JP2001043359A - Image processor and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain the fast rotation, enlargement and reduction and clipping of an image by means of a memory of small capacity with no use of a page memory and with no redundant reading operations required. SOLUTION: A band data storing means 40 reads in sequence the input images, and the local data stored in the means 40 are sequentially read out to undergo the conversion processing such as the rotation and enlargement/ reduction via a local data conversion part 50 and then stored in an intermediate buffer of a converted data storing means 60. Then the local data are read out of the means 60 in their outputting order, stored temporarily in a band buffer of a band data storing means 70 and outputted via an image output means 80.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing technique, and more particularly to an inexpensive and high-speed rotation, enlargement / reduction, and clipping of an arbitrary angle.

[0002]

2. Description of the Related Art Generally, image processing can be broadly divided into geometric processing for changing the position of a pixel and pixel value processing for changing a pixel value. Specific examples of the former include affine transformation such as image rotation, enlargement, and reduction, and clipping for clipping pixels. Specific examples of the latter include color correction and filter processing.

[0003] Among these image processes, the rotation process is different from other processes. That is, in general image processing, image data is processed in the raster scanning order, and the result is also output in the raster scanning order. On the other hand, in the rotation processing, the input / output order of data differs between the input, that is, the source side and the output, that is, the destination side.

Specifically, when data is input from the source side in the raster scanning order, the rotated data is written diagonally on the destination side. Therefore,
On the destination side, random access to image data is required. This is hereinafter referred to as a forward conversion method. Conversely, if the image data is transferred on the destination side in the raster scanning order, random access is required on the source side. This method is hereinafter referred to as an inverse conversion method.

[0005] In either system, random access is required on either the source side or the destination side.
The prior art uses a page memory to make this possible. That is, in the case of the forward conversion method, a rotated image is generated on the page memory on the destination side, and in the case of the inverse conversion method, the entire image is written in the page memory and then the rotation processing is performed. On the premise of this method, a method for reducing the transfer time of image data and the coordinate calculation time has been proposed.

As a conventional example, a technique described in Japanese Patent Laid-Open No. 61-161576, which is one of such systems, will be described. In the conventional example, the image is processed locally, that is, in units of blocks, thereby speeding up the rotation process. Hereinafter, a conventional example will be described with reference to FIG. In FIG. 17, words and the like are changed so as to match the description of the present invention, but there is essentially no change from the description in the publication of the conventional example. In FIG. 17, reference numeral 10 denotes an image input unit;
9 is an inverse transformation coordinate calculation unit, 48 is a page memory read unit, 49 is a source page memory, 52 is a conversion ROM,
51 is a local data conversion unit, 78 is a page memory writing unit, 79 is a destination page memory, 80 is an image output unit, 200 is input image data, 228 is source address information, 229 is inverse transformation coordinates, and 230 is local Data 231 is conversion information of the conversion ROM, 240 is converted local data, 269 is destination-side address information, and 270 is output image data.

Next, each part will be described. Image input unit 1
0 indicates that the input image data 200 is stored in the source page memory 49.
To store. The inverse-transformed coordinate calculating unit 29 calculates the coordinates on the source side obtained by inversely converting the center pixel of the local data on the destination side, and outputs the coordinates as the inverse-transformed coordinates 229. The page memory read unit 48 has the inverse transformed coordinates 229.
Is transmitted to the source-side page memory 49. The source side page memory 49 stores the input image data 200 and sends out the local data 230 to the local data converter 61 according to the source side address 228. Local data converter 51 converts RO
Based on the conversion information 231 of M52, conversion processing such as rotation, enlargement, and reduction is performed on the read local data 230,
The converted local data 240 is output to the destination page memory 79. Page memory writing unit 78
Stores the destination-side address 269 corresponding to the inverse transformed coordinate 229 in the destination-side page memory 79.
Send to The destination-side page memory 79 stores the converted local data 240, and sends out the output image data 270 to the image output unit 80 according to the destination-side address 269. The image output unit 80 outputs the output image data 270 to the outside.

