JP2007110419A - Image magnification method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image magnification method which subjects an image to a magnification processing by the block and is capable of preventing a line from appearing at a boundary between the adjacent blocks after a magnification processing is carried out. <P>SOLUTION: When an image is subjected to a magnification processing in a first direction, a first storage means keeps original data or data obtained from data as to a block which is one of blocks located adjacent to a processing object block in the first direction, and whose processing turn is immediately followed by that of the processing object block, and data necessary for subjecting the processing object block to a magnification processing in the first direction are chosen from the above data and stored in the first storage means. The processing object block is subjected to a magnification processing in the first direction using the data stored in the first storage means. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image scaling method.

  In the case of enlarging an image in the main scanning direction (or sub-scanning direction), as shown in FIG. 1, the pixel data adjacent in the main scanning direction (or sub-scanning direction) is linearly complemented so that Is generated. When the image is reduced in the main scanning direction (or sub-scanning direction), as shown in FIG. 2, the data after reduction is obtained by averaging pixel data adjacent in the main scanning direction (or sub-scanning direction). Is generated.

  When such a scaling process is performed by a device in which pixel data is input from an external storage device in a raster form, a buffer in units of lines is required to perform scaling in the sub-scanning direction.

Therefore, in order to reduce the buffer capacity, it is considered to divide the image into M × N blocks A, B, C, D,... And perform scaling processing in units of blocks as shown in FIG. It is done. When the scaling process is performed in units of blocks in this way, streaks may appear at the boundary between adjacent blocks after the scaling process.
JP-A 63-74369

  An object of the present invention is to provide an image scaling method capable of preventing streaks from appearing at the boundary between adjacent blocks after the scaling process in an image scaling method that performs a scaling process in units of blocks.

  According to the first aspect of the present invention, there is provided an image scaling method for dividing an image into a plurality of blocks and performing a scaling process on a block-by-block basis, in either one of a horizontal direction and a vertical direction (first In the image scaling method for performing the scaling process in the other direction (second direction) using the scaling process result in the (direction), when performing the scaling process in the first direction, Of the blocks adjacent to each other in the direction, the data necessary for the scaling process in the first direction with respect to the processing target block among the original data of the block one processing order before the processing target block or the data obtained from those data When the scaling process in the first direction is performed on the processing target block using the data stored in the first storage unit and stored in the first storage unit, and the scaling process in the second direction is performed. Is the target bro Among the blocks adjacent in the second direction to the processing target block, the first scaling processing result data in the first direction of the block whose processing order is one before the processing target block or the data obtained from those data is Data necessary for the scaling process in the two directions is stored in the second storage unit, and the scaling process in the second direction is performed on the processing target block using the data stored in the second storage unit. It is characterized by.

  According to the present invention, in the image scaling method that performs the scaling process in units of blocks, it is possible to prevent streaks from appearing at the boundary between adjacent blocks after the scaling process.

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

[1] Circuit configuration for scaling processing

FIG. 4 shows a configuration of a circuit for performing the scaling process.
The external storage device 1 stores image data to be subjected to scaling processing. The image processing circuit 2 performs a scaling process on the image data input from the external storage device 1 and stores the image data after the scaling process in the external storage device 1.

  The image processing circuit 2 includes a main scanning scaling circuit 11 and a sub scanning scaling circuit 12. The main scanning zoom circuit 11 includes a line memory 21. A sub-scanning zoom circuit 12, a line memory 22 and a line memory 23 are provided.

  As shown in FIG. 3, the image processing circuit 2 divides the image into M × N blocks A, B, C, D,... And performs a scaling process on a block basis.

FIG. 5 shows blocks A to H among the plurality of blocks shown in FIG. FIG. 6 shows block A. A _ij in block A (i represents a column and j represents a row) represents pixel data in block A. B _ij , C _ij , D _ij , and E _{ij in the} blocks B, C, D, and E also represent pixel data in the block.

  The pixel data in the block is processed in a raster shape as shown in FIG. In block units, as shown in FIG. 5, processing is performed in alphabetical order (ABC order).

  When scaling processing is performed in units of blocks, streaks appear at the boundary between adjacent blocks after scaling processing, depending on the magnification, it is necessary to perform linear interpolation or averaging of pixel data across block boundaries However, at this time, the processing order does not use the data of the block immediately before the processing target block. Therefore, in this embodiment, among the data obtained from the block whose processing order is one before the processing target block, data necessary for the scaling process for the processing target block is stored in the line memory or the external storage device 1. The data is used when the scaling process is performed on the processing target block.

