JPH1093805A - Picture processing method and picture processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a picture processing method for reducing cost of a processor and to provide a picture processor by executing a picture rotation processing through a use of a memory of small capacity. SOLUTION: Pixel data which is read in a line unit in a read part 2 is stored in grid line buffers 7a and 7b for m-lines. Pixel data for m-lines is divided into blocks at every n-pixels in a main scanning direction. An encoding part 3 executes encoding for respective block units and data are accumulated in an accumulation memory 9. Encoded data are decoded in the block unit in a decoding part 4 and picture element data of m-pixels × n-pixels is rotated in a 90 deg. unit. The respective blocks are decoded in a prescribed order different from an encoding order in accordance with the rotation angle of the designated restored picture and a conversion processing to pixel array corresponding to the rotation angle of the designated restored picture is execute. Data is stored in a line buffer 8 from the prescribed direction corresponding to the rotation direction and pixel data of n.lines or m.lines are sequentially generated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and apparatus, and more particularly to image rotation processing in a multifunction device having both a facsimile function and a copier function.

[0002]

2. Description of the Related Art Conventionally, the main scanning direction of reading / recording in a copying machine is generally the long side direction for A4 documents and the short side direction for A3 documents. This has the advantage that, as shown in FIG. 21, the length of the A3 document and the A4 document in the main scanning direction can be unified, and in the case of the A4 document, the number of scanning lines can be reduced, that is, the copy speed is increased. Because there is.

On the other hand, in facsimile machines, IT
According to the recommendations of U-T3, T4, T5, etc., it is standardized that the main scanning direction of the transmission original is the short side direction of the original.

Therefore, if an attempt is made to realize a multifunction machine of a copying machine and a facsimile, inconsistency occurs in handling of A4 images.

As one solution to this reading, for an A4-size facsimile transmission original, it is conceivable to specify the direction in which the original is set in the reading section such that the short side is the main scanning direction. However, this solution imposes a complicated operation on the user, and if the user makes a mistake in the operation, there is a danger that the device may mistakenly recognize the A3 size and transmit the reduced size, and the user may not notice it. There is a serious disadvantage that there is no.

Therefore, it is desirable to read the A4 original in the same direction as that during copying (ie, in the direction in which the long side direction is the main scanning direction) during facsimile transmission. To realize this, processing for rotating the read image by 90 ° at the time of facsimile transmission is required.

[0007] Also, the printing process has the same inconsistency as the reading process. As the simplest solution, A
Although there is a method including a paper feeding unit capable of feeding two types of recording paper, that is, four lengths and four A4 widths, there are disadvantages such as an increase in cost and an increase in size of the apparatus.

[0008] Therefore, the A4 size paper feed unit can be reduced to one by rotating the facsimile reception image by 90 ° in the print processing unit and then performing the print processing.

Further, when printing is performed on both sides of the recording paper, and when the short side of the recording paper is set in the main scanning direction, it is necessary to rotate the image to be printed on the back side by 180 °.

As described above, in a multifunction device of a facsimile and a copying machine, image rotation processing is performed with ease of operation, cost, and cost.
This is an indispensable process from the viewpoint of the volume of the device.

Conventionally, this rotation processing has been performed by performing vertical and horizontal conversion one pixel at a time using a bitmap page memory for one page of the image to be rotated.

The capacity of the bit map page memory is as follows:
Even with A4 size 1 page / resolution 200 dpi,
Requires 0.5 MB. If the facsimile transmission operation and the printing of the already received image are to be performed simultaneously, an additional 0.5 MB is required. Further, if a bitmap page memory for two pages for writing an image before rotation and for reading an image after rotation is provided to improve the throughput of the apparatus, a double memory capacity is required. Become.

The above is the memory capacity in the case of "fine" resolution that is common in facsimile machines. However, when viewed as a copying machine, a resolution of at least 400 dpi is required. Further, in the case of 400 dpi, the above memory capacity needs to be further increased by four times, and 2M for one page of A4 size.
Requires as much capacity as bytes.

[0014]

As described above, the conventional image rotation processing method requires a bitmap page memory for at least one image page.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and provides an image processing method and apparatus in which the cost of the apparatus is reduced by performing image rotation processing using a small-capacity memory. With the goal.

[0016]

To achieve the above object, an image processing apparatus according to the present invention comprises a reading means for reading a document line by line, and at least m lines of pixel data read by the reading means can be stored. Storage means, encoding means for encoding pixel data stored in the storage means, and accumulation means capable of accumulating a plurality of pages of encoded data encoded by the encoding means. The means divides the m-line pixel data stored in the storage means into blocks of n pixels in the main scanning direction, and performs encoding in units of each block.

According to another aspect of the present invention, there is provided an image processing method comprising: reading a document in units of lines; storing read pixel data in a storage unit for at least m lines; The pixel data is divided into blocks of n pixels in the main scanning direction, encoded in units of blocks, the encoded data is decoded in units of blocks, and the restored pixel data of m pixels × n pixels is rotated in units of 90 °. ,
In accordance with the rotation angle of the specified restored image, each block is decoded in a predetermined order different from the encoding order, and a conversion process to a pixel array corresponding to the rotation angle of the specified restored image is performed. Store from a predetermined direction according to the rotation direction,
It is characterized in that pixel data of n or m lines is sequentially created.