In the above configuration, it is assumed that the local data 230 is, for example, block data of 3 × 3 pixels.

The operation of processing the rotation of an image based on the above configuration by dividing the rotation into a local rotation and a global rotation will be described with reference to FIG. In FIG. 18, 400 is local data before rotation,
Reference numeral 410 denotes a center pixel before rotation, 420 denotes a center pixel of rotation, 43
0 is the local data after the ideal rotation, 440 is the central pixel after the rotation,
450 is local data after rotation.

When the pre-rotation local data 400 is rotated around the rotation center pixel 420, the ideal post-rotation local data 43
It becomes 0. However, since pixels must be arranged in the horizontal and vertical directions, the local data 45 after rotation is actually used.
The purpose of the rotation process is to obtain 0 and its coordinates. Now, in the first embodiment, the rotated local data 4
50 is obtained by performing local rotation processing on the post-rotation center pixel 440 on the ideal post-rotation local data 430. Since this processing is constant regardless of the local data, if the rotation destination in the local data is tabulated and stored in the ROM according to the rotation angle α, it is not necessary to recalculate the coordinates for each local data. The post-rotation center pixel 440 needs to be calculated from the pre-rotation center pixel 410, the rotation center pixel 420, and the rotation angle α, but this calculation may be performed for each local data.

As described above, the rotation processing in the conventional example is characterized in that it is performed by two processings: a local processing based on the conversion ROM 52 and a global processing for obtaining the address of the local data of the rotation destination. According to the conventional example, high-speed rotation processing is possible. That is, since the former local processing is simplified, it can be speeded up by hardware or the like, and the wide area processing can reduce the number of operations itself.

However, the conventional example is premised on having two page memories, and since the page memories are generally expensive, there is a problem that the cost of the entire apparatus increases.

[0013]

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and uses a small-capacity memory without having a page memory to rotate and scale these images at high speed and at low cost. The purpose is to enable clipping.

[0014]

According to the present invention, in order to achieve the above object, a configuration as described in the claims is adopted.

According to the present invention, the second rectangular block is set so as to include the first rectangular block tilted according to the rotation angle, and the input image is sequentially read out in units of the second rectangular block, and read out. Since the first rectangular block subjected to the rotation processing according to the rotation angle is obtained from the second rectangular block, the rotation of the input image before rotation is performed without using a page memory when performing rotation at an arbitrary rotation angle. Scanning and reading from the output image after rotation can be performed only in the raster scan direction.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image processing apparatus according to the present invention will be described in detail.

FIG. 1 is a block diagram for explaining the basic concept of the present invention. In the figure, 10 is an image input unit, 20 is an address control unit, 40 is a band data storage unit, 50 is a local data conversion unit, 60 is a converted data storage unit, 70 is a band data storage unit, 80 is an image output unit, 200 Is input image data, 220 is address information on the source side, 230
Is local data, 240 is transformed local data, 250
Is address information on the destination side, 260 is converted local data, and 270 is output image data.

The configuration of FIG. 1 will be described in detail. The image input means 10 inputs image data 2 one line at a time in raster scanning order.
00 to the band data storage means 40. The address control means 20 sends the source-side address information 220 to the band data storage means 40 and sends the destination-side address information 250 to the converted data storage means 60. The band data storage means 40 has a band buffer which is large enough to store local data and which always discards the oldest one-line data when one-line image data is input in raster scanning order. (Hereinafter referred to as band data) in the band buffer, and the local data 230 in the band buffer indicated by the address information 220 on the source side.
Is sent. The local data conversion unit 50 has a conversion function such as rotation and enlargement / reduction, and sends out the converted local data 240 of the local data 230 of the image. The converted data storage means 60 has a random accessible intermediate buffer capable of temporarily storing the converted local data 240, and sends out the local data 260 indicated by the address information 250 on the destination side. The band data storage means 70 has a band buffer large enough to store local data,
After storing the converted local data 260 in the band buffer, the output image data 270 is sent to the image output means 80. The image output means 80 receives the output image data 270 for each band, and sends it out to another storage device or outside.