[2] Enlargement process

  In the following, the scaling process method will be described by taking as an example the case where the scaling ratio is 200%. The scaling process with a scaling factor of 200% is a process of generating double image data from the original image in each of the main scanning direction (horizontal direction) and the sub-scanning direction (vertical direction). In this embodiment, a case where a scaling process in the sub-scanning direction (vertical direction) is performed using a scaling process result in the main scanning direction (horizontal direction) will be described.

[2-1] 200% enlargement processing in the main scanning direction

(1) Processing for block A is performed.

Input pixel data is represented by A _ij , and pixel data after enlargement processing is represented by A ′ _ij .
As shown by the following equation (1), the main scanning scaling circuit 11 performs enlargement processing in order from the top pixel, and sends the processed data to the sub-scanning scaling circuit 12.

A _'00 = A ₀₀
A ′ ₀₁ = A ₀₀ + (A ₀₁ −A ₀₀ ) / 2
A _'02 = A ₀₁
A ′ ₀₃ = A ₀₁ + (A ₀₂ −A ₀₁ ) / 2
...
A ' ₀₁₄ = A ₀₇ (1)

When the processing of the first line of block A is completed, the final data A ₀₇ of the first line of block A is written into the line memory 21.

  Similar processing is repeated for the second and subsequent lines of block A.

(2) When processing for all lines in block A is completed, processing for block B is performed.

  As shown by the following equation (2), the main scanning scaling circuit 11 performs enlargement processing in order from the top pixel, and sends the processed data to the sub-scanning scaling circuit 12.

B ′ ₀₀ = A ₀₇ + (B ₀₀ −A ₀₇ ) / 2
B _'01 = B ₀₀
B ′ ₀₂ = B ₀₀ + (B ₀₁ −B ₀₀ ) / 2
B _'03 = B ₀₁
...
B ′ ₁₅ = B ₀₇ (2)

To generate the B _'00 uses the data of the A ₀₇ stored in the line memory 21. When the processing of the first line of block B is completed, the data A ₀₇ used in the processing of the first line in the line memory 21 is replaced with the final data B ₀₇ of the first line of block B.

  Similar processing is repeated for the second and subsequent lines of block B.

(3) A similar process is performed on blocks after block C.

[2-2] 200% enlargement process in the sub-scanning direction

(1) Processing for block A is performed.
The final output data is represented by A ″ _ij . The sub-scanning scaling circuit 12 has data A ′ _{00 to} A ′ _0X (X is a variable after processing of the first line of the block A from the main scanning scaling circuit 11. Is obtained in accordance with the magnification) and written to the line memory 22. Also, the data A ′ _{10 to} A ′ _1X after the processing of the second line of the block A is obtained from the main scanning scaling circuit 11, and the line memory 23 is written.

  Based on the data written in the line memories 22 and 23, the sub-scan scaling circuit 12 performs an enlargement process as shown by the following equation (3).

A ” ₀₀ = A ₀₀ '
A ” ₀₁ = A _'01
A ” ₀₂ = A _'02
...
_{A "10 = A 00 '+} (A' 10 -A '00) / 2
A " ₁₁ = A ₀₁ '+ (A' ₁₁ -A _'01 ) / 2
_{A "12 = A 02 '+} (A' 12 -A '02) / 2
...
... (3)

When the data for the first line and the second line are obtained in this way, the sub-scanning zoom circuit 12 receives the data A ′ _{20 to} A after processing of the third line of the block A from the main scanning zoom circuit 11. 'Get _2X and overwrite line memory 22

  Based on the data written in the line memories 22 and 23, the sub-scan scaling circuit 12 performs an enlargement process as shown in the following equation (4).

A ” ₂₀ = A ₁₀ '
A _"21 = A _'11
A _"22 = A _'12
...
_{A "30 = A 10 '+} (A' 20 -A '10) / 2
_{A "31 = A 11 '+} (A' 21 -A '11) / 2
_{A "32 = A 12 '+} (A' 22 -A '12) / 2
...
... (4)

When the processing for the block A is completed, the processed data A ′ _{70 to} A ′ _7X for the last line of the block A from the main scanning scaling circuit 11 are written as auxiliary data in the external storage device 1.

Similar processing is performed for blocks B, C, and D. At this time, the data B ′ _{70 to} B ′ _7X , C ′ _{70 to} C ′ _7X , and D ′ _{70 to} D ′ _7X after the processing for the final lines of the blocks B, C, and D from the main scanning scaling circuit 11 are external. It is written in the storage device 1 as auxiliary data.