[0018]

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

The rotation processing in this embodiment is realized by dividing an image into blocks (hereinafter, referred to as "blocks") each having m pixels.times.n pixels.

The exchange of image data among the reading section, the recording section, and the facsimile transmitting / receiving section is performed via a storage memory for storing code data compression-encoded in block units.

<Storage Processing> When image data is stored in the above-described storage memory, pixel data for m lines (hereinafter, referred to as "stripe") is temporarily stored in a line buffer. Next, the encoding processing unit extracts n pixels from the m line buffer in the main scanning direction. This is called a first block of the stripe.

Subsequently, the encoding processing unit compresses and encodes the first block, and stores three types of information, block code information / block code length information / code information storage start position information of the first block, in the storage memory. In a predetermined area. Here, two pieces of information of the block code length information and the code information storage start position information of the first block are fixed length information.

Subsequently, n pixels are successively taken out of the line buffer, and the second to last blocks are taken out of the first block.
The data is compressed and encoded in the same manner as the block, and is stored in the storage memory. However, the information stored in the storage memory is only two types of information, block code information / block code length information.

Thereafter, the processing from the second stripe to the last stripe of the document is processed in the same manner as the first stripe.

As a result, the storage memory has three types of information: (1) block code information, (2) block code length information of each block, and (3) storage start position information of the code data of the first block of each stripe. Is stored.

<Decompression and Rotation Processing> In Case of Decompression Only First, the decryption processing unit refers to the storage start position information and
The storage position of the code information of the first block of the stripe is determined. Next, since the first information of the block code length information indicates the code length of the first block, a predetermined amount of block code information is read from the storage position and decoding processing of the first block is performed. The image data of n pixels × m lines obtained as a result of the decoding is stored in each line of the line buffer of m lines by n pixels in the main scanning direction.

Next, the storage start position information of the second and subsequent blocks of the first stripe is sequentially obtained by adding the block code length information to the storage start position information of the previous block. Therefore, based on the information, similar to the first block,
The second and subsequent blocks are decoded and stored in the line buffer.

Thus, the pixel data of the first stripe composed of m lines is restored. After the second stripe,
The decoding process is performed in the same order as the first stripe. When rotating by 90 ° after decompression First, the decoding processing unit determines the storage position of the target block code information using the storage start position information and the block code length information, and performs the decoding process of the block itself. Is the same as the case of only decoding

Here, two points are different in the order of the blocks to be decoded and the decoding result of each block is rotated by 90 ° and then stored in the line buffer.

In other words, the block decoding process is performed in the order of processing all the first blocks from the first stripe to the last stripe, processing all the second blocks in the same manner, and so on.

In the 90 ° rotation processing, image data of n pixels × m lines as a result of decoding is rotated to the right by 90 ° for each block, converted to m pixels × n lines, and then sequentially stored in a line buffer of n lines. The data is stored from the direction opposite to the main scanning direction. -Rotation by 180 degrees after stretching Rotation by 180 degrees is necessary when two-sided printing is performed on an image whose main scanning is in the short side direction of the paper. The order of stripes to be decoded is from the last stripe toward the first stripe. The decoding order of the blocks is performed in the order of the first block, the second block,... For each stripe. Then, the image data of n pixels × m lines, which is the decoding result of each block, is stored in the line buffer in 18 lines.
After the rotation by 0 °, the data is sequentially stored in the line buffer of m lines in the direction opposite to the main scanning direction.

FIG. 1 is a block diagram showing the configuration of a multifunction device according to the present embodiment, which is a device in which a facsimile and a copying machine are combined. In FIG. 1, reference numeral 1 denotes a system control unit which controls the present system and performs the following image coding / decoding processing at the time of facsimile transmission / reception.

At the time of facsimile transmission, the code data in the storage memory is once decoded, re-encoded by an encoding method that matches the decoding capability of the other party's facsimile receiving apparatus, and the facsimile transmission processing is performed. At the time of facsimile reception, the received facsimile code data is once decoded, and the image data obtained as a result is re-encoded according to “encoding suitable for image rotation processing” and stored in the storage memory.

Reference numeral 2 denotes a reading unit, which reads an image in line units with A3 and B4 originals on the short side and A4 originals on the long side in the main scanning direction, and sequentially stores them in the line buffers 7a and 7b. Reference numeral 3 denotes an encoding unit, which is a line buffer 7a or 7
The image data stored in b is encoded according to “encoding suitable for image rotation processing” and stored in the storage memory 9. 4
Is a decoding unit that decodes the code data stored in the storage memory 9 in accordance with “encoding suitable for image rotation processing”, and performs non-rotation 90 ° according to the document size and the recording paper size.
The pixel data is stored in the line buffer 7a or 7b by performing a rotation or a 180 ° rotation.

Reference numeral 5 denotes a printing unit, which is a line buffer 7a,
The image data stored in 7b is sequentially read out and printed on recording paper. Reference numeral 6 denotes a buffer control unit which manages the presence / absence of image data stored in the line buffers 7a and 7b, and controls the system control unit 1 for both line buffers.
It arbitrates for access requests from / reading unit 2 / encoding unit 3 / decoding unit 4 / printing unit 5. Reference numerals 7a and 7b denote line buffer memories having a capacity for storing image data of at least one stripe.

