JP4732183B2 - Image processing apparatus, image processing method, and program for causing computer to execute the method - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program for causing a computer to execute the method.

  Conventionally, in an image forming apparatus, in order to rotate an image, the image is divided into certain rectangles, read from the memory in units of blocks, rotated, and then written back to the memory in units of blocks. For example, when rotating a 1-bit binary image, if one block is 256 × 256 pixels, a block of 64 bytes in the main scanning direction and 256 lines in the sub-scanning direction is transferred from the memory to the rotator SRAM (Static RAM). Rotate while reading and writing to the memory write position.

  Here, when accessing the memory in units of blocks, the start address of the block is generally limited by the HW restriction of the memory bus width. If this is ignored and the start address is set in units of bytes for a 32-bit width memory, for example, the memory access address generation method becomes complicated, and if the HW is not designed according to the implementation, the performance of the memory access increases. Decrease occurs.

  For example, when an image is rotated 90 degrees or 270 degrees, performance is degraded unless the image is blocked in consideration of the image layout (address on the memory) of the rotated image. In general, both the pre-rotation image and the post-rotation image can be rotated only in the memory address, main scanning length, and sub-scanning length that satisfy specific boundary constraints. Therefore, the block for rotation is fixed, and an image having an odd size or an image placed at an odd memory location (address) cannot be rotated.

  In order to solve such a problem, in the technique of Patent Document 1, the rotation processing unit for performing the rotation processing in units of blocks is performed by 4 bytes in the image rotation processing, and the access unit to the memory is also set to 4 bytes accordingly. Set to byte. In the technique of this document, when the input head address is not aligned with the rotation processing unit, that is, when the grid and the head part of the rotation processing unit do not match, the lower-order bits of the input head address are set to 0. At the same time as aligning, the masking amount is calculated and masked, and then output to the output means.

  FIG. 11 is a diagram for explaining generation of an input address in the image rotation processing according to the conventional example. Here, a lattice 1101 (a lattice represented by a dotted line in the figure) for determining the rotation processing unit 1110 is one in which the lower 2 bits are 0, that is, the input address is [1: 0] = 0. As shown in FIG. 11, the rotation processing unit 1110 is 4 bytes (32 × 32 bits), and the memory access unit is 4 bytes. The memory address is in bytes.

  In this conventional example, when the input head address is not aligned in the rotation processing unit, for example, in the case of the upper left vertex 1111a in the figure in the region 1111, the input address conversion means simultaneously sets the lower-order bits of the input head address to 0. The mask amount is calculated and output to the output means. In the drawing, in the unaligned rotation processing unit, the data 1121 in the data 1120 is a masked portion. Therefore, the valid data is only 3 bytes obtained by subtracting 1 byte from 4 bytes, as represented by data 1122 to 1124 in the figure.

JP 2001-274975 A

  However, in this method, when the rotating block is small, the number of times of converting the input head address and calculating the mask amount is increased, and the memory access is wasted. Further, when the rotation processing unit is increased with this method, the number of bits to be masked increases, reading of unused addresses increases, and memory access is wasted.

  The present invention has been made in view of the above problems, and an object thereof is to provide an image processing apparatus, an image processing method, and a program for causing a computer to execute the method, which can eliminate useless memory access and have high image processing efficiency. .

In order to solve the above-described problems and achieve the object, the invention according to claim 1 includes a dividing means for dividing an image memory for storing image data into rectangular areas, and an input memory for reading out the image data from the image memory. Input address generation means for generating an address; memory access control means for reading out image data by burst access to the image memory based on the input memory address generated by the input address generation means; and the input address generation means Rotation processing means for reading the image data on the basis of the input address generated by the input address and performing rotation processing, and the input address generation means reads the image data stored in the image memory in the main scanning direction. beginning a of the divided rectangular areas by said division means a memory address The first memory address generating means for generating less, and, the top non-access amount is non-accessible volume of the head pixel in a predetermined width in the main scanning direction, a non-access up to the rear end pixel in the predetermined width of the main scanning direction rear non-access amount is an amount, and the by the memory access control means for calculating a burst length in the rectangular region to burst access, a pixel corresponding to the first memory address previously Atamaa dress generated by the generating means the non-access amount calculating means for calculating the number of pixels to stop reading after having the memory access control means, after a pixel corresponding to the destination Atamaa dress, the leading non-access by the non-access amount calculation means the amount, the rear end non-access amount, and the image data of the number of pixels that have been calculated based on the burst length Skip viewed burst accesses in the image memory reads the stored image data, the image data subjected to rotation processing by the rotation processing unit, and characterized in that the burst transfer to the image memory.

Such invention in claim 2, the image processing apparatus according to claim 1, the image data read by the image memory access control means, generating an output address for generating an output memory address for storage in the image memory The memory access control means burst-transfers the read image data to the image memory based on the output memory address generated by the output address generation means. To do.

Such invention in claim 3, the image processing apparatus according to claim 2, wherein the memory access control means, said output memory address, and on the basis of the burst length, the read image data, the image memory It is characterized by burst transfer.

According to a fourth aspect of the present invention, in the image processing apparatus according to the second or third aspect , the memory access control unit masks a predetermined number of pixel data when storing the image data in the image memory. It has a mask generation means for performing processing and storing it.

According to a fifth aspect of the present invention, in the image processing apparatus according to the fourth aspect , the mask generation unit performs a mask process for a predetermined number of pixels subsequent to the head reading skip, and the rear A rear end mask generation unit that performs mask processing for a predetermined number of pixels preceding skipping of the end reading, and the image data is subjected to mask processing by the head mask generation unit and mask processing by the rear end mask generation unit; The mask processing is performed.

According to a sixth aspect of the present invention, in the image processing apparatus according to any one of the first to fifth aspects, the dividing unit divides the image memory into rectangular areas in units of bytes. To do.

Such invention in claim 7, an image processing method in images processing device, the dividing means, a dividing step of dividing an image memory for storing image data in the rectangular area, the input address generating unit, the image memory An input address generation step for generating an input memory address for reading the image data from the memory, and a burst access to the image memory based on the input memory address generated in the input address generation step by the memory access control means, A memory access control step for reading data; and a rotation processing step for reading the image data and performing rotation processing based on the input address generated in the input address generation step by a rotation processing means, and generating the input address The process is performed by the top memory address generating means. The first memory address generating step of generating the start address of the divided rectangular area in the division step a first memory address in the main scanning direction for reading the image data stored in the memory, and, by non-access amount calculation means, a main A head non-access amount that is a non-access amount from the head pixel in a predetermined width in the scanning direction, a rear end non-access amount that is a non-access amount up to a rear end pixel in the predetermined width in the main scanning direction, and the memory access by controlling means for calculating a burst length within the rectangular area to be burst access, non-access for calculating the number of pixels to stop reading the subsequent pixel corresponding to the first memory address generated by the generating step the previously Atamaa dress the amount calculating step, wherein, the memory access control step, pixels corresponding to the destination Atamaa dress Descending, wherein the first non-access amount in a non-access amount calculation step, the rear end non-access volume and burst accesses in the image memory skip image data of the number of pixels is calculated based on the burst length, stores The read image data is read out, and the image data subjected to the rotation process in the rotation process step is burst-transferred to the image memory.

