 "and" Transfer buffer 22 "
It is clear that the transfer of the encoded data between the two and the transfer of the encoded data between the "transfer buffer 22" and the "encoding circuit 31" can be processed in parallel.

Next, an embodiment corresponding to claim 4 will be described with reference to FIG. That is, in this embodiment, encoding (compression)
"Quantization table" and "Huffman table" during processing
If one or both of the set values of is the same as the value used immediately before, the "encoding mechanism" (encoding circuit 31 in FIG. 1)
It is a method of speeding up the encoding process of the image by omitting the setting.

First, as shown in FIG. 6, each of them is given a unique number and patterned into n types of quantization tables, m types of Huffman tables, and the immediately preceding encoding. The numbers of the quantization table and the Huffman table used in the above are configured to store the memories "previous quantization table ID" and "previous Huffman table ID", respectively, and are notified for encoding (compression). The "quantization table ID" and the "Huffman table I" included in the current compression condition
"D", "previous quantization table ID" and "previous Huffman table ID" are stored in the central processing unit (C
PU) 24 compares and if they match, sending of the matched table value (actual value) to the encoding circuit 31 is omitted. still,
The quantization table and the Huffman table may be stored in the main storage device and may be stored in the file device 1 or the like.

Next, an embodiment corresponding to claim 5 will be described with reference to FIG. That is, in this embodiment, at the time of restoration processing,
The "quantization table ID" and the "Huffman table I" stored in the header portion of the encoded data shown in the figure
When one or both of the set values of “D” are the same as the previously used “previous quantization table ID” and “previous Huffman table ID” values used immediately before, “encoding mechanism” (see FIG. 1). This is a method in which setting in the encoding circuit 31) is omitted.

First, an n-type patterned quantization table and a m-type patterned Huffman table, each of which is given a unique number, are defined. In the header part, as the quantization table ID and the Huffman table ID,
A memory "previous quantization table ID" and a "previous Huffman table ID" that store the numbers and respectively store the quantization table numbers and the Huffman table numbers used in the immediately preceding decoding are configured to be used at the time of restoration processing. , "Quantization table ID" and "Huffman table ID" to be restored, "previous quantization table ID" and "previous Huffman table ID", the computer device
The central processing unit (CPU) 24 of No. 2 makes a comparison, and if they match, the sending of the matched table value (actual value) to the coding circuit 31 is omitted to speed up the coding process of the image. You can

As described above, according to the present invention, in order to perform two or more types of image compression (encoding), the original image data of a predetermined size is transferred only once, and only the encoding parameters are set thereafter. A plurality of types of image compression (encoding) for the size is performed. Further, at the time of the image encoding, an original image ID for identifying an original image is added, a plurality of types of encoded data are stored in a database, the encoding type and the original image ID are designated, and the encoding is performed. The image of the data is restored, the image subjected to the instructed image processing is displayed, and the coded data of the unnecessary image is deleted from the database according to the deterioration of the image quality of the displayed image. Further, at the time of encoding / restoring processing, if the setting ID of the quantization table or Huffman table is the same as the ID used immediately before, setting of the values of the quantization table or Huffman table is omitted in the encoding circuit. There is a feature in doing so.

[0073]

As described above in detail, according to the image coding method of the present invention, it is possible to shorten the processing time of a plurality of types of image compression, and to reduce the image quality required for image interpretation / diagnosis / saving. The compressed image can be easily selected, and the unnecessary data is deleted, whereby the amount of image data held by the image coding apparatus can be reduced. or,
Omitting the transfer of the quantization table and Huffman table has the effect of speeding up the image compression / decompression processing time.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention (No. 1).

FIG. 2 is a diagram showing an embodiment of the present invention (Part 2).

FIG. 3 is a diagram showing an embodiment of the present invention (part 3).

FIG. 4 is a diagram showing an embodiment of the present invention (No. 4).

FIG. 5 is a diagram showing an embodiment of the present invention (No. 5).

FIG. 6 is a diagram showing an embodiment of the present invention (No. 6).