Reference numeral 8 denotes a line buffer memory for temporarily storing image data when the system control unit 1 performs the same encoding / decoding processing as the encoding unit 3 and the decoding unit 4 for the facsimile transmission / reception operation. Memory. Reference numeral 9 denotes a storage memory for storing encoded code data.
A communication unit 10 modulates a signal to perform facsimile transmission and reception via a communication line.

S1 to S4 denote an access request signal from each of the processing units 2 to 5 to the buffer control unit 6 to access the line buffer 7a or 7b, and an access generated by the control unit 6 in response to the signal to permit access. This is a permission signal.
D1 to D4 are image data buses through which the line buffer 7a or 7b inputs and outputs image data via the processing units 2 to 5 and the buffer control unit 6. S5 is a line buffer 7a or 7b generated by the buffer control unit 6 and previously associated with the processing contents of each of the processing units 2 to 5 that have permitted the access request.
Address signal and access control signal. D5 is a data input / output bus between the line buffers 7a and 7b and the buffer controller 6. In step S6, when the system control unit 1 requests access to the line buffer 7a or 7b, the system control unit 1 waits until the buffer control unit 6 permits the access.
Is a wait signal for extending the access cycle of the data.

In the present embodiment, the size of one block is determined by considering the ease of handling by software, the capacity of a buffer memory, the amount of data at the time of encoding, and the like.
Although described as 16 pixels, it is clear that the features and effects of the present invention are not limited to this size.

FIG. 2 exemplifies the concept of rotating an image by 90 °. The upper part shown in the figure is an image PIX (0) before rotation, and the lower part is an image PIX (9) after rotating 90 ° to the right.
0), and the direction from left to right with respect to the plane of the paper is the main scanning direction.

In the image PIX (0), the sub-scanning direction is divided every 16 lines,
Stripes, and the main scanning direction of each stripe is 16
The image is divided for each pixel and the first block to the X-th block.

Therefore, the image PIX (0) uses s and b as variables, and changes the main scanning direction from [s, 1] to [s, X] and the sub-scanning direction from [1, b] to [Y, b]. It is configured as an aggregate of blocks to be executed. [S, b] is a symbol indicating the b-th block in the s-th stripe.

The encoding order of the image PIX (0) is as follows.
First, the first stripe is processed in the following order. That is, code processing is performed from left to right of the stripe.

[1,1], [1,2], [1,3],..., [1, X] The second and subsequent stripes are similarly processed in the following order.

[2, 1], [2, 2], [2, 3],..., [2, X] [Y, 1], [Y, 2], [Y, 3],. , X] The storage memory 9 is provided with the following three areas of different code information in advance.

(1) Block code information (hereinafter abbreviated as COD) (2) Block code length information of each block (hereinafter CNT)
(3) Storage start position information of the code data of the first block of each stripe (hereinafter abbreviated as ADD) In the COD area, the code data is stored in the order of the encoded blocks.

The CNT area stores the code length data of each block in the order of the coded blocks.

In the ADD area, the address of the COD area storing the code of the first block of the stripe from 1 to Y is stored.

The image PIX (90) after the 90 ° rotation decoding processing shown in FIG. 2 can be realized by arranging the first blocks of the respective stripes at the top of the image in a form rotated 90 ° to the right. That is, a new stripe after decoding is obtained by decoding blocks in the following order.

New first stripe = [1, 1], [2,
1] to [Y, 1] new second stripe = [1, 2], [2, 2] to [Y, 1
2] New X-th stripe = [1, X], [2, X] to [Y,
X] The pixel data of each block after decoding has an arrangement of 90 °
After performing the rotation process, the image data is stored in the line buffer right-justified in the opposite direction to the main scanning direction.

As described above, the arrangement of the blocks after decoding is set as a new stripe, and the pixel arrangement is rotated by 90 ° when the decoded block pixels are stored in the line buffer, so that the entire image can be rotated by 90 °. .

FIG. 3 illustrates the concept when the image is rotated by 180 °. The upper part shown in FIG. 3 is an image PIX (0) before rotation, and the lower part is an image PIX after rotation by 180 °.
(180), and the direction from the left to the right with respect to the page is the main scanning direction.

The order of the blocks to be coded is the same as that in the case of the above-mentioned 90 ° rotation, so that the description is omitted.

The image PIX (1) after the 180 ° rotation decoding process
80) can be obtained by arranging the blocks of the last stripe in the reverse direction at the top of the image and rotating the pixel array in each block by 180 °. That is, a new stripe after decoding is obtained by decoding blocks in the following order.

New first stripe = [Y, 1], [Y2,
1] to [Y, X] new second stripe = [Y-1, 1], [Y-1, 2] to
[Y-1, X] New (Y-1) th stripe = [2, 1], [2, 2] to
[2, X] New Y-th stripe = [1, 1], [1, X] to [1,
X] The pixel data of each block after decoding has an array of 180
After performing the rotation process, the data is stored in the line buffer right-justified in the opposite direction to the main scanning direction.

As described above, the arrangement of the blocks after decoding is set as a new stripe, and the pixel arrangement is rotated by 180 ° when the decoded block pixels are stored in the line buffer, whereby the entire image is rotated by 180 °. .

FIG. 4 is a flowchart for explaining the original reading operation of the reading section 2. The read pixel data is sequentially stored in the line buffer 7a or 7b having a capacity of 16 lines according to line arbitration control by the buffer control unit 6. It is also assumed that the line buffers 7a and 7b are determined to be used recursively from 7a.