According to an eighth aspect of the present invention, in the image processing method according to the seventh aspect , the output memory address for storing the image data read out in the image memory access control step by the output address generating means in the image memory. An output address generation step for generating the memory access control step, wherein the memory access control means reads the image data read out based on the output memory address generated in the output address generation step; It is characterized by burst transfer to the image memory.

According to a ninth aspect of the present invention, in the image processing method according to the eighth aspect , in the memory access control step, the memory access control means reads the read image based on the output memory address and the burst length. The data is burst-transferred to the image memory.

According to a tenth aspect of the present invention, in the image processing method according to the eighth or ninth aspect , the memory access control step generates a mask when the memory access control means stores the image data in the image memory. The method includes a mask generation step of performing mask processing on a predetermined number of pixel data by means and storing.

According to an eleventh aspect of the present invention, in the image processing method according to the tenth aspect , in the mask generation step, the mask generation step performs mask processing for a predetermined number of pixels following the head reading skip by the head mask generation means. And a rear end mask generating step of performing mask processing by a predetermined number of pixels preceding the rear end skipping by the rear end mask generating means, and mask processing by the head mask generating means and the rear end mask A mask process is performed on the image data by a mask process by a generation unit.

According to a twelfth aspect of the present invention, in the image processing method according to any one of the seventh to eleventh aspects, the dividing step divides the image memory into rectangular areas in units of bytes. It is characterized by being.

According to a thirteenth aspect of the present invention, a program causes a computer to execute the image processing method according to any one of the seventh to twelfth aspects.

  According to the first aspect of the present invention, the input memory address for reading the image data from the image memory storing the image data is generated, and the image memory is burst-accessed based on the generated input memory address to Since reading is performed, memory access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.

According to the first aspect of the present invention, the top memory address in the main scanning direction for reading the image data stored in the image memory is generated, and the pixels that stop reading from the pixel corresponding to the generated top memory address are stopped. After the pixel corresponding to the head memory address is calculated, the image data corresponding to the calculated pixel number is skipped, the image memory is burst-accessed, and the stored image data is read to cope with unnecessary pixels. Since the memory is not accessed, the memory access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.
According to the first aspect of the present invention, the image data is read based on the generated input address, is subjected to rotation processing, and the rotation-processed image data is burst transferred to the image memory. As a result, memory access can be omitted even in image rotation processing, and an image processing apparatus with high image processing efficiency can be provided.

According to the first aspect of the present invention, the head non-access amount, which is the non-access amount from the head pixel in the predetermined width in the main scanning direction, is calculated, and non-access until the rear end pixel in the predetermined width in the main scanning direction is reached. Since the rear end non-access amount, which is the amount, is calculated, the burst length for burst access is calculated, and the burst access to the image memory is performed based on the head memory address, the head non-access amount, the rear end non-access amount, and the burst length. Since there is no access to the memory corresponding to unnecessary pixels at the head and rear ends along the main scanning direction, memory access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.
According to the first aspect of the invention, the image memory is divided into rectangular areas, the start address of the divided rectangular area is generated, the burst length in the rectangular area is calculated, and burst transfer is performed. Access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.

According to the invention of claim 2 , an output memory address for storing the read image data in the image memory is generated, and the read image data is burst to the image memory based on the output memory address. Since the transfer is performed without accessing a memory corresponding to an unnecessary pixel, the memory access can be omitted and an image processing apparatus with high image processing efficiency can be provided.

According to the third aspect of the present invention, since the read image data is burst-transferred to the image memory based on the output memory address and the burst length, burst writing is performed only by accessing the output memory address. Access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.

According to the invention of claim 4 , when storing the image data in the image memory, the predetermined number of pixel data is masked and stored, so unnecessary writing can be omitted, and the efficiency is high. Memory access can be omitted, and an image processing apparatus with high image processing efficiency can be provided.

According to the fifth aspect of the present invention, in the mask process, the mask process is performed for a predetermined number of pixels following the head non-access amount, and the mask process is performed for a predetermined number of pixels preceding the rear end non-access amount. Therefore, when storing the image data in the image memory, the predetermined number of pixel data are masked and stored at both the front and rear ends in a predetermined width in the main scanning direction, so unnecessary writing is omitted. Therefore, it is possible to provide an image processing apparatus with high efficiency, high memory efficiency, and high image processing efficiency.

According to the invention of claim 6 , in the division, the image memory is divided into rectangular areas in units of bytes, so that burst access can be performed in units of bytes, so that memory access can be omitted, and image processing efficiency is high. There is an effect that an image processing apparatus can be provided.

According to the invention of claim 7 , the input memory address for reading the image data from the image memory storing the image data is generated, and the image memory is burst-accessed based on the generated input memory address, and the image data is Since reading is performed, memory access can be omitted, and an image processing method with high image processing efficiency can be provided.

According to the seventh aspect of the present invention, a top memory address in the main scanning direction for reading image data stored in the image memory is generated, and a pixel that stops reading after the pixel corresponding to the generated top memory address After the pixel corresponding to the head memory address is calculated, the image data corresponding to the calculated pixel number is skipped, the image memory is burst-accessed, and the stored image data is read to cope with unnecessary pixels. Since the memory is not accessed, the memory access can be omitted, and an image processing method with high image processing efficiency can be provided.
According to the seventh aspect of the present invention, the image data is read based on the generated input address, the rotation process is performed to perform the rotation process, and the image data subjected to the rotation process is burst transferred to the image memory. As a result, memory access can be omitted even in image rotation processing, and an image processing method with high image processing efficiency can be provided.

According to the seventh aspect of the present invention, the head non-access amount that is the non-access amount from the head pixel in the predetermined width in the main scanning direction is calculated, and the non-access until the rear end pixel in the predetermined width in the main scanning direction is reached. Since the rear end non-access amount that is the amount is calculated, the burst length for burst access is calculated, and burst access is performed to the image memory based on the top memory address, the top non-access amount, the rear end non-access amount, and the burst length. Since there is no access to the memory corresponding to unnecessary pixels at the head and rear ends along the main scanning direction, memory access can be omitted, and an image processing method with high image processing efficiency can be provided.
According to the invention of claim 7 , the image memory is divided into rectangular areas, the start address of the divided rectangular area is generated, the burst length in the rectangular area is calculated, and burst transfer is performed. Access can be omitted, and an image processing method with high image processing efficiency can be provided.