The operation of the basic concept based on the above configuration will be described. FIG. 2 is a flowchart for explaining the operation of the basic concept. Hereinafter, description will be made with reference to FIG.

In S10, one line of the input image is read out to the band buffer. In S20, the band data storage means 4
If the local data indicated by the address information on the source side exists in the band buffer held by 0, the process proceeds to S30. If the local data does not exist in the band buffer, one line of the input image is read again in S10. In S30, the local data in the band buffer indicated by the address information on the source side is sent to the local data conversion unit 50. In S40, the transmitted local data is subjected to conversion processing such as rotation and enlargement / reduction, and is stored in the intermediate buffer held by the converted data storage unit 60. S
In 50, if all the necessary processing has not been performed on the input image, the process proceeds to S10, and if it has been performed, the process proceeds to S60. In S60, the converted local data stored in the intermediate buffer is sent to the band buffer held by the band data storage unit 70 in accordance with the address information on the destination side. In S70, the local data in the band buffer is sent to the image output means 80 according to the address information on the destination side. In S80, if all necessary processes have not been performed on the input image, the process proceeds to S60, and if performed, the process ends.

In the band data storage means 40 for reading local data from an input image by the above means,
There is no need to read the input image many times, and by reading the input image only once in the raster scanning order, the input image can be rotated in local data units obtained by locally dividing the input image. And then store it again on the band buffer (band data storage means 70), thereby performing a wide-area rotation (parallel movement) of the local data and obtaining a converted output image such as rotation and enlargement / reduction processing. At the same time, a page memory is not required, and these conversion processes can be realized with only a small number of buffers.

[0022]

Embodiments of the present invention will be described below. [First Embodiment] A method of performing a rotation process using a band buffer will be described as a first embodiment of the present invention. In addition,
In the embodiment described below, the rotation processing is performed by the inverse conversion method. When the inverse conversion method is adopted, the rotation processing can be performed with reference to the pixel position of the output pixel, so that more accurate rotation processing can be performed. Of course, a forward conversion method can also be adopted.

FIG. 3 is a block diagram for explaining the first embodiment of the present invention. In the figure, 10 is an image input unit, 21 is an address list generation unit, 30 is an address list sort unit, 4
1 is a read band buffer, 42 is a band buffer read control unit, 51 is a local data conversion unit, 52 is a conversion ROM, 61 is an intermediate buffer, 62 is an intermediate buffer control unit, 71 is a write band buffer, and 80 is an image output. , Reference numeral 200 denotes input image data, 210 denotes a source-side inverted address list, 220 denotes a sorted source-side address list, 221 denotes a read band buffer control signal, 230 denotes local data, and 231 denotes local data. Transformed coordinates, 240 is transformed local data, 250 is a destination address list, 251 is an intermediate buffer control signal, 260 is transformed local data,
70 is output image data.

The configuration of FIG. 3 will be described. The image input unit 10 sequentially sends the input image data 200 in a raster scanning order.
The address list generation unit 21 sends out the destination address list 250 and the source address list 210 obtained by inversely converting the address list 250. The address list sorting unit 30 has a sorting function for the y-coordinate and sends out the address list 220 on the source side. The band buffer read control unit 42 has a function of controlling the read band buffer 41 based on the source-side address list, and sends out a control signal. The local data conversion unit 51 sends out the local data 230 and the local data 240 converted based on the converted coordinates 231 of the local data. The intermediate buffer control unit 62 sends an intermediate buffer control signal 251 based on the destination address list 250. The intermediate buffer 61 has a function of storing the converted local data, and sends out the local data 260 as needed. The write band buffer 71 has a function of storing local data, and sequentially sends out buffer data. The image output unit 80 sends out an output image.

The operation of this embodiment will be described based on the above configuration. FIGS. 4, 5, and 8 are flowcharts for explaining the operation of this embodiment. The operation of the present embodiment will be described below with reference to FIGS.