First, in step S1, a flag DOC_END_F indicating that reading of one page of a document has been completed is performed.
Is reset, and the variable rsw is initialized to “0”. still,
The variable rsw has a value of "0" or "1" and is used for selecting the line buffers 7a and 7b. Next, step S
In 2, the document to be read is transported to a predetermined reading position. Then, in step S3, it is determined whether or not the line buffer 7a or 7b to be stored selected by the variable rsw is in a state where image data can be written. Here, if the encoding process is not completed while the image data read in the past is still stored, the end of the encoding process is waited for.

If it is in a writable state, the process proceeds to the next step S4, where "16" is substituted as an initial value into a variable loop for counting the number of lines stored in the line buffer. Then, in step S5, one line is read, and pixel data is stored in the target line buffer 7a or 7b. Next, in step S6, it is determined whether the flag DOC_END_F has been reset. If it has been set, step S12
The process proceeds to step S7 if it is in the reset state. In step S7, it is determined whether a sensor installed before the reading position has detected the rear end of the read document. If the rear end has not been reached yet, the process proceeds to step S9 to prepare for reading the next line.

On the other hand, if the trailing edge is detected, step S8
, The flag DOC_END_F is set, and the remaining number of lines determined by the distance from the sensor to the reading position is substituted with N for the variable line. Then, in step S9, the variable loop is decremented by 1, and in step S10, it is determined whether the variable loop has become "0", that is, whether 16 lines have been read into the line buffer to be stored. Here, if it has not reached the 16th line, the process returns to the step S5 to read the next line. If 16 lines have been read, the process proceeds to step S11, inverts the variable rsw for selecting a storage target, and returns to step S3.

In step S6, the flag DO is set.
If C_END_F has been set, step S1
Proceed to 2 and subtract 1 from the variable line. In step S13, it is determined whether the variable line has reached "0", that is, whether the original has been read to the end. Here, "0"
If not, the process proceeds to step S9, and the above-described reading process is repeated. If it has reached “0”, step S14
To reset the flag DOC_END_F. Then, the line buffer storing the last line read in step S15 forms one stripe, that is, 16 lines.
If the number of lines is less than the number of lines, an end process for storing all white lines in the remaining line buffer area is performed, and the buffer controller 6 is notified of the end of reading one page of the document. Next, in step S16,
The process waits until the encoding process of the document is completed, and when it is completed, it is determined in step S17 whether there is a document to be continuously read. If there is a document, the process returns to step S1 to prepare for reading the next document. If not, the reading operation ends.

FIG. 5 is a flowchart illustrating the encoding operation of the encoding unit 3.

The flags, variables,
The constant has the following meaning. LAST_F is a flag indicating that the encoding process is for the last stripe. s
w is a flag for switching the line buffers 7a and 7b used for encoding, and has a value of "0" or "1". B
UF [sw] represents a line buffer to be encoded,
BUF [0] indicates the line buffer 7a, and BUF [1] indicates the line buffer 7b. s is a variable representing a stripe number. b is a variable representing a block number. p is a pointer variable having an address value for storing the code data in the storage memory 9. c is a counter variable for counting the code data length of the block. ADD [s] is an address value at which the code of the first block of the S-th stripe is stored in the storage memory 9. CNT
[S, b] is the code data length of the B-th block of the S-th stripe. X is the last block number of the stripe uniquely determined by the main scanning length of the document. Pnext is an address value of the storage memory 9 to be assigned to a pointer variable in order to change a code storage destination to another section of the storage memory 9.

First, in step S21, the flag L
AST_F is reset, the line buffer to be encoded is set to 7a (flag sw = 0), the storage start address value is set in the pointer variable p, and s = 0, b = 1, and c = 0 are initialized. Next, in step S22, it is determined whether 16 read lines are stored in the line buffer to be encoded and encoding can be started. Here, if the process can be started, the process proceeds to step S26; otherwise, the process proceeds to step S23. In step S23, it is determined whether the reading operation has already been completed, and if not, the process returns to step S22.

If the reading operation has been completed, the flow advances to step S24 to determine whether data to be encoded is stored in the encoding target line buffer. Here, if there is no encoding process, the encoding process ends. If there is, the process proceeds to step S25, and the flag LAST_F is set. Next, in step S26, the stripe number s
, And the block number b is initialized to “1”. Then, in step S27, the value of the pointer p is set to A as the start address for storing the code of the stripe.
DD [s].

Next, in step S28, the block [s, b] is encoded, the code is sequentially stored at the address indicated by the pointer p, and the code length c is counted. Next, in step S29, the block code length CNT
The counted value of the code length c of the block is stored in [s, b]. Then, in step S30, the block number b
, And it is determined in step S31 whether encoding up to the last block of the stripe has been completed. If the last block has not been reached yet, the process returns to step S28, and the above-described encoding process is repeated.

If the encoding has been completed up to the last block, the flow advances to step S32 to determine whether the encoding of the last stripe has been completed. If the encoding has been completed, the encoding process is terminated. If not, the process proceeds to step S33, in which the flag sw is inverted to change the line buffer to be encoded, and the process returns to step S22.

Here, the processing for storing the code data in the storage memory 9 in the block coding processing in step S28 will be described in further detail.