According to the eighth aspect of the present invention, an output memory address for storing the read image data in the image memory is generated, and the read image data is burst to the image memory based on the output memory address. Since the transfer is performed without accessing a memory corresponding to an unnecessary pixel, the memory access can be omitted and an image processing method with high image processing efficiency can be provided.

According to the ninth aspect of the present invention, since the read image data is burst-transferred to the image memory on the basis of the output memory address and the burst length, burst writing is performed only by accessing the output memory address. Access can be omitted, and an image processing method with high image processing efficiency can be provided.

According to the invention of claim 10 , when storing the image data in the image memory, the predetermined number of pixel data is masked and stored, so unnecessary writing can be omitted, and the efficiency is high. Memory access can be omitted, and an image processing method with high image processing efficiency can be provided.

According to the invention of claim 11 , in the mask process, the mask process is performed for a predetermined number of pixels subsequent to the head non-access amount, and the mask process is performed for a predetermined number of pixels preceding the rear end non-access amount. Therefore, when storing the image data in the image memory, the predetermined number of pixel data are masked and stored at both the front and rear ends in a predetermined width in the main scanning direction, so unnecessary writing is omitted. It is possible to provide an image processing method that is high in efficiency, can eliminate memory access, and can provide high image processing efficiency.

According to the twelfth aspect of the present invention, in the division, the image memory is divided into rectangular areas in units of bytes, whereby burst access can be performed in units of bytes, so that memory access can be omitted, and image processing efficiency is high. There is an effect that an image processing method can be provided.

According to the invention concerning Claim 13, there exists an effect that the program which makes a computer perform the image processing method as described in any one of Claims 7-12 can be provided.

  DETAILED DESCRIPTION Exemplary embodiments of an image processing apparatus, an image processing method, and a program for causing a computer to execute the method according to the present invention will be described below in detail with reference to the accompanying drawings.

(Embodiment)
When the image processing apparatus according to the embodiment divides the image information into read rotation processing units and performs rotation processing, the image processing apparatus acquires the start address and burst length information of the rotation processing unit, and performs burst access from the read start address. Read. When writing after rotation, burst transfer is performed from the start address of writing.

  If the start address of the pixel data in the rotation processing unit is not aligned in the rotation processing unit, the portion that does not access the memory is skipped, the mask processing is performed on the portion without the image data, and burst transfer is performed for writing. That is, determine the start address, set and read the part that does not access the memory, set the mask of only the required number of bits, and burst transfer, the start address of the pre-rotation image and post-rotation image in bytes It can be set, and unnecessary memory access is stopped in areas that are not aligned, and unnecessary memory access can be reduced by performing mask transfer and burst transfer, thus suppressing a reduction in rotation speed in image rotation processing. The efficiency of the rotation process can be increased.

  FIG. 1 is a functional block diagram of an image forming apparatus including an image processing apparatus according to an embodiment. The image forming apparatus includes a scanner 1, an input processing unit 2, an operation display unit 3, an output processing unit 4, an image output unit 5, a storage unit (HDD) 6, and an image processing device 10.

  The scanner 1 reads a document image. The scanner 1 irradiates an original moving in the sub-scanning direction with reading light, and photoelectrically converts the reflected light by a photoelectric conversion element such as a CCD (Charge Coupled Device) to read an image of the original. When the scanner 1 includes an ADF (Auto Document Feeder) that sends out a plurality of documents to be mounted one by one, the document sent from the ADF is sequentially read as analog image data, and the analog image data is input. It transmits to the processing unit 2.

  The input processing unit 2 receives the analog data read by the scanner 1, converts it into digital image data, outputs it to the storage unit 6, stores it, and sends it to the image processing apparatus 10 together with it.

  The operation display unit 3 displays a preview image and accepts a setting input by the operator. The operation display unit 3 reads by the scanner 1, input processing by the input processing unit 2, display of a preview image by the operation display unit 3, image output processing by the output processing unit 4, image output processing by the image output unit 5, and the like. Accept setting input. Moreover, the operation display part 3 displays the image after a process, when various image processing is performed.

  For example, a liquid crystal display or the like is used as the operation display unit 3, and a touch panel is arranged on the upper part of the liquid crystal monitor. The operation display unit 3 includes various operation keys and a touch panel disposed on the liquid crystal display. Various operations necessary for operating the image forming apparatus are performed from the operation keys and the touch panel. In particular, image processing (image processing) of a document image on a document to be printed by the user, settings regarding printing conditions, and post-processing It accepts various setting operations such as settings.

  The operation display unit 3 receives a request input for performing rotation processing on the read image. The accepted rotation processing requests are, for example, 90 degrees, 180 degrees, and 270 degrees clockwise. Or it can also be set as the structure which receives the input by numerical information. Moreover, it can also be set as the structure which displays a protractor-like icon and receives an angle by contact input.

  The image processing apparatus 10 receives the image data processed by the input processing unit 2, and in accordance with the rotation processing request received by the operation display unit 3, the read image information is converted into a rectangular rotation processing unit in the two-dimensional distribution. The image is divided and subjected to a rotation process. After the rotation process, the image data is output as a display on the operation display unit 3 as image information obtained by combining the divided processing units. The image is output as an image output in the output processing unit 4.

  The output processing unit 4 performs output processing on the image processed by the image processing apparatus 10 based on the received processing setting. The output processing unit 4 performs various necessary image processing such as gamma conversion on the image data.

  The output processing unit 4 is configured to be read from the image memory when performing the output process, to perform the process, and to be read and written by the image processing apparatus 10 according to the embodiment when the output is performed. Can do.

  The image output unit 5 outputs an image according to the setting of the output process performed by the output processing unit 4. Here, the output processing performed by the image output unit 5 not only forms an image on output paper and outputs the image, but also includes post-printing post-processing such as stapling processing and punching processing.

  The storage unit 6 is composed of a large-capacity storage medium such as a large-capacity RAM (Random Access Memory) or a hard disk (HDD), and classifies and stores image data read by the scanner 1 for each file.

  The image processing apparatus according to the present invention will be described with respect to an example in which rotation processing is performed in the embodiment. However, not only image rotation processing but also image processing using a predetermined block as a processing unit, The present invention can be applied to those arranged in a predetermined rule in the main scanning direction. For example, it can be applied when the output processing unit 4 performs output processing as described above. In the input processing unit 2, for example, when performing the γ correction processing, the image file once stored in the storage unit 6 is held in the image memory, read from here, processed, and stored again in the image memory. It can also be applied to processing.