Referring to FIG. 4, the operation of the present embodiment will be described as a whole. FIG. 4 shows that the operation of this embodiment can be roughly divided into two. In S100, the input image is divided and divided into local data and converted, and the conversion result is stored in the intermediate buffer 61 in the conversion order. In S200, the intermediate buffer 6
The local data stored in the order of the addresses sorted in 1 is rotated (translated) in a wide area, and an output image is transmitted. The detailed operations of S100 and S200 correspond to FIGS. 5 and 8, respectively.

The operation of S100 (FIG. 4) of this embodiment will be described with reference to FIG. In S110, an address list on the destination side and an address list on the source side obtained by inversely converting the address list are generated. In S120, the address list on the source side is sorted for the y coordinate. In S130, an address pointer for specifying the address of the sorted source-side address list is initialized. In S140, the image data of one line of the input image is read out to the readout band buffer 41. S150
Then, it is determined whether or not the data of the n × n pixel block serving as the local data centering on the address indicated by the address pointer exists in the band buffer 41, and if all the pixels of the local data do not exist, it is input again in S140. One line of the image is read out to the band buffer 41. S160
In S150, when all the pixels of the local data exist in S150, the data of the n × n pixel block centering on this address is transmitted. In S170, the local data is converted based on the previously obtained coordinate conversion table and stored in the intermediate buffer 61. In S180, the address pointer on the address list is advanced by one, and the process preparation for the next address is performed. In S190, it is determined whether or not there is another address in the address list sorted on the source side.
If it exists, the local buffer indicated by the address may exist in the band buffer 41 read in S140 in S150, so that the process jumps to S150. If no address remains in the address list, the entire input image is displayed. Is locally converted and stored in the intermediate buffer 61.

The above-mentioned band buffer 4 for reading during operation
1 will be described. FIG.
5A shows the flow of data in the read band buffer 41 of FIG. The reading process when a ring buffer is used as the reading band buffer 41 is performed in the following flow.

The oldest read line data among the line data in the ring buffer is discarded. Shifting is performed sequentially from the line data read next to the discarded line data. After shifting the line data of (n-1), one new line is read from the input image to an empty line buffer.

By the above operation, local data can be extracted while reading data line by line from the input image using the ring buffer.

FIG. 6B shows the flow of data when a double buffer is used as the read band buffer 41. Generally, when a buffer is used for data transfer, a double buffer is frequently used for efficiency. The present invention can also employ a double buffer. The reading process using the double buffer is performed according to the following flow.

When all the local data in the band buffer (for example, buffer B) that has been read out of the double buffer has been read into the hardware, the data in the band buffer B is discarded. The line data of the next n lines in the input image is read into the band buffer B of the step. After reading the buffer A or the local data spanning the buffers A and B into the hardware, the line data in the buffer A is discarded, and the readout of the next input image is prepared.

By the above operation, two band buffers having a width of n pixels can be alternately read in units of n lines, so that the reading efficiency is improved.

Next, the intermediate buffer storing process (S170) during the operation of FIG. 5 will be described. FIG. 7 shows an example of an address list and a rough generation procedure in the present invention. The process of storing local data in the intermediate buffer 61 is performed in the following flow.

An address list of all blocks on the destination side is generated (processing in S110 of FIG. 5). The address of the step is inversely converted, and the source side address is generated by performing a sort process along the y-coordinate (process in S120 in FIG. 5). The local data converted in the order of the source address is stored, and the stored address is recorded in correspondence with the source address.

With the above operation, data corresponding to the destination block stored in the intermediate buffer 61 can be obtained from the address list.