FIG. 6 is a flowchart showing the storing process. In general, the storage area of the storage memory 9 is divided and managed for each predetermined capacity in order to improve the use efficiency of the memory. Therefore, when the amount of code data to be stored exceeds a predetermined capacity, it is necessary to continue the storing process for discontinuous addresses. Which storage section should be updated can be determined by issuing an interrupt request to the system control unit 1 which manages the storage memory 9 and issuing the address value Pnex of the new storage section.
Request exception processing that returns the value of t.

First, in step S41, the code data is stored at the address indicated by the pointer p. Then, in step S42, it is determined whether or not the value of the pointer p corresponds to the boundary of a predetermined storage section. Here, if it is not a boundary, the process proceeds to step S45, 1 is added to the pointer p,
finish. If it is the boundary, the process proceeds to step S43, where an interrupt request is issued to the system control unit 1, and an exception process is requested to input an address value Pnext to start storage next. Then, in step S44, Pnext is substituted for the pointer p.

FIG. 7 shows three types of code information stored in the storage memory 9 as a result of the above-described coding process. As shown in the figure, ADD [s] decomposes an image of one page into stripes of 16 lines, and encodes the first block of code data COD [s,
1] shows fixed-length address information storing an address value storing “1”. COD [s, b] indicates indefinite length code data obtained by sequentially coding each block in the following order.

The first block [1, 1] of the first stripe
To the X-th block [1, X], the first block [2,1] of the second stripe to the X-th block [2, X], the first block [Y, 1] of the Y-th stripe to the X-th block [Y ,
X] and CNT [s, b] have the same block order as COD [s, b], and indicate fixed-length code length data indicating the code length of each block.

FIG. 8 is a flowchart showing a decoding process for restoring an image in the same direction as the original image using the code information shown in FIG. First, in step S101, the buffer selection sw = 0 and the stripe counter s =
1. Initialize each variable to block counter b = 1. Then, the process proceeds to step S102, where it is determined whether or not the selected buffer is capable of storing the restored image. If not, the process waits. When the buffer is ready, the process proceeds to step S103, and the block [s,
The code length information of [b] is read from CNT [s, b] and substituted into a variable c. Then, in step S104,
The address information of the storage memory 9 in which the code of the stripe s to be decoded is stored is read from ADD [s] and substituted for the variable p.

Next, in step S105, the contents of ADD [s] are updated to the value of p + c in preparation for decoding of the next block. Then, in step S106, the block [s,
b] to restore an image of 16 × 16 pixels. Upon completion of the decoding process, the process proceeds to step S107, where the restored image is stored in the line buffer BUF [sw], left-justified in the same manner as in the main scanning direction, and the subsequent step S1
At 08, 1 is added to the block number b. Then, in step S109, it is determined whether decoding of the last block of the stripe has been completed by comparing the block number b per stripe with the block number b uniquely determined according to the predetermined image size. Here, if not, the process returns to step S103, and the above-described decoding process is repeated.

On the other hand, if the processing is completed in step S109, the flow advances to step S110 to invert the contents of the buffer selection variable sw to change the line buffer to be used in the future.
Then, in step S111, 1 is added to the stripe number s. In step S112, the number of stripes Y at the time of encoding is calculated.
If the last stripe has not been decoded yet, the process returns to step S102, and the above-described stripe decoding process is executed again. Upon completion, the process advances to step S113 to notify the buffer controller 6 that the decoding process for one page has been completed.

FIG. 9 is a flowchart showing a decoding process for restoring an image obtained by rotating the original image by 90 ° to the right similarly to FIG. First, in step S201, the buffer selection sw = 0 and the stripe counter s =
1. Initialize each variable to block counter b = 1. Then, the process proceeds to step S202, and it is determined whether the selected buffer is in a state where the restored image can be stored. If not, the process waits. When the buffer is ready, the process proceeds to step S203, and the block [s,
The code length information of [b] is read from CNT [s, b] and substituted into a variable c. Then, in step S204,
The address information of the storage memory 9 in which the code of the stripe s to be decoded is stored is read from ADD [s] and substituted for the variable p.

Next, in step S205, the contents of ADD [s] are updated to the value of p + c in preparation for decoding of the next block. Then, in step S206, the block [s,
b] to restore an image of 16 × 16 pixels. Next, in step S207, the restored 1
The image of 6 pixels × 16 pixels is rotated 90 ° clockwise. Then, in step S208, the image resulting from the restoration rotation is stored in the line buffer BUF [sw], right-justified in the opposite direction to the main scanning direction, and 1 is added to the stripe number s in step S209. Next, in step S210, the number of stripes Y at the time of encoding is compared with the stripe number s, and if the final stripe has not been decoded yet, step S210 is executed.
Returning to step 203, the decoding processing of the same-numbered block of the next stripe is executed.

On the other hand, when the processing ends in step S209, the flow advances to step S211 to invert the contents of the buffer selection variable sw and change the line buffer to be used in the future. Then, in step S212, 1 is added to the block number b, and in step S213, the number of blocks per stripe X uniquely determined according to a predetermined image size is compared with the block number b, thereby obtaining the last block of the stripe. It is determined whether the decryption has been completed. Here, if not, the process returns to step S202, and the above-described stripe decoding process is repeated. If the decoding has been completed, the process proceeds to step S214, and the buffer controller 6 is notified that the decoding process for one page has been completed.