  FIG. 2 is a functional block diagram of the image processing apparatus according to the embodiment. The image processing apparatus 10 includes an image size setting unit 11, a rotation angle setting unit 12, an input address setting unit 13, an output address setting unit 14, an image dividing unit 15, an input address generating unit 16, an output address generating unit 17, a memory 18, An input unit 19, a rotation unit 20, and an output unit 21 are provided.

  The image size setting unit 11 sets the image size of the image to be processed from the image data subjected to the input processing by the input processing unit 2. Here, the image data is processed for each page.

  The rotation angle setting unit 12 sets the rotation angle of the image selected and input by the operator from the operation display unit 3 to the image data to be processed.

  When the image information stored in the memory 18 is rotated, the input address setting unit 13 sets a start address for reading the entire page of image information stored in the input unit 19. The input address setting unit 13 sets two parameters, the horizontal width and the vertical length of the input image data.

  The output address setting unit 14 sets a start address for writing one page of image information when the image information stored in the memory 18 is rotated and output by the output unit 21. The output address setting unit 14 sets two parameters, the horizontal width and the vertical length of image data when writing to the memory.

  The image dividing unit 15 divides the memory address corresponding to the image data according to the rotation processing unit according to the image size set by the image size setting unit 11. Here, the rotation processing unit to be divided is a square area of 4 bytes × 4 16 bytes in the horizontal direction (main scanning direction) and 128 lines in the vertical direction (sub-scanning direction). In this way, since it is a considerably large area compared to the conventional divided area, it is not necessary to access memory addresses corresponding to individual pixel points by burst-transferring up to 16 bytes in the main scanning direction. So memory access can be reduced. The present invention can also be applied to a case where this rotation processing unit is not aligned in the memory.

  The input address generation unit 16 generates read address and burst amount information in the rotation processing units divided by the image dividing unit 15. Here, the memory 18 stores image data for at least one page. Alternatively, the memory 18 preferably has a capacity capable of storing image data of two pages or more for reading and writing.

  The input address generation unit 16 receives the head address in the main scanning direction in the rotation processing unit from the image division unit 15, calculates the non-access amount and the mask amount, and transmits the information on the head address and the burst amount to the input unit 19. .

  The input unit 19 reads image information from the memory 18 in accordance with the leading address, non-access amount, and burst amount of one line of the rotation processing unit generated by the input address generation unit 16. Here, the head address information in the main scanning direction in the rotation processing unit is received, skipped by the non-access amount, and the memory access after the non-access amount after the head address is performed by burst access to the rest of one line. Stop. Subsequently, the head address information of the next line in the rotation processing unit is received, and the same reading process is performed. Such reading processing is performed for each rotation processing unit.

  The rotating unit 20 is SRAM (can be configured with FF) and has a size including a rotation processing unit. The rotation unit 20 temporarily holds the image information of the rotation processing unit read by the input unit 19 in a form subjected to rotation processing.

  The output unit 21 writes the image information that has been subjected to the rotation processing by the rotation unit 20 and is held by the head address generated by the output address generation unit 17, and thereafter the non-access amount in the main scanning direction of the rotation processing unit. One line is written by burst access based on the mask amount and the burst amount. During burst access and writing, there is no memory reading. Subsequently, the output unit 21 receives the top address information of the next line in the rotation processing unit and performs the writing process by the same burst access. Such writing processing is performed for each rotation processing unit.

  FIG. 3 is a functional block diagram of the input address generation unit 16. The input address generation unit 16 generates memory head address information and burst length information in the main scanning direction when reading pixels in a rotation processing unit. Further, it is also possible to generate the head mask amount and the rear end mask amount, and the input unit 19 performs a mask process.

  The input address generation unit 16 includes a transfer amount calculation and next line start address generation unit 161, a mask generation unit 162, a memory address generation unit 163, and a non-access amount calculation unit 164.

  The transfer amount calculation and next line start address generation unit 161 receives information on the main scanning length of the image before division from the image size setting unit 11 from the outside of the input address generation unit 16. Further, the transfer amount calculation and next line head address generation unit 161 receives information on the image main scanning length in the rotation processing unit after the division from the image dividing unit 15. Further, the transfer amount calculation and next line head address generation unit 161 receives the information of the image memory head address in the rotation processing unit after the division from the image division unit 15. Further, the transfer amount calculation and next line head address generation unit 161 receives information on the image sub-scanning length in the rotation processing unit after the division from the image dividing unit 15.

  Further, the transfer amount calculation and next-line head address generation unit 161 receives mask information from the mask generation unit 162 and receives burst length information from the non-access amount calculation unit 164 from the inside of the input address generation unit 16. Receive.

  The mask generation unit 162 includes a head mask generation unit 162a and a rear end mask generation unit 162b, and generates information on the head mask amount and the rear end mask amount, respectively.

  The non-access amount calculation unit 164 includes a head non-access amount calculation unit 164a, a rear end non-access amount calculation unit 164b, and a burst length calculation unit 164c, and the head non-access amount, rear end non-access amount, and burst length, respectively. The information of is calculated.

  The transfer amount calculation and next line start address generation unit 161 calculates the transfer amount from these pieces of information, and calculates the transfer amount information of the calculation result as the mask generation unit 162, the memory address generation unit 163, and the non-access amount calculation. To the unit 164.

  The memory address generation unit 163 generates memory start address information in the main scanning direction of the rotation processing unit to be read by the input unit 19 and transmits the information to the input unit 19.

  The burst length calculation unit 164 c included in the non-access amount calculation unit 164 transmits burst length information to the input unit 19.

  The input unit 19 uses the two pieces of information to perform burst access to the memory from the head address in the main scanning direction for one line in the rotation processing unit, so that it can be read from the memory at high speed and efficiently.

  The head mask generation unit 162a receives the information of the memory head address, and transmits the information of the bit to be masked at the head of the rotation processing unit in the sub-scanning direction to the transfer amount calculation and the head address generation unit 161 of the next line. By masking, an unaligned portion is read as if there is no apparent data. Here, the mask process is performed only in the write process.

  The head non-access amount calculation unit 164a calculates the amount of information to be skipped without masking in the main scanning direction from the information of the unaligned portion of the head portion. The rear end non-access amount calculation unit 164b calculates the amount to be skipped without masking in the main scanning direction from the information of the unaligned portion of the rear end portion. Accordingly, the input unit 19 can perform input processing without memory access and reading of pixel values by skipping the non-access amount among the unaligned portions.