(2) When the process for block D is completed, the process for block E is performed.
The sub-scanning scaling circuit 12 acquires data E ′ _{00 to} E ′ _0X (X varies depending on the scaling factor) after processing the first line of the block E from the main scanning scaling circuit 11, and the line memory 22 is written. In addition, data E ′ _{10 to} E ′ _1X after the processing of the second line of the block E is acquired from the main scanning scaling circuit 11 and written into the line memory 23.

The sub-scanning scaling circuit 12 is a main-scanning scaling circuit 11 for the last line of the block A above the block E, which is written as auxiliary data in the external memory 1 and the data written in the line memories 22 and 23. Based on the data A ′ _{70 to} A ′ _7X after the processing according to, enlargement processing is performed as shown in the following equation (5).

_{E "00 = A 70 '+} (E' 00 -A 70 ') / 2
E " ₀₁ = A ₇₁ '+ (E' ₀₁ -A ₇₁ ') / 2
_{E "02 = A 72 '+} (E' 02 -A 72 ') / 2
...
E ” ₁₀ = E _'00
E ” ₁₁ = E _'01
E ” ₁₂ = E _'02
...
E " ₂₀ = E ₀₀ '+ (E' ₁₀ -E ₀₀ ') / 2
E ” ₂₁ = E ₀₁ '+ (E' ₁₁ -E ₀₁ ') / 2
E ” ₂₂ = E ₀₂ '+ (E' ₁₂ -E ₀₂ ') / 2
...
... (5)

When the processing for the block E is completed, the processed data E ′ _{70 to} E ′ _7X for the last line of the block E from the main scanning scaling circuit 11 are written as auxiliary data in the external storage device 1.

  Thereafter, processing is performed in the order of blocks F, G, and H.

[3] Reduction processing

  In the following, the scaling process method will be described by taking the case where the scaling ratio is 33% as an example. The scaling process with a zoom ratio of 33% is a process for generating 1/3 times image data from the original image in each of the main scanning direction and the sub-scanning direction.

[2-1] 33% reduction processing in the main scanning direction

(1) Processing for block A is performed.
Input pixel data is represented by A _ij , and pixel data after reduction processing is represented by A ′ _ij .
As shown in the following equation (6), the main scanning scaling circuit 11 performs reduction processing in order from the top pixel and sends the processed data to the sub-scanning scaling circuit 12.

A ′ ₀₀ = (A ₀₀ + A ₀₁ + A ₀₂ ) / 3
A ′ ₀₁ = (A ₀₃ + A ₀₄ + A ₀₅ ) / 3 (6)

Since the addition value (A ₀₆ + A ₀₇ ) of the remaining data A ₀₆ and A _{07 on} the first line is necessary to generate the reduced data B ′ ₀₀ on the first line of the block B, the processing on the first line Is completed, the data is written in the line memory 21. Each data A ₀₆ and A ₀₇ may be separately written in the line memory 21 instead of the added value (A ₀₆ + A ₀₇ ) of the remaining data A ₀₆ and A _{07 on} the first line.

  Similar processing is repeated for the second and subsequent lines of block A.

(2) When processing for all lines in block A is completed, processing for block B is performed.

  As shown by the following equation (7), the main scanning scaling circuit 11 performs reduction processing in order from the first pixel, and sends the processed data to the sub-scanning scaling circuit 12.

B ′ ₀₀ = (A ₀₆ + A ₀₇ + B ₀₀ ) / 3
B _'01 = (B ₀₁ + B ₀₂ + B ₀₃ ) / 3
B ′ ₀₂ = (B ₀₄ + B ₀₅ + B ₀₆ ) / 3 (7)

When B ′ ₀₀ is generated, data (A ₀₆ + A ₀₇ ) written in the line memory 21 is used. Since the remaining data B ₀₆ of the first line is necessary for generating the reduced data C ′ ₀₀ of the first line of the block C, when the processing of the first line is completed, (A ₀₆ in the line memory 21 is completed. + A ₀₇ ) is replaced with B ₀₆

  Similar processing is repeated for the second and subsequent lines of block B.

(3) A similar process is performed on blocks after block C.

[3-2] 33% reduction processing in the sub-scanning direction

(1) Processing for block A is performed.
The final output data is represented by A ″ _ij . The sub-scanning scaling circuit 12 has data A ′ _{00 to} A ′ _0X (X is a variable after processing of the first line of the block A from the main scanning scaling circuit 11. Is changed according to the magnification) and written to the line memory 22.