FIG. 10 is a flowchart showing a decoding process for restoring an image obtained by rotating the original image by 180 ° using the code information shown in FIG. First, step S
At 101, each variable is initialized to a buffer selection sw = 0, a stripe counter s = Y, and a block counter b = 1. “Y” is the number of the last stripe. Then, the process proceeds to step S302, where it is determined whether the selected buffer is in a state where the restored image can be stored. If not, the process waits. When the buffer is ready, the process proceeds to step S303, and the block [s,
The code length information of [b] is read from CNT [s, b] and substituted into a variable c. Then, in step S304,
The address information of the storage memory 9 in which the code of the stripe s to be decoded is stored is read from ADD [s] and substituted for the variable p.

Next, in step S305, the contents of ADD [s] are updated to the value of p + c in preparation for decoding of the next block. Then, in step S306, the block [s,
b] to restore an image of 16 × 16 pixels. Upon completion of the decoding processing, the flow advances to step S307 to store the restored image in the line buffer BUF [sw], right-justified in the opposite direction to the main scanning direction, and then to step S3.
At 08, 1 is added to the block number b. Then, in step S309, it is determined whether decoding of the last block of the stripe is completed by comparing the block number b per stripe with the block number b uniquely determined according to the predetermined image size. Here, if not, the process returns to step S303, and the above-described decoding process is repeated.

On the other hand, if the processing is ended in step S309, the flow advances to step S310 to change the line buffer used next time by inverting the contents of the buffer selection variable sw.
Then, 1 is subtracted from the stripe number s in step S311, and it is determined whether the stripe number is "0" in step S312. Here, if all stripes have not been decoded yet, the process returns to step S302, and the above-described stripe decoding process is executed again. When the processing is completed, step S31 is performed.
Proceeding to step 3, the buffer control unit 6 is notified that the decoding process for one page has been completed.

FIG. 11 is a flowchart showing the operation of the printing unit 5 shown in FIG. 1 for performing processing for printing one page of the image stored in the line buffers 7a and 7b. First, in step S401, the selection flag psw of the line buffer to be used is initialized to “0”. In addition, psw is
Wait for a value of “0” or “1”. Then, step S40
Proceed to 2 to transport the recording paper to the printing start position. Next, in step S403, the line buffer BU to be printed is
It is determined whether image data to be printed is stored in F [psw]. If the image data has been prepared, the process proceeds to step S404, and the line buffer BUF
The image data is read from [psw] and printing of 16 lines is executed. Then, in step S405, the value of the selection flag psw of the line buffer to be used is inverted.

On the other hand, if it is determined in step S403 that the image data has not been prepared, the flow advances to step S406 to determine whether the printing operation of one page is completed. If not, the process returns to step S403. If the process is completed, the printed recording paper is discharged, and the printing operation for one page is completed.

FIG. 12 is a diagram showing a configuration of a circuit for rotating pixel data. In the present embodiment, one block is 16 pixels × 16 pixels, but here, for simplification of description, 2 pixels × 2 pixels are selectively set to 0 ° / 90 ° / 1.
An example in which the output is rotated at any angle of 80 ° / 270 ° will be described. It would be easy to guess to extend this to 16 × 16 pixels.

First, as input signals, Di1 and Di0 are data input signals of 2-bit pixel data and data for selecting a rotation angle, nCS is a negative logic select signal for validating the data input / output operation of this circuit, and nWR. Is a negative logic write signal for writing the input data Di1, Di0,
nRD = negative logic read signal for reading output data Do1 and Do0, and A1 and A0 are address signals for selecting data input / output targets. Do1 and Do0 as output signals are output signals of pixel data as a result of the rotation processing.

The circuit block 101 is 2 to
4 has the function of the truth table shown in FIG. Reference numerals 102 to 104 denote 2-bit D flip-flops, which have input terminals D at the rising edge of the clock CK.
The state of 1 / D0 is latched and output to output terminals Q1 / Q0. Reference numeral 105 denotes a selector having the function of the truth table shown in FIG. 14, and four types of four-bit data input terminals D * 0 / D * 1 according to the states of the selection input terminals S1 / S0.
/ D * 2 / D * 3 is selected and output to the output terminal Q *.

Reference numeral 106 denotes a selector having the function of the truth table shown in FIG. 15, and two types of two-bit data input terminals D * 0 / D * according to the state of the selection input terminal SEL.
1 is selected and output to the output terminal Y *. 107
Is a 3-input NOR gate. And 108a, 1
08b is 3 of “1”, “0”, and “high impedance”.
An output buffer with two output states.

Next, the operation of the above-described rotation processing circuit will be described. FIG. 16 is a diagram for explaining the operation of this circuit. First, input pixels pix00, pix01, pix
10 and pix11 are D flip-flops 10
The positional relationship is changed in accordance with the rotation angle selection data (S1, S0) latched in No. 4 and read out from the output terminals Do1 / Do0. Hereinafter, the operation will be described by taking a case of performing a 90 ° rotation as an example. (1) nWR with nCS = 0, A1 = 0, A0 = 0
, A D1 flip-flop 102 latches the state of Di1 / Di0 as the first two pixels pix01 and pix00. (2) nWR with nCS = 0, A1 = 0 and A0 = 1
, A D1 flip-flop 103 latches the state of Di1 / Di0 as the next two pixels pix11 and pix10. (3) nWR with nCS = 0, A1 = 1, A0 = 1
By applying a negative pulse to the D flip-flop 104 as the rotation angle selection data.
Latched. Here, since it is a case of 90 ° rotation, D
“01” is set in the flip-flop 104. (4) Since the output of the D flip-flop 104 is connected to the selection input S1 / S0 of the selector 105, 4-bit output data is output from the selector 105 according to the truth table shown in FIG.