  The burst length calculation unit 164c calculates the burst length for burst access in the main scanning direction without reading the memory address after reading and skipping the memory start address. Information on the burst length calculated by the burst length calculation unit 164 c is transmitted to the input unit 19. Thereby, the input unit 19 can efficiently read by accessing the memory for the necessary burst length.

  The input unit 19 reads out the pixel value corresponding to the head address in the main scanning direction of the rotation processing unit according to the head address of one line in the rotation processing unit, and performs burst access for the necessary burst length as it is to obtain the pixel value. read out. Since the memory is not accessed from the aligned memory address to the address where the actually required image exists, the memory access time can be shortened. Then, after reading the burst length in the main scanning direction within the rotation processing unit, if there is a line in the sub-scanning direction that is not accessed within the rotation processing unit, the next line generated by the input address generation unit 16 A head address is received, and burst access similar to the above is performed according to the received head address. In this way, the pixel data in the rotation processing unit is read by burst access by sequentially repeating accesses in the main scanning direction.

  FIG. 4 is a schematic diagram illustrating burst access to the memory by the input unit. The memory addresses 400 corresponding to the pixels distributed on the two-dimensional memory are divided into a lattice shape with the rotation processing unit as a lattice unit. Here, the vertical lines of the lattice are those whose lower 4 bits are 0. In the symbol, the address represented by [3: 0] = 0 in the figure is a vertical line.

  Here, the rotation processing unit is 16 bytes in the main scanning direction width, and is composed of 128 lines in the sub-scanning direction. The rotation processing unit is, for example, an area surrounded by one grid as represented by reference numeral 420 in the drawing. Here, in the memory area 451 corresponding to the rectangular image data, the input head address is not aligned in units of rotation processing as seen at the vertex 431 of the image data. That is, regions 430 and 450 are not aligned. On the other hand, the region 420 and the like are aligned.

  The address data 451 is schematically shown by extracting image data for one line in the main scanning direction in which the region 450 is present, and 16 bytes are schematically expressed in the main scanning direction. In this address data 451, since the head address is not aligned in the rotation processing unit, there is no memory access (reference numeral 452) from the first 8 bytes. Then, in the subsequent 4 bytes from the 9th byte, the pixel value starts in the middle of this byte, and therefore mask processing is performed on 1 byte up to the middle (reference numeral 453). Then, the valid data portion for 3 bytes and the remaining valid data for 4 bytes (reference numeral 454) without the mask follow. Access to the actual pixel value is after the portion of the mask 453.

  Accordingly, the data 451 is skipped without memory access, the input head address is set to data 453, mask processing is performed on the data 453, and the subsequent valid data 454 is read by burst access.

  FIG. 5 is a diagram for explaining that the rotation unit 20 performs the rotation process on the pixel value of the image processing unit. The rotation unit 20 is configured by, for example, an SRAM, and reads and holds a pixel value corresponding to a rotation processing unit for one line in the main scanning direction of the rotation processing unit. That is, P11, P12,..., P1n as the first line, P21, P22,... As the next line, and Pnn as the last pixel point are read and held. Table 510 shows this as a matrix.

  The rotation unit 20 performs rotation processing on the image information according to the rotation angle set by the rotation angle setting unit 12. The stored pixel point is subjected to a set angle, for example, a 90 degree rotation process to obtain a table 550 that schematically represents the rotated pixel. Here, Pn1,... P21, P11 are arranged and held in the main scanning direction.

  The output unit 21 performs burst write processing on the pixel information rotated by the rotation unit 20 based on the output head address, non-access amount information, mask amount information, and burst length information generated by the output address generation unit 17.

  FIG. 6 is a functional block diagram of the output address generation unit 17. The output address generation unit 17 includes a transfer amount calculation and start address generation unit 171 of the next line, a mask generation unit 172, a memory address generation unit 173, and a non-access amount calculation unit 174.

  Similarly to the case of the input address generation unit 16, the transfer amount calculation and next line start address generation unit 171 receives information on the main scanning length of the image before division from the image size setting unit 11. In addition, the image division unit 15 receives the divided image main scanning length information. Further, the information of the start address of the divided image memory is received from the image dividing unit 15. Further, the image dividing unit 15 receives information on the divided image sub-scanning length.

  Further, the transfer amount calculation and next line start address generation unit 171 receives mask information from the mask generation unit 172, and receives non-access amount information and burst length information from the non-access amount calculation unit 174.

  The transfer amount calculation and start address generation unit 171 of the next line calculates the transfer amount from these pieces of information, and uses the information on the transfer amount of the calculation result as a mask generation unit 172, a memory address generation unit 173, and a non-access amount calculation unit 174. Send to.

  The memory address generation unit 173 transmits the information of the memory start address to be written to the memory by the output unit 21 to the output unit 21, and the non-access amount calculated by the non-access amount calculation unit 174a and the burst length calculation unit 174c calculate. Information on the burst length is transmitted to the output unit 21.

  Here, the mask generation unit 172 includes a head mask generation unit 172a and a rear end mask generation unit 172b. The head mask generation unit 172a receives the information of the memory head address, and transmits bit information masked at the head in the main scanning direction of the rotation processing unit to the transfer amount calculation and head address generation unit 171 of the next line. In addition, the head mask generation unit 172a transmits information on bits to be masked at the head to the output unit 21. The output unit 21 stops writing pixel data that is not valid at the tip by masking in writing.

  The rear end mask generation unit 172b receives the information on the remaining image main scanning length, and transmits the information on the bit to be masked at the rear end in the main scanning direction of the rotation processing unit to the transfer amount calculation and the next line head address generation unit 171. To do. Further, the rear end mask generation unit 172b transmits bit information to be masked at the rear end to the output unit 21. The output unit 21 stops writing pixel data that is not valid at the rear end by masking in writing.

  The memory address generation unit 173 generates memory start address information in the main scanning direction in the rotation processing unit and transmits the information to the output unit 21.

  The non-access amount calculation unit 174 includes a head non-access amount calculation unit 174a, a rear end non-access amount calculation unit 174b, and a burst length calculation unit 174c. The head non-access amount, the rear end non-access amount, and the burst length, respectively. The information is calculated.

  The head non-access amount computing unit 174a computes the amount to be skipped without masking the information of the unaligned portion of the head portion. The rear end non-access amount calculation unit 174b calculates the amount to be written without masking the information of the unaligned portion of the rear end portion. The burst length calculation unit 174c calculates the number of bursts from the memory address after adding the amount to be skipped to the rear end determined by the main scanning length. Information on the burst length calculated by the burst length calculation unit 174 c is transmitted to the output unit 21.