Referring to FIG. 8, S200 of this embodiment (FIG. 4)
Will be described. In S201, an address pointer for specifying an address in the address list on the destination side is initialized. In S210, the address indicated by the address pointer is read from the address list on the destination side. In S220, the address pointer is advanced by one for the next processing. In S230, the local data corresponding to the read address on the destination side is transmitted from the intermediate buffer 61 and stored in the write band buffer 71. In S240, it is determined whether or not the address indicated by the address pointer is the end of the image. If not, the process proceeds to S210 for the next address. In S250, if it is the image end in S240, all the data in the write band buffer 71 is sent to the image output unit 80 in the raster scanning order. In S260, it is determined whether or not an unprocessed address exists in the address list. If the address exists, the process for the next address is performed in S210. If the address does not exist, the input image stored in the intermediate buffer 61 is deleted. The process is terminated because all have been output.

Next, the data flow of this embodiment will be described.
FIG. 9 shows a data flow (-) of this embodiment. The processing in this embodiment is performed according to the following flow.

The center coordinates of the blocks (blocks 1 to 24) blocked on the destination side are obtained in the raster scanning order, and are used as an address list on the destination side. The coordinates obtained by inversely transforming the coordinates of the address list on the destination side are sorted with respect to the y-coordinate to generate an address list on the source side. The input image is read into the band buffer 41 line by line. If there is local data centered on the coordinates indicated by the address pointer in the band buffer 41, the local data is read. The conversion is performed based on the conversion coordinates of the local data generated in advance in the conversion ROM 52, and is written to the intermediate buffer 61 in the order of conversion. When the processing is completed for the source-side address list, the local data is read from the intermediate buffer 61 to the write band buffer 71 in accordance with the destination-side address list generated in the step. If the destination is the image end from the address list, the data in the write band buffer 71 is transmitted as output image data. This is all performed for the address list on the destination side.

The conversion coordinates of the conversion ROM 52 in the steps in the above operation will be described. FIG. 10 is 8 ×
It is an example of a conversion table in the case of rotating a block of 8 pixels by 45 degrees. FIG. 10A shows the positions of the pixels after the conversion, and FIG. 10B shows the positions of the pixels before the 45-degree rotation.
The method of calculating the coordinates can be easily obtained by equation (1). The rotation destination of each pixel with respect to the center of the block before rotation is calculated.

[0041]

(Equation 1)

[Second Embodiment] As a second embodiment of the present invention, a method for performing a rotation process using a macroblock buffer will be described. FIG. 11 is a block diagram illustrating a second embodiment of the present invention. In the figure, a configuration different from that of the first embodiment will be described. Reference numeral 22 denotes a macroblock (hereinafter referred to as MB) address list generation unit, 31 denotes an MB address list sort unit, 43 denotes a read MB band buffer, 44
Is an MB buffer read control unit, and 72 is a write MB.
Band buffer 211 is the source-side inversely converted MB
Address list, 222 is the source MB sorted
An address list 223 is a read MB buffer control signal, 252 is a destination MB address list, and 253 is an intermediate buffer control signal.

The configuration of FIG. 11 different from that of the first embodiment will be described. The MB address list generation unit 22 sends out the destination-side MB address list 252 and the source-side address list 211 obtained by performing the inverse conversion process on the destination MB address list 252. The MB address list sort unit 31 has a sort function for the y-coordinate and has a source side MB.
The address list 222 is sent. The MB buffer read controller 44 has a function of controlling the read MB band buffer 43 based on the source-side MB address list, and sends out a control signal. The write MB band buffer 72 has a function of storing all MB data (hereinafter referred to as MB data), and sequentially sends out MB data.

Next, the operation of the second embodiment will be described. FIG.
2 shows a data flow of the second embodiment. Less than,
This will be described with reference to FIG. 12, but the description of the parts performing the same operations as in the first embodiment will be omitted.

The combination of several blocks blocked on the destination side is referred to as MB (MB: a to
f), the center coordinates of the MB and each block in the MB are determined in the raster scanning order, and are set as the MB address list on the destination side. The coordinates obtained by inversely transforming the coordinates of the destination-side MB address list are sorted with respect to the y-coordinate to generate a source-side MB address list. At this time, the sorting process is performed only on the center coordinates of the MB. The input image is read into the MB band buffer 43 line by line. If there is MB data centered on the coordinates indicated by the MB address pointer in the MB band buffer 43, the MB data is read for each local data unit. The conversion is performed based on the conversion coordinates of the local data generated in advance in the conversion ROM 52, and the MB data is written to the intermediate buffer 61 in the conversion order. When the processing is completed for the source side MB address list, the destination side M
The MB data is read from the intermediate buffer 61 to the write MB band buffer 72 according to the B address list. If the destination is the image end according to the address list, the data in the write MB band buffer 72 is transmitted as output image data. This is all performed for the address list on the destination side.