Next, the processed data A ′ _{10 to} A ′ _1X of the second line of the block A are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

Next, the processed data A ′ _{20 to} A ′ _2X of the third line of the block A are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

  Based on the data stored in the line memory 22, the sub-scanning scaling circuit 12 performs a reduction process as shown by the following equation (8).

_{A "00 = (A 00 '} + A 10' + A 20 ') / 3
A " ₀₁ = (A ₀₁ '+ A ₁₁ ' + A ₂₁ ') / 3 (8)

When the reduced data of the first line of the block A is obtained in this way, the sub-scanning scaling circuit 12 receives the data A ′ _{30 to} A after processing of the fourth line of the block A from the main scanning scaling circuit 11. ' _3X (X changes according to the magnification) is written in the line memory 22.

Next, the processed data A ′ _{40 to} A ′ _4X of the fifth line of the block A are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

Next, the processed data A ′ _{50 to} A ′ _5X of the sixth line of the block A are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

  Based on the data stored in the line memory 22, the sub-scanning scaling circuit 12 performs a reduction process as shown by the following equation (9).

A " ₁₀ = (A ₃₀ '+ A ₄₀ ' + A ₅₀ ') / 3
A ″ ₁₁ = (A ₃₁ ′ + A ₄₁ ′ + A ₅₁ ′) / 3 (9)

When the reduced data of the second line of the block A is obtained in this way, the sub-scanning scaling circuit 12 sends the data A ′ _{60 to} A ′ after the processing of the seventh line of the block A from the main scanning scaling circuit 11. ' _6X (X changes according to the scaling factor) is acquired and written to the line memory 22.

Next, the processed data A ′ _{70 to} A ′ _7X of the eighth line of the block A are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

Since there is no data after processing of the 9th line of block A, it is not possible to add data for 3 lines. Therefore, the addition data for two lines stored in the line memory 22 (in this example, (A ′ ₆₀ + A ′ ₇₀ ) and (A ′ ₆₁ + A ′ ₇₁ )) is written in the external storage device 1 as auxiliary data. . Instead of writing the addition data for two lines into the external storage device 1, the data A ′ ₆₀ , A ′ ₆₁ , A ′ ₇₀ , A ′ ₇₁ of each line may be written as auxiliary data into the external storage device 1. Good.

Similar processing is performed for blocks B, C, and D. At this time, the addition data (B ′ ₆₀ + B ′ ₇₀ ) and (B ′ ₆₁ + B ′ ₇₁ ), (C ′ ₆₀ + C ′ ₇₀ ) and (C ′ ₆₁ + C ′ ₇₁ ), (D ′ ₆₀ + D) for two lines _'70 ) and (D' ₆₁ + D' ₇₁ ) are written to the external storage device 1 as auxiliary data.

(2) When the process for block D is completed, the process for block E is performed.
The sub-scanning scaling circuit 12 acquires data E ′ _{00 to} E ′ _0X (X varies depending on the scaling factor) after processing the first line of the block E from the main scanning scaling circuit 11, and the line memory 22 is written. Next, (A ′ ₆₀ + A ′ ₇₀ ) and (A ′ ₆₁ + A ′ ₇₁ ) stored as auxiliary data in the external storage device 1 are read out, added to the corresponding data in the line memory 22, and the addition result is displayed as a line. Store in the memory 22.

  Based on the data stored in the line memory 22, the sub-scanning scaling circuit 12 performs a reduction process as shown by the following equation (10).

E " ₀₀ = ( _A60 '+ _A70 ' + _E00 ') / 3
E " ₀₁ = (A ₆₁ '+ A ₇₁ ' + E ₀₁ ') / 3 (10)

When the reduced data of the first line of the block E is obtained in this way, the sub-scanning scaling circuit 12 receives the data E ′ _{10 to} E after processing of the second line of the block E from the main scanning scaling circuit 11. ' _1X (X changes according to the magnification) is written in the line memory 22.

Next, the processed data E ′ _{20 to} E ′ _2X of the third line of the block E are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

Next, the processed data E ′ _{30 to} E ′ _3X of the fourth line of the block E are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

  Based on the data stored in the line memory 22, the sub-scanning scaling circuit 12 performs a reduction process as shown by the following equation (11).