  The output unit 21 writes the image information in the memory according to the head address, head non-access amount, head mask information, burst length, rear end mask information, and rear end non-access amount generated by the output address generation unit 17. That is, the output unit 21 masks only the mask amount while skipping according to the head mask information and the information on the rear end non-access amount and the rear end mask amount from the address after adding the address for the head non-access amount. While writing the information for the burst length.

  In this manner, the output unit 21 outputs the head address and writes the image data continuously in the continuous area during the burst access and writing the pixel information, so that the memory access can be omitted. After writing the burst length in the main scanning direction within the rotation processing unit, if there is a next main scanning direction line not accessed within the same rotation processing unit, the same burst access as described above is performed. To write to the memory 18. In this way, the pixel data for each rotation processing unit subjected to the rotation processing is output by sequentially performing the writing processing by the burst access in the main scanning direction.

  Here, when the rotation processing unit is aligned with the lattice in the memory space, there is no need for skip processing and mask processing. Only the width in the main scanning direction of the rotation processing unit can be written according to the memory head address.

  FIG. 7 is a flowchart for explaining an input address generation procedure according to the embodiment. In the input address generation unit 16, the transfer amount calculation and next line start address generation unit 161 first sets the image main scan length before division, the image main scan length after division, the image memory start address, and the image sub-scan length. Information is received (step S101). The transfer amount calculation and next line head address generation unit 161 calculates the memory head address and the remaining image main scanning length from the received information (step S102).

  The non-access amount calculation unit 164 calculates the front end non-access amount and the rear end non-access amount (step S103). The area without memory access (reference numeral 452) shown in FIG. 4 corresponds to the tip non-access amount. The rear end non-access amount may occur when there is an unaligned area at the rear end.

  The memory address generation unit 163 generates a reading start address in the main scanning direction in the rotation processing unit and transmits it to the input unit 19 (step S104).

  On the other hand, the mask generation unit 162 generates a head mask amount and a rear end mask amount (step S105). The mask area (reference numeral 453) shown in FIG. 4 corresponds to the head mask amount. The trailing edge mask amount may occur when there is an unaligned region at the trailing edge, but the description is omitted because it is the same as the leading edge.

  The burst length calculation unit 164c calculates the burst length and transmits it to the input unit 19 (step S106). The burst amount is generally between the leading address and the trailing edge in the rotation processing unit. During this time, the input unit 19 can perform burst access to the continuous area at once after the burst start address is output.

  The transfer amount calculation and head address generation unit 161 of the next line determines whether or not the processing is completed within the rotation processing unit (step S107). If it is determined that the processing is not completed (No in step S107), step S102 is performed. Return to. To read the next line that has not been accessed, repeat the same steps as above. At this time, since it is not the first line in the rotation processing unit, each piece of information obtained by calculation is fed back (step S102).

  On the other hand, if it is determined that the processing within the rotation processing unit has been completed (Yes in step S107), the transfer amount calculation and next line head address generation unit 161 determines whether or not the rotation processing for one page has been completed. Is determined (step S108). If it is determined that the rotation process for one page has been completed (Yes in step S108), the process ends as it is. When it is determined that the rotation processing for one page has not been completed (No in step S108), the process returns to step S101 again, and the same processing is repeated for the rotation processing unit that has not been accessed.

  FIG. 8 is a flowchart for explaining an output address generation procedure according to the embodiment. In the output address generation unit 17, the transfer amount calculation and next line start address generation unit 171 receives information on the image main scan length before division, the image main scan length after division, the image memory start address, and the image sub-scan length. Receive (step S201). The transfer amount calculation and next line start address generator 171 calculates the memory start address and the remaining image main scan length (step S202).

  The non-access amount calculation unit 174 calculates the front end non-access amount and the rear end non-access amount (step S203). If the schematic diagram of the input data shown in FIG. 4 is viewed as a schematic diagram of the output data subjected to the rotation processing as it is, the no memory access area (reference numeral 452) is a portion corresponding to the tip non-access amount. The rear end non-access amount may occur when there is an unaligned area at the rear end.

  The memory address generation unit 173 generates a write head address and transmits it to the output unit (step S204).

  The mask generation unit 172 generates a head mask amount and a rear end mask amount (step S205). The mask generation unit 172 transmits the generated head mask amount and rear end mask amount to the output unit 21 (step S206). Here, the output unit 21 masks and outputs the pixels corresponding to the received leading mask amount and trailing edge mask amount, and thus apparently does not perform the writing process.

  The burst length calculation unit 174c calculates the burst length and transmits it to the output unit 21 (step S207). The output unit 21 receives the non-access amount and skips the unaligned portion according to the non-access amount, stops writing, receives the head mask amount, and performs mask processing on the portion corresponding to the received head mask amount. After that, pixel values are burst-accessed and written based on the burst length. Then, a trailing edge mask process is performed, and then a writing process is performed at the trailing edge in accordance with the non-access amount information.

  The transfer amount calculation and head address generation unit 171 of the next line determines whether or not the process is completed within the rotation processing unit (step S208). If it is determined that the process is not completed (No in step S208), the process proceeds to step S208. Return to 202. When it is determined that the transfer has been completed (Yes in step S208), the transfer amount calculation and next-line head address generation unit 171 determines whether or not the rotation processing for one page has been completed (step S209), and the processing ends. If it is determined that it has not (No in step S209), the process returns to step 201, and if it is determined that the process has ended (Yes in step S209), the image rotation processing on this page ends.

  FIG. 9 is a diagram for explaining memory access in image processing between the image processing apparatus according to the conventional example and the image processing apparatus according to the embodiment. Data 910 in FIG. 9 is a mode when burst access is performed on 16-byte data in the main scanning direction by the technique of Patent Document 1. Here, the start addresses of rotation processing units that are not aligned (when [1: 0] is not 0) are aligned to [1: 0] = 0, and unnecessary portions are masked. The masked area is indicated by reference numeral 911. This is followed by valid data 912. However, since the mask area 911 is actually read by the input unit 19 and subjected to mask processing in writing, useless reading occurs.

  On the other hand, data 950 in FIG. 9 is a processing mode when the technique according to the embodiment is applied to burst access to 16-byte data. Here, an area 951 without memory access is set for unnecessary pixels, after which only a necessary part is masked (reference numeral 952), and then valid data 953 follows. For this reason, the reading of the area 951 without memory access can be stopped and the mask portion can be reduced. Consequently, the unnecessary reading is stopped and the image processing efficiency is superior to the conventional example. It has become.

  In addition, in the conventional example, when the rotation processing unit is set to be small and waste of reading is eliminated, the burst length that can be burst-accessed by the rotation processing unit that is set small is small, and the number of accesses to the memory increases. Therefore, the image processing efficiency is not high. On the other hand, in the image processing apparatus 10 according to the embodiment, it is not necessary to set the rotation processing unit small, and wasteful memory access can be eliminated. Also, the larger the rotation processing unit is set, the longer the burst length can be, so that the number of memory accesses can be further reduced, and the image processing efficiency can be improved.