Qd = Dd1 = pix10, Qc = Dc1
= Pix00, Qb = Db1 = pix11, Qa = Da
1 = pix01, (5) nRD with nCS = 0, A1 = 0, A0 = 0
, A negative pulse is applied to two output terminals of the selector 106 according to the truth table shown in FIG.
/ Da0 side, that is, pix11 / pix01 is output and read out to the external output terminals Do1 / Do0 via the output buffers 108b / 108a. (6) nRD with nCS = 0, A1 = 0 and A0 = 1
, A negative pulse is applied to two terminals of the selector 106 according to the truth table shown in FIG.
a1, that is, pix10 / pix00 is output, and the external output terminal D is output via the output buffers 108b / 108a.
o1 / Do0.

As a result of the above, pi of 2 × 2 pixels
x00, pix10, and pix11 are converted into the positional relationship shown in FIG.

[Other Embodiments] In the above embodiment,
Although the method of realizing the right 90 ° rotation and the 180 ° rotation has been described, as another embodiment, a method of realizing the case where the encoded image is rotated to the left 90 ° will be described.

FIG. 18 is a diagram showing the concept of rotating an image by 90 degrees to the left. The upper part shown in the figure is the image PIX (0) before rotation, and the lower part is the image PIX (2) after rotating 90 °
70), and the main scanning direction is from left to right with respect to the page.

As shown, in PIX (270),
The block number X must be arranged on the first stripe. However, as described in the above-described embodiment, the stripe address information ADD [s] is the address information of the first block.
Processing slightly different from 0 ° rotation is required.

First, as the stripe address information, only the last stripe address information ADD [Y] is provided in the above-described embodiment, but in addition to this, the block [Y, X] is encoded and stored. The next address information after the last address is required. This is called ADD [Y + 1]. On the other hand, the information of ADD [1] becomes unnecessary. As a basic decoding procedure, CNT [s, b] is subtracted from ADD [s + 1] to decode block [s, b], and the result is used as address pointer p and stored in ADD [s + 1]. .

FIG. 19 is a flow chart showing the decoding processing of the 90-degree left rotation. First, in step S501, the buffer selection sw = 0 and the stripe counter s =
1. Initialize each variable to a block counter b = X. Then, the process proceeds to step S502, where it is determined whether the selected buffer is in a state capable of storing the restored image. If not, the process waits. When the buffer is ready, the process proceeds to step S503, and the block [s,
The code length information of [b] is read from CNT [s, b] and substituted into a variable c. Then, in step S504,
The variable c is subtracted from the address information ADD [s + 1] of the stripe next to the stripe s to be decoded, and is substituted for the variable p.

Next, in step S505, the value of p is stored in ADD [s + 1] in preparation for decoding of the next block. Then, in step S506, the decoding process of the block [s, b] is performed to restore an image of 16 pixels × 16 pixels. Next, in step S507, the restored 16-pixel × 16-pixel image is rotated 90 ° to the left.
Then, in step S508, the restored rotation result image is stored left-justified in the line buffer BUF [sw] in the same manner as in the main scanning direction, and the stripe number s is incremented by 1 in step S509. Next, in step S510, the number of stripes Y at the time of encoding is compared with the stripe number s, and if the final stripe has not been decoded yet, step S50.
Then, the process returns to step S3, and the decoding processing of the same numbered block of the next stripe is executed.

On the other hand, when the processing ends in step S510, the flow advances to step S511 to invert the contents of the buffer selection variable sw to change the line buffer to be used in the future. Then, in step S512, 1 is subtracted from the block number b,
In step S513, it is determined whether the block number b has reached “0”. Here, if not, the process returns to step S502. On the other hand, if the processing has been completed, the process proceeds to step S514,
The completion of the decoding process for one page is notified by the buffer controller 6.
Notify.

In the embodiment, although no particular mention is made of the coding method of each block, coding methods such as MH, MR, and MMR generally used in facsimile apparatuses can be used. However, the compression ratio is deteriorated as compared with the normal line-by-line coding. From the viewpoint of the compression ratio, the “arithmetic code” defined in ITU-T recommendation T82 is particularly excellent. This method is easy to handle because the code unit is bytes.

It is apparent that the same effect can be obtained regardless of which encoding method is used, with respect to "reducing the memory capacity required for rotation processing", which is the main feature of the present invention.

In encoding, a necessary condition for implementing the present invention is that the independence of each block must be ensured. That is, when the target pixel is encoded with reference to the surrounding pixels, if the reference pixel does not exist in the block, the virtual pixel is not set without referring to the actual pixel of the adjacent block. No. In block coding, a block in which all pixels are “white” is represented by code length information = “0” and code information = none.

As a result, it is possible to improve the coding compression ratio of a white background portion having a high probability of occurrence in a document image, and to increase the processing speed of coding and decoding.

The present invention can be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, and the like), and can be applied to a single device (for example, a copier, a facsimile). Device).

An object of the present invention is to supply a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (CPU or MP) of the system or apparatus.
It goes without saying that U) can also be achieved by reading and executing the program code stored in the storage medium.