As described above, in this embodiment, an MB in which some blocks are collected is introduced. By preparing an MB band buffer capable of sufficiently storing MB data, the number of addresses to be sorted can be reduced to the number of MBs. That is, in the example of FIG. 12, if one block is composed of four blocks, the number of sorts is １ ／. Since the conversion of the local data is the same as in the first embodiment, the H / W for the conversion is unchanged. By introducing the MB as described above, the processing load of software can be suppressed.

To further reduce the processing load of software, a method may be adopted in which the address list is made up of only MBs and the address of each block in the MB is calculated in hardware from the offset value between blocks. The processing load of the software can also be reduced by using only the MB for the corresponding address list on the source side and recording one storage address collectively for several blocks of data stored in the intermediate buffer 61.

[Third Embodiment] As a third embodiment of the present invention, a description will be given of a method in which a compression process is also used when storing data in the intermediate buffer 61. FIG. 13 is a block diagram illustrating a third embodiment of the present invention. In the figure, a configuration different from that of the second embodiment will be described. 90 is a data compression unit, 91 is a data decompression unit, 242 is compressed local data, and 261 is compressed local data.

The configuration of FIG. 13 different from that of the second embodiment will be described. The data compression section 90 has a function of performing compression in units of local data, and the compressed local data 242
Is sent. The data decompression unit 91 has a function of decompressing a local data unit, and sends out the decompressed local data 260.

In this embodiment, block data of n × n pixels is used as local data. For this reason, JPEG (Jo
int Photographic Experts
By using block coding such as Group method, not only the size of the intermediate buffer 61 can be reduced, but also the transfer time can be reduced. The compression method of this embodiment includes, for example, DCT, DFT, D
ST, LOT, Hadamard transform, Wavelet transform,
Any of sub-band codings using any frequency conversion can be adopted.

FIG. 14 theoretically explains the present embodiment (the same applies to the fourth embodiment). That is, as shown in FIG. 14, the effect of the affine transformation is divided into local pixel unit processing (first term) and wide area partial image unit processing (second term), and both processes are independent. Can be performed. In the present embodiment, the processing (first term) in units of pixels in the partial image is executed by the local data conversion unit 51, and thereafter, this is compressed by the data compression unit 90. Then, at the time of writing to the write MB band buffer 72, a wide area process (second term) is executed in units of partial images. By doing so, the processing of the second term can be executed in a state where the image data is compressed, and the time required for data transfer and writing to the memory can be reduced.

[Fourth Embodiment] As a fourth embodiment of the present invention, a method of transmitting a compressed image of a processed image by removing the data decompression unit 91 in the third embodiment will be described. FIG. 15 is a block diagram for explaining a fourth embodiment of the present invention. In the figure, a configuration different from that of the third embodiment will be described. Reference numeral 81 denotes a compressed image output unit, and 271 denotes compressed local data.

The configuration of FIG. 15 different from that of the third embodiment will be described. The compressed image output unit 81 has a function of merging the compressed local data, and sends out a compressed output image.

In general, an image has an enormous amount of information and is usually compressed and handled. For this reason, although it is expected that compression processing will be performed after output, wide-area processing can be performed with local data compressed by this method.
In the present embodiment, when block coding such as the JPEG method is performed, it is necessary to code the difference between DC components in each block, but this is performed while the compressed image output unit 81 captures the difference.

Fifth Embodiment As a fifth embodiment of the present invention, a description will be given of a rotation processing method involving clipping processing of a plurality of raster images. FIG. 16 is a block diagram for explaining a fifth embodiment of the present invention. In the figure, a configuration different from that of the third embodiment will be described.
0 is clip information.