FIG. 7 is a diagram showing an embodiment of the present invention (No. 7).

FIG. 8 is a diagram illustrating a conventional image encoding device.

[Explanation of symbols]

1 File device 10 Original image data file 11 Compressed coded data file 2 Computer device 20 Image area 21 Transfer buffer 22,22a, 22b
Transfer buffer 23 Compressed code area 2a Main memory 30a, 30b Image buffer 31 Encoding circuit, Encoding mechanism Quantization table 1-n Huffman table 1-m Previous quantization table ID Previous Huffman table ID This quantization table ID This time Huffman table ID of the Huffman table I of the header section of the compressed code data
D

Claims (5)

[Claims] 1. An original image data file of a file device (1)
Read a part or all of the original image data from (10),
The image data (20) of the main storage device (2a) of the computer device (2) is copied, and the image data for the capacity of the transfer buffer (21) of the main storage device (2) is transferred from the image area (20). Copy to the transfer buffer (21), set the predetermined encoding parameters in the encoding circuit (31) of the image encoding mechanism (3), activate the encoding circuit (31), and transfer the transfer buffer (21). ) To the image buffer (30a or 30b)
Copy a predetermined amount of image data to, switch the image buffer (30a, or 30b), connect the image buffer (30a, or 30b) to the encoding circuit 31,
Connect the image buffer (30b or 30a) to the computer device,
A method for encoding image data in the image buffer (30a, 30b) by using switching of the image buffer (30a, 30b) as a trigger and transferring the encoded data to a computer device, the file device Image buffer from (1) (30a, or 30
For one transfer of image data to b), a computer device
The image code characterized by repeating the setting of a plurality of types of encoding parameters from (2) to encode the plurality of types of image data for the capacity of the image buffer (30a, 30b). Method. 2. The image encoding method according to claim 1, wherein the original image data file (1
(0) Each original image data is added with an original image ID, and a plurality of types of encoded data encoded by the encoding circuit (31) are used as a database to form a compressed encoded data file (11).
In this case, the encoding type and the original image ID are designated, the plural types of images of the original image ID are restored, and the restored image of the designated original image ID and the designated encoding type is restored. On the other hand, when the predetermined image processing is instructed, after performing the predetermined image processing on all the restored images corresponding to the plurality of coding types,
Image coding characterized by displaying an image of the above-mentioned coding type and selectively deleting the coded data of an unnecessary image from the displayed image according to image quality deterioration Method. 3. An image coding method, wherein the image coding according to claim 2 is performed for each predetermined area of original image data. 4. The image coding method according to claim 1 or 2, wherein a unique number is assigned to each of the coding circuits (31) to set the coding parameters. The previously stored n types of patterned quantization tables (), m types of patterned Huffman tables (), the numbers of the quantization tables used in the immediately preceding encoding process, and the Huffman table numbers are stored. Quantization table ID () and previous Huffman table ID () are provided, and in order to encode original image data from this, a computer device
The quantization table ID () included in the encoding condition notified from the central processing unit (24) of (2), the Huffman table ID (), the previous quantization table ID (),
The previous Huffman table ID () is compared by the central processing unit (24), and when they match, the matched table value is controlled so as not to be transferred to the encoding circuit (31). Image coding method. 5. The image coding method according to claim 1 or 2, wherein decoding parameters are set in the coding circuit (31) to restore coded data, The n kinds of patterned quantization tables () with unique numbers, the m kinds of patterned Huffman tables (), the numbers of the quantization tables used in the immediately preceding encoding process, and the Huffman The previous quantization table ID () for storing the table number and the previous Huffman table ID () are provided, and in order to restore the encoded data by the encoding circuit (31), the header portion of the encoded data is used. The stored quantization table ID (), Huffman table ID (), the previous quantization table ID (), and the previous Huffman table ID () are stored in the central processing unit of the computer device (2).
The image decoding method, characterized in that the table values are matched so as not to be transferred to the coding circuit (31) when compared in (24) and if they match.