E " ₁₀ = (E ₁₀ '+ E ₂₀ ' + E ₃₀ ') / 3
E " ₁₁ = (E ₁₁ '+ E ₂₁ ' + E ₃₁ ') / 3 (11)

When the reduced data of the second line of the block E is obtained in this way, the sub-scanning scaling circuit 12 receives the data E ′ _{40 to} E after processing of the fifth line of the block E from the main scanning scaling circuit 11. ' _4X (X changes according to the magnification) is written into the line memory 22.

Next, the processed data E ′ _{50 to} E ′ _5X of the sixth line of the block E are obtained from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

Next, the processed data E ′ _{60 to} E ′ _6X of the seventh line of the block E are acquired from the main scanning scaling circuit 11 and added to the corresponding data in the line memory 22, and the addition result is stored in the line memory 22. Store.

  Based on the data stored in the line memory 22, the sub-scanning scaling circuit 12 performs a reduction process as shown by the following equation (12).

E " ₂₀ = (E ₄₀ '+ E ₅₀ ' + E ₆₀ ') / 3
E " ₂₁ = (E ₄₁ '+ E ₅₁ ' + E ₆₁ ') / 3 (12)

When the reduced data of the third line of the block E is obtained in this way, the sub-scanning scaling circuit 12 receives the data E ′ _{70 to} E after processing of the eighth line of the block E from the main scanning scaling circuit 11. ' _7X (X changes according to the magnification) is written in the line memory 22.

Since there is no data after processing of the 9th line of block E, the data for 3 lines cannot be added. Therefore, the data for one line stored in the line memory 22, in this example, E ′ ₇₀ and E ′ ₇₁ are written in the external storage device 1 as auxiliary data.
Thereafter, processing is performed in the order of blocks F, G, and H.

[4] General scaling

  The enlargement process of [2] described the enlargement process of integer multiples, and the reduction process of [3] described the reduction process of 1 / integer. The present invention can also be applied to the magnification.

  In other words, when the scaling factor is other than an integer multiple or a fraction of an integer, by combining an integer multiple enlargement process as described above and an integral fraction reduction process as described above, Perform scaling processing.

  For example, when the scaling factor is 99%, the enlargement process of 99 times is performed, and then the reduction process of 1/100 is performed.

  According to the above embodiment, since the scaling process is performed in units of blocks, the capacity of the live memory can be reduced. Further, among the original data of the block one processing order before the processing target block or the intermediate data of the enlargement processing, the data necessary for the scaling process for the processing target block is stored in the line memory or the external storage device, Since the data is used when the scaling process is performed on the processing target block, it is possible to prevent a streak from appearing at the boundary between adjacent blocks after the scaling process.

  In the above embodiment, the scaling process result in the sub-scanning direction (vertical direction) is performed using the scaling process result in the main scanning direction (horizontal direction), but the scaling process result in the sub-scanning direction (vertical direction). May be used to perform scaling processing in the main scanning direction (horizontal direction).

It is a schematic diagram for demonstrating the conventional image expansion method. It is a schematic diagram for demonstrating the conventional image reduction method. It is a block diagram which shows a mode that the image was divided | segmented into the block A, B, C, D, ... of the number of MxN. It is a block diagram which shows the structure of the circuit for performing a scaling process. FIG. 4 is an enlarged view showing blocks A to H among a plurality of blocks shown in FIG. 3. FIG. 6 is an enlarged view showing a block A among a plurality of blocks shown in FIG. 5.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 External storage device 2 Image processing circuit 11 Main scanning scaling circuit 12 Sub scanning scaling circuit 21, 22, 23 Line memory

Claims (1)

An image scaling method that divides an image into a plurality of blocks and performs a scaling process in units of blocks, and uses a scaling process result in one of the horizontal direction and the vertical direction (first direction). In the image scaling method for performing scaling processing in the other direction (second direction),
When performing the scaling process in the first direction, among the blocks adjacent to the processing target block in the first direction, the original data of the block whose processing order is one before the processing target block or the data obtained from those data Among them, data necessary for the scaling process in the first direction for the processing target block is stored in the first storage unit, and the first data for the processing target block is stored using the data stored in the first storage unit. Perform direction scaling,
When performing the scaling process in the second direction, among the blocks adjacent to the processing target block in the second direction, the scaling process result data in the first direction of the block whose processing order is one before the processing target block or Of the data obtained from these data, data necessary for the scaling process in the second direction for the processing target block is stored in the second storage unit, and the data stored in the second storage unit is used. An image scaling method characterized by performing a scaling process in the second direction on the processing target block.

Images