The configuration of FIG. 16 different from that of the second embodiment will be described. The clip information generation unit 90 has a function of generating clip information from the position and shape information of a plurality of raster images, and sends the clip information 280 to the MB address list generation unit 22.

Clipping is a process of superimposing a plurality of pieces of raster data input in raster scanning order according to their position information. Therefore, it is necessary to perform raster scanning even if it is only the number of pieces of raster data to be overlapped in order to generate one output image. Therefore, the clipping process requires a page memory because the order of input and output is different as in the rotation process. Further, the clipping process needs to be performed for each pixel in accordance with the overlapping information such as the edge list. In this case, data access alone is enormous.

As described above, the clip processing is also divided into the local processing and the wide area processing as in the rotation processing. That is, clip processing for each pixel is performed on local data, and wide processing is performed on a local data unit basis. As shown in FIG. 15, when an address is generated in local data, that is, in block units, it is performed only for a local region that finally comes to the top based on clip information. For example, raster A,
If the rasters A and B exist in the block when the clip processing of B is performed, the block address is generated for each raster. When the raster A is completely hidden by the raster B, only the raster B generates an address for this block. As a result, blocks that do not include the boundaries of the raster written in the intermediate buffer 61 are written to the write MB band buffer 72 as they are, and blocks that include the boundaries are subjected to pixel-based local processing on the write MB band buffer 72. be able to.

According to the present embodiment, the clipping process is performed without using the page memory even when the rotation process is not involved, so that the local process and the global process of the clipping process can be performed, and the process can be performed in a short calculation time. Furthermore, there is no need to perform rotation processing of a portion hidden in the case of clipping processing involving rotation, and the amount of calculation can be reduced.

[0060]

As is apparent from the above description, according to the present invention, the second rectangular block is set so as to include the first rectangular block tilted according to the rotation angle, and the second rectangular block is set. Since the input image is sequentially read in block units, and a first rectangular block subjected to a rotation process in accordance with the rotation angle is obtained from the read second rectangular block, a page memory is used when rotating at an arbitrary rotation angle. Without reading, reading an input image before rotation and reading from an output image after rotation can be performed only in the raster scan direction.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of the present invention.

FIG. 2 is a flowchart illustrating the operation of the present invention.

FIG. 3 is a block diagram illustrating a configuration of a first exemplary embodiment of the present invention.

FIG. 4 is a flowchart illustrating an operation of the first exemplary embodiment of the present invention.

FIG. 5 is a flowchart illustrating an operation of the first exemplary embodiment of the present invention.

FIG. 6 is an explanatory diagram of a read band buffer according to the first embodiment of the present invention.

FIG. 7 is an explanatory diagram of an address list according to the first embodiment of this invention.

FIG. 8 is a flowchart illustrating an operation of the first exemplary embodiment of the present invention.

FIG. 9 is an explanatory diagram of a wide area process in the first embodiment of the present invention.

FIG. 10 is an explanatory diagram of a local process in the first embodiment of the present invention.

FIG. 11 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 12 is an explanatory diagram of a wide area process in the second embodiment of the present invention.

FIG. 13 is a block diagram showing a configuration of a third example of the present invention.

FIG. 14 is a diagram illustrating a third embodiment of the present invention.

FIG. 15 is a block diagram illustrating a configuration of a fourth exemplary embodiment of the present invention.

FIG. 16 is a block diagram showing a configuration of a fifth example of the present invention.

FIG. 17 is a block diagram showing a configuration of a conventional example.

