JP2005322049A - Image processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a memory capacity required at the time of image processing in a pipe line system by preventing the useless assignment of a buffer. <P>SOLUTION: This processor is provided with a storage means, a control means for driving a module for image processing, an assignment means for assigning a storage region in a storage means to the module as a buffer before the module is driven by the control means and a writing means for writing the data generated by the module driven by the control means in the buffer when the memory capacity of the buffer is sufficient for the data, and for writing the data after the buffer is extended when the memory capacity of the buffer is not sufficient for the data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique for performing image processing by a pipeline method.

  Various image processing apparatuses that perform various kinds of image processing on input image data and output them have been proposed. As an example, a pipeline-type image processing apparatus disclosed in Patent Document 1 can be cited. In Patent Document 1, a module that executes basic processing such as image enlargement and reduction, resolution conversion, and color conversion is provided for each processing, and a pipeline is formed by appropriately combining these modules. An image processing apparatus that flexibly realizes various image processing by sequentially driving modules to be formed is disclosed.

Japanese Patent Laid-Open No. 07-105020

  By the way, in the pipeline type image processing apparatus, it is necessary to transfer the data as the former processing result from the module that has completed the processing to the module that performs the next processing. This is done. That is, prior to driving each module, a storage area (hereinafter referred to as a buffer) for storing data as a result of processing is allocated to each module in the memory. Each module writes the data that is the processing result to the buffer assigned to itself, while the next driven module refers to the buffer assigned to the module that has completed processing immediately before, and the processing result To get. The assignment of buffers to modules means, for example, a buffer identifier (for example, the start address of the storage area) uniquely identifying each buffer in the management table as shown in FIG. 7, the size of the buffer, and the buffer. This is to write a module identifier (for example, a character string representing the name of the module) that uniquely identifies a module to which data is written, in association with each other. The management table is used to prevent a storage area corresponding to a buffer allocated to a certain module from being allocated as a buffer of another module. When a buffer is newly allocated, the management table This is for allocating the buffer while avoiding the storage area of the head address stored in.

  As described above, when allocating the buffer to each module forming the pipeline, it is necessary to specify the data size of the data stored in the buffer. However, the data size of the processing result cannot be uniquely specified for all modules forming the pipeline. For example, in a module that encodes and compresses input data according to a predetermined compression algorithm, data having different data sizes is generated according to the contents of the input data. For this reason, for this type of module, the data size of the processing result cannot be specified until the actual compression processing is completed. As described above, when the data size of the processing result cannot be uniquely specified, it is generally performed that the buffer is allocated with a theoretically assumed maximum value for the data size of the processing result.

  However, even when data having a different data size is generated according to the contents of the input data, the data size rarely reaches the maximum value. In other words, if a buffer is allocated with the above maximum value, most of it will not be used. As described above, since a storage area corresponding to a buffer allocated to a certain module is not allocated to another module, it is extremely useless to allocate a buffer that is not used in this way, It is not preferable. If the pipeline contains a large number of modules that cannot uniquely identify the data size of the processing result, the maximum of the above-mentioned maximum values for each module is used even though most of the modules are not actually used. Since a buffer must be allocated by value, a large amount of memory is required to execute the pipeline. For this reason, in an image processing apparatus having only a small memory, such as an image processing apparatus incorporated in a scanner device or a printer device, it is often difficult to execute pipeline image processing. .

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique for avoiding unnecessary buffer allocation and reducing the memory capacity required for pipelined image processing.

  In order to solve the above-described problems, the present invention provides a storage unit, a control unit that drives a module that performs image processing, and a memory stored in the storage unit before the module is driven by the control unit. If the data generated by the module driven by the controller driven by the control means and the allocation means for allocating an area as a buffer is stored in the buffer, the data is written in the buffer, and if the data does not fit in the buffer, An image processing apparatus having writing means for writing the data after expanding a buffer is provided.

  In order to solve the above-described problem, the present invention provides a computer device with a control function for driving a module that performs image processing, and before the module is driven by the control function, An allocation function for allocating a storage area in a memory as a buffer, and when data generated by the module driven by the control function fits in the buffer, the data is written to the buffer and does not fit in the buffer Provides a program for realizing a write function for writing the data after the buffer is expanded.

  According to such an image processing apparatus and program, a buffer having a predetermined size is allocated to the module before the module is driven, while data as a result of processing of the module does not fit in the buffer. The data is written after the buffer is expanded.

  According to the present invention, before each module forming the pipeline is driven, a buffer for storing data as a result of processing of the module is allocated, but when the data does not fit in the buffer, The data is written after the buffer is expanded. For this reason, it is necessary to allocate a buffer with the maximum value that can be theoretically assumed for the data size of the processing result even for modules that generate data having different data sizes depending on the contents of the input data. In addition, the memory capacity required for pipelined image processing is reduced.

The best mode for carrying out the present invention will be described below with reference to the drawings.
[A: Configuration]
FIG. 1 is a diagram illustrating an example of the overall configuration of an image reading apparatus 10 according to an embodiment of the present invention. As illustrated in FIG. 1, the image reading apparatus 10 includes a storage unit 110, an image reading unit 120, and an image processing unit 130. The image reading apparatus 10 performs predetermined image processing on image data corresponding to an image read by the image reading unit 120 by the image processing unit 130 and then writes the image data in the storage unit 110. In the present embodiment, the case where the image processing unit 130 compresses and encodes the image data according to a predetermined compression algorithm will be described. However, the processing performed by the image processing unit 130 is not limited to such compression processing. However, any process may be used as long as data with different data sizes is generated according to the