In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, based on the instructions of the program code, It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

[0107]

As described above, according to the present invention,
The capacity of a memory for storing an image required when rotating an image as pixel data is at least one stripe.
Even if the improvement of the throughput is taken into consideration, it is possible to reduce the number of stripes to at least two stripes, and it is possible to greatly reduce the apparatus cost.

FIG. 20 is a comparison of the memory capacity required between the conventional bitmap system and the block system of the present invention when the A4 image is rotated by 90 °. Thus, the higher the resolution of the target image, the greater the cost effectiveness of this memory reduction.

In the block method of the present invention, the block size is 16 pixels × 16 pixels, and the stripe memory is A3.
It is assumed that there is a width of 16 lines × 2.

[0110]

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a multifunction peripheral according to an embodiment.

FIG. 2 is a diagram illustrating a positional relationship between an image stripe and a block before and after a 90 ° rotation process;

FIG. 3 is a diagram illustrating a positional relationship between an image stripe and a block before and after a 180 ° rotation process;

FIG. 4 is a flowchart illustrating an original reading operation of the reading unit 2.

FIG. 5 is a flowchart illustrating an encoding operation of the encoding unit 3.

FIG. 6 is a flowchart illustrating a code storing process.

FIG. 7 is a conceptual diagram illustrating three types of code information stored in a storage memory 9;

FIG. 8 is a flowchart of the decoding unit 4 for explaining an operation of decoding a code image stored in the storage memory 9 in the original image direction.

FIG. 9 is a flowchart of the decoding unit 4 for explaining an operation of restoring a code image stored in the storage memory 9 into an image rotated by 90 °.

FIG. 10 shows a case where the code image stored in the storage memory 9 is 180.
35 is a flowchart of the operation of the decoding unit 4 for describing an operation of restoring a rotated image.

FIG. 11 is an operation flowchart of the printing unit 5 for explaining an operation of printing an image.

FIG. 12 is a diagram showing a configuration of a circuit for selectively rotating a matrix image of 2 pixels × 2 pixels by 0 °, 90 °, 180 °, and 27 ° and outputting the result;

13 is a diagram showing a truth table of the decoder 101 in FIG.

FIG. 14 is a diagram showing a truth table of a selector 105 in FIG. 12;

FIG. 15 is a diagram showing a truth table of a selector 106 in FIG. 12;

FIG. 16 is a diagram for explaining the operation of the circuit shown in FIG.

FIG. 17 is a diagram showing a state in which a matrix of 2 pixels × 2 pixels is rotated by 90 °.

FIG. 18 is a diagram illustrating a positional relationship between an image stripe and a block before and after a left 90 ° rotation process;

FIG. 19 shows a code image stored in the storage memory 9 at 90 left
35 is a flowchart of the operation of the decoding unit 4 for describing an operation of restoring a rotated image.

FIG. 20 is a diagram comparing the memory capacities required for the conventional bitmap method and the block method according to the present invention when the A4 image is rotated by 90 °.

FIG. 21 is a diagram illustrating a main scanning direction of reading / recording of a document in a copying machine.

Claims (6)

[Claims] A reading unit that reads a document in line units; a storage unit that can store at least m lines of pixel data read by the reading unit; and an encoding unit that encodes the pixel data stored in the storage unit. Means, and accumulating means capable of accumulating code data encoded by the encoding means for a plurality of pages, wherein the encoding means stores the m-line pixel data stored in the storage means in the main scanning direction. An image processing apparatus which divides an image into blocks of n pixels and performs encoding on a block basis. 2. The encoding means as an encoding result,
Code information obtained by compressing and encoding the image of each block, code length information indicating the amount of the code information, and address information indicating the position where the code information of the first block is stored in the storage means are generated. The image processing apparatus according to claim 1. 3. A decoding means for decoding the code data stored in the storage means in units of blocks, and a pixel obtained by rotating the pixel data of m pixels × n pixels reconstructed by the decoding means in units of 90 °. And a block rotation processing unit for converting the data into an array, wherein the decoding unit decodes each block in a predetermined order different from the order of encoding by the encoding unit according to a designated rotation angle of the restored image. The block rotation processing unit also performs a conversion process into a pixel array corresponding to the rotation angle of the specified restored image, sequentially stores the converted image in the storage unit from a predetermined direction corresponding to the rotation direction, and stores n or m lines. 2. The image processing apparatus according to claim 1, wherein the pixel data is sequentially created. 4. A printing means for inputting pixel data of n or m lines sequentially created and printing it on a recording medium, wherein an image for one page rotated by a predetermined angle is printed. The image processing apparatus according to claim 3, wherein: 5. A second encoding unit which receives sequentially created n or m lines of pixel data and performs predetermined encoding in line order according to a transmission destination, and a second encoding unit. The image processing apparatus according to claim 3, further comprising a communication unit that transmits the encoded data, wherein the image processing unit transmits an image of one page rotated by a predetermined angle. 6. A document is read line by line, read pixel data is stored in at least m lines of storage means, and the stored m lines of pixel data are divided into blocks of n pixels in the main scanning direction. Encode each block, decode the coded data in block units, rotate the restored m-pixel x n-pixel data in units of 90 °, and specify the encoding order according to the specified rotation angle of the restored image. Decode each block in a predetermined order different from the above, perform a conversion process to a pixel array according to the rotation angle of the specified restored image, sequentially store from a predetermined direction according to the rotation direction,
An image processing method characterized by sequentially creating pixel data of n or m lines.