FIG. 18 is an explanatory diagram of a rotation process in a conventional example.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Image input means 20 Address control means 21 Address list generation part 22 Macro block (hereinafter referred to as MB) address list generation part 29 Inverse transformation coordinate calculation part 30 Address list sort part 31 MB address list sort part 40 Band data storage means 41 Reading Band buffer for band 42 Band buffer read controller 43 MB band buffer for read 44 MB buffer read controller 46 Buffer reader 47 Buffer for read 48 Page memory reader 49 Source page memory 50 Local data converter 51 Local data converter 52 Conversion ROM 60 Conversion data storage means 61 Intermediate buffer 62 Intermediate buffer control unit 70 Band data storage means 71 Write band buffer 72 Write MB buffer 76 buffer File writing unit 77 Writing buffer 78 Page memory writing unit 79 Destination page memory 80 Image output unit 90 Clip information generation unit 200 Input image data 210 Inverted address list on source side 211 Inverted MB on source side Address list 220 Source side address information 221 Sorted source side address list 222 Read band buffer control signal 223 Sorted source MB address list 224 Read MB buffer control signal 227 Source side address information 228 Source side address Information 229 Inverse conversion coordinates 230 Local data 231 Conversion coordinates of local data 231 Conversion information of conversion ROM 240 Converted local data 250 Address information on destination side 51 Destination-side address list 252 Intermediate buffer control signal 253 Destination-side MB address list 254 Intermediate buffer control signal 260 Converted local data 268 Destination-side address information 269 Destination-side address information 270 Output image data 280 Clip information 400 local data before rotation 410 center pixel before rotation 420 center pixel of rotation 430 ideal local data after rotation 440 center pixel after rotation 450 local data after rotation

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Hiroaki Tomita 430 Nakaicho, Nakaicho, Ashigarakami-gun, Kanagawa Prefecture Inside Green X-Tech Fuji Xerox Co., Ltd. Inside

Claims (10)

[Claims] 1. An image processing apparatus for obtaining an output image by rotating an input image in units of a first rectangular block composed of a plurality of pixels, wherein the image processing apparatus includes a first rectangular block tilted according to a rotation angle. Reading means for setting two rectangular blocks and sequentially reading the input image in units of the second rectangular blocks; and performing local rotation according to a rotation angle from the second rectangular blocks read by the reading means. An image processing apparatus comprising: means for generating the first rectangular block to which the first rectangular block is applied; and storage control means for sequentially storing the first rectangular block generated by the generating means in a storage unit. . 2. The reading means sets and reads the second rectangular block so that the first rectangular blocks included in the second rectangular block share one side with each other. The image processing device according to claim 1. 3. The reading means sequentially reads the second rectangular blocks included in the band data in a band data unit consisting of a plurality of lines, and each band data sets a line partially overlapping with another band data. The image processing apparatus according to claim 1, further comprising: 4. An image input unit for transmitting the input image line by line to the band data, wherein the reading unit determines whether or not the second rectangular block exists in the band data, 4. The image processing apparatus according to claim 3, wherein when the second rectangular block is present in the band data according to the determination result, the second rectangular blocks included in the band data are sequentially read. apparatus. 5. The image processing apparatus according to claim 1, further comprising control means for reading out the first rectangular blocks from the storage unit in an output order according to a rotation angle. 6. The storage unit stores the first rectangular block together with identification information, and the reading unit reads the first rectangular block from the storage unit in a predetermined order according to a rotation angle. 6. The image processing apparatus according to claim 5, further comprising a unit. 7. An image processing method for obtaining an output image by rotating an input image in units of a rectangular block composed of a plurality of pixels, wherein a second rectangular block including a first rectangular block tilted according to a rotation angle is provided. A reading step of sequentially reading from the input image; a step of generating, from the second rectangular block read in the reading step, the first rectangular block subjected to local rotation according to a rotation angle; Storing the first rectangular block generated in the step in a storage unit. 8. The reading step includes setting and reading the second rectangular block so that the first rectangular blocks included in the second rectangular block share one side with each other. The image processing method according to claim 7, wherein: 9. The reading step sequentially reads the second rectangular blocks included in the band data in a band data unit consisting of a plurality of lines, and each band data includes a line partially overlapping with other band data. The image processing method according to claim 7, further comprising: 10. The image processing method according to claim 7, further comprising a control step of reading out the first rectangular blocks from the storage unit in an output order according to a rotation angle.