  As described above, when the image processing apparatus according to the embodiment reads the image information and divides it into rotation processing units and performs rotation processing, the image processing device acquires the start address and burst length of the rotation processing unit, and only the head address is acquired. By outputting and performing burst transfer in the main scanning direction after that, it is possible to reduce useless memory access and minimize a decrease in image rotation speed.

  In addition, for a rotation processing unit in which a predetermined grid of addresses defined in two dimensions and the start address are not aligned, a portion where no memory is accessed is set, the start address is determined, and unnecessary data is skipped. By masking only the required number of bits and performing burst transfer, the start address of the pre-rotation image and post-rotation image can be set in bytes, and burst transfer reduces the unnecessary memory access and rotates the image. In this case, a decrease in rotational speed can be minimized.

(Modification 1)
The image processing apparatus according to the embodiment has been described as applied to an image forming apparatus. However, the image processing apparatus of the present invention is not applicable only to an image forming apparatus, but can be applied to an apparatus that requires image processing to store and read image data at least temporarily and write it to another memory. is there. For example, the present invention can also be applied as an image processing apparatus in a mobile phone that receives an image including a character image, rotates it, or performs some image processing and displays it on a display screen.

(Modification 2)
As another modification, for example, the present invention can be applied as an image processing apparatus in a digital camera that captures a subject image via a CCD (ChargeCoupled Device), rotates the image, or performs some image processing and displays the image on a display screen. .

(Modification 3)
As another modification, for example, the present invention can be applied as an image processing program in a personal computer (PC) that reads an image including a character image, rotates it, or performs some image processing and displays it on a monitor screen.

(Hardware configuration etc.)
FIG. 10 is a block diagram illustrating a hardware configuration of an image forming apparatus including the image processing apparatus according to the embodiment. This image forming apparatus is configured as a multifunction peripheral (MFP) having multiple functions such as a fax machine and a scanner. As shown in the figure, this MFP has a configuration in which a controller 2210 and an engine unit 2260 are connected by a PCI (Peripheral Component Interconnect) bus. The controller 2210 is a controller that controls inputs from the FCUI / F 2230 and the operation display unit 3 such as control of the entire MFP, image display control, various controls, and image processing control. The engine unit 2260 is an image processing engine that can be connected to a PCI bus, and includes, for example, an image processing portion such as error diffusion and gamma conversion for acquired image data.

  The controller 2210 includes a CPU 2211, a North Bridge (NB) 2213, a system memory (MEM-P) 2212, a South Bridge (SB) 2214, a local memory (MEM-C) 2217, and an ASIC (Application Specific Integrated Circuit). 2216 and the hard disk drive 6, and the north bridge 2213 and the ASIC 2216 are connected by an AGP (Accelerated Graphics Port) bus 2215. The MEM-P 2212 further includes a ROM (Read Only Memory) 2212a and a RAM (Random Access Memory) 2212b.

  The CPU 2211 performs overall control of the MFP, has a chip set including the NB 2213, the MEM-P 2212, and the SB 2214, and is connected to other devices via the chip set.

  The NB 2213 is a bridge for connecting the CPU 2211 to the MEM-P 2212, SB 2214, and AGP 2215, and includes a memory controller that controls reading and writing to the MEM-P 2212, a PCI master, and an AGP target.

  The MEM-P 2212 is a system memory used as a memory for storing programs and data, a memory for developing programs and data, and the like, and includes a ROM 2212a and a RAM 2212b. The ROM 2212a is a read-only memory used as a program / data storage memory, and the RAM 2212b is a writable / readable memory used as a program / data development memory, an image drawing memory during image processing, and the like.

  The SB 2214 is a bridge for connecting the NB 2213 to a PCI device and peripheral devices. The SB 2214 is connected to the NB 2213 via a PCI bus, and an FCUI / F 2230 and the like are also connected to the PCI bus.

  The ASIC 2216 is an IC (Integrated Circuit) for multimedia information processing having hardware elements for multimedia information processing, and has a role of a bridge for connecting the AGP 2215, the PCI bus, the HDD 6, and the MEM-C 2217.

  The ASIC 2216 includes a PCI target and an AGP master, an arbiter (ARB) that is the core of the ASIC 2216, a memory controller that controls the MEM-C 2217, and a plurality of DMACs (Direct Memory) that perform rotation of image data by hardware logic or the like. A Universal Serial Bus (USB) 2240 and IEEE (The Institute of Electrical and Engineering Engineers) 1394 interface 2250 are connected between the Access Controller and the engine unit 2260 via a PCI bus.

  The MEM-C 2217 is a local memory used as a transmission image buffer and a code buffer, and the HDD 6 is a storage for storing image data, programs, font data, and forms.

  The AGP 2215 is a bus interface for a graphics accelerator card that has been proposed for speeding up graphics processing, and speeds up the graphics accelerator card by directly accessing the MEM-P 2212 with high throughput.

  The operation display unit 3 connected to the ASIC 2216 receives an operation input from the operator and transmits the operation input information received by the ASIC 2216.

  An image processing program executed by the MFP in which the image processing apparatus according to the embodiment is incorporated is provided by being incorporated in advance in a ROM or the like.

  An image processing program executed by an MFP incorporating the image processing apparatus according to the embodiment is a file in an installable format or an executable format, and is a CD-ROM, a flexible disk (FD), a CD-R, a DVD (Digital Versatile). It may be configured to be recorded on a computer-readable recording medium such as Disk).

  Further, the image processing program executed by the MFP incorporating the image processing apparatus according to the embodiment is stored on a computer connected to a network such as the Internet, and is provided by being downloaded via the network. Also good. Further, the image processing program executed by the MFP incorporating the image processing apparatus according to the embodiment may be provided or distributed via a network such as the Internet.

  The image processing program executed by the MFP incorporating the image processing apparatus according to the embodiment includes the above-described units (image size setting unit 11, rotation angle setting unit 12, input address setting unit 13, output address setting unit 14, image division). Unit 15, input address generation unit 16, output address generation unit 17, input unit 19, rotation unit 20, output unit 21, and the like). As actual hardware, a CPU (processor) is the ROM By reading and executing the image display program from the above, each of the above units is stored on the main storage device, the image size setting unit 11, the rotation angle setting unit 12, the input address setting unit 13, the output address setting unit 14, the image dividing unit 15, and the input An address generation unit 16, an output address generation unit 17, an input unit 19, a rotation unit 20, an output unit 21, and the like are provided on the main storage device. It is adapted to be made.

  The embodiment or modification of the present invention described above is an example for description, and the present invention is not limited to these specific examples described here.

  As described above, the image processing apparatus, the image processing method, and the program for causing the computer to execute the method according to the present invention are useful for the image processing technology, and in particular, the image processing that reads and processes the image data from the storage device. Suitable for technology.

1 is a functional block diagram of an image forming apparatus including an image processing apparatus according to an embodiment. 1 is a functional block diagram of an image processing apparatus according to an embodiment. 3 is a functional block diagram of an input address generation unit 16. FIG. It is a schematic diagram explaining burst access. It is a figure explaining rotation part 20 performing rotation processing to the pixel value of an image processing unit. 3 is a functional block diagram of an output address generation unit 17. FIG. It is a flowchart explaining the input address generation procedure by embodiment. It is a flowchart explaining the output address generation procedure by embodiment. It is a figure explaining the memory access in the image processing of the image processing apparatus by a prior art example, and the image processing apparatus by embodiment. 1 is a functional block diagram of an image forming apparatus including an image processing apparatus according to an embodiment. It is a figure explaining the production | generation of the input address in the rotation process of the image by a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Scanner 2 Input processing part 3 Operation display part 4 Output processing part 5 Image output part 6 Memory | storage part (HDD)
DESCRIPTION OF SYMBOLS 10 Image processing apparatus 11 Image size setting part 12 Rotation angle setting part 13 Input address setting part 14 Output address setting part 15 Image division part 16 Input address generation part 17 Output address generation part 18 Memory 19 Input part 20 Rotation part 21 Output part

Claims (13)

A dividing means for dividing an image memory for storing image data into rectangular areas;
An input address generating means for generating an input memory address to read out the image data from the image memory,
Memory access control means for performing burst access to the image memory and reading out image data based on the input memory address generated by the input address generation means;
Rotation processing means that reads the image data based on the input address generated by the input address generation means and performs rotation processing;
The input address generating means
Head memory address generating means for generating a head address of a rectangular area divided by the dividing means, which is a head memory address in the main scanning direction for reading out image data stored in the image memory; and
A head non-access amount that is a non-access amount from the top pixel in a predetermined width in the main scanning direction, a rear end non-access amount that is a non-access amount up to a rear end pixel in the predetermined width in the main scanning direction, and the memory by access control means for calculating a burst length in the rectangular region to burst access, non for calculating the number of pixels to stop reading the subsequent pixels corresponding to the previously Atamaa dress generated by the first memory address generating means Having access amount calculation means,
Said memory access control means, said destination Atamaa subsequent pixels corresponding to the dress, the top non-access amount by said non-access amount calculating means, the rear end non-access volume, and number of pixels that have been calculated based on the burst length The image data is skipped, the image memory is burst accessed, the stored image data is read, and the image data subjected to the rotation processing by the rotation processing means is burst transferred to the image memory. A featured image processing apparatus. Output address generation means for generating an output memory address for storing the image data read out by the image memory access control means in an image memory;
It said memory access control means on the basis of the output memory address generated by the output address generating means, the image data read out, according to claim 1, characterized in that the image memory is to burst transfer Image processing apparatus. 3. The image processing according to claim 2 , wherein the memory access control means burst-transfers the read image data to the image memory based on the output memory address and the burst length. apparatus. The said memory access control means has a mask production | generation means which performs a mask process with respect to a predetermined number of pixel data, and stores it, when storing the said image data in the said image memory. The image processing apparatus according to 2 or 3 . The mask generating means includes
A head mask generating means for performing mask processing for a predetermined number of pixels following the head reading skip; and
Rear end mask generating means for performing mask processing for a predetermined number of pixels preceding the rear end skipping,
The image processing apparatus according to claim 4 , wherein the image data is subjected to mask processing by mask processing by the leading mask generation unit and mask processing by the trailing edge mask generation unit. The dividing means, the image processing apparatus according to any one of claims 1 to 5, characterized in that for dividing the image memory in the rectangular area in bytes. An image processing method in an image processing apparatus,
A dividing step of dividing the image memory storing the image data into rectangular areas by a dividing unit;
By the input address generating unit, an input address generating step of generating an input memory address to read out the image data from the image memory,
Based on the input memory address generated in the input address generation step by the memory access control means, a memory access control step of performing burst access to the image memory and reading out image data;
A rotation processing step by which the rotation processing means reads the image data based on the input address generated in the input address generation step and performs the rotation processing;
The input address generation step includes
A leading memory address generating step for generating a leading memory address in the main scanning direction for reading out image data stored in the image memory by a leading memory address generating means, and generating a leading address of the rectangular area divided in the dividing step; and ,
The leading non-access amount that is the non-access amount from the leading pixel in the predetermined width in the main scanning direction by the non-access amount calculating means, and the trailing edge that is the non-accessing amount up to the trailing pixel in the predetermined width in the main scanning direction non-access volume, and the by the memory access control means for calculating a burst length within the rectangular area to be burst access, stops reading of the subsequent pixel corresponding to the first memory address generating step in the generated previously Atamaa dress A non-access amount calculation step for calculating the number of pixels to be performed,
Said memory access control step, the destination Atamaa subsequent pixels corresponding to the dress, the top non-access amount by said non-access quantity calculation step, the rear end non-access volume, and number of pixels that have been calculated based on the burst length The image data is skipped, the image memory is burst-accessed, the stored image data is read, and the image data subjected to the rotation process in the rotation process step is burst-transferred to the image memory. A featured image processing method. An output address generation step of generating an output memory address for storing the image data read in the image memory access control step by the output address generation means in the image memory;
In the memory access control step, the memory access control means performs burst transfer of the read image data to the image memory based on the output memory address generated in the output address generation step. The image processing method according to claim 7 . 2. The memory access control step, wherein the memory access control means performs burst transfer of the read image data to the image memory based on the output memory address and the burst length. 9. The image processing method according to 8 . The memory access control step includes a mask generation step in which the memory access control means performs mask processing on a predetermined number of pixel data by the mask generation means when storing the image data in the image memory. The image processing method according to claim 8, wherein: The mask generation step includes
A head mask generating step of performing mask processing for a predetermined number of pixels following the head reading skip by the head mask generating means; and
A rear end mask generating means for performing mask processing for a predetermined number of pixels preceding the rear end skipping by the rear end mask generating means;
The image processing method according to claim 10 , wherein mask processing is performed on the image data by mask processing by the head mask generation unit and mask processing by the rear end mask generation unit. 12. The image processing method according to claim 7, wherein in the dividing step, the dividing unit divides the image memory into rectangular areas in units of bytes. A program for causing a computer to execute the image processing method according to any one of claims 7 to 12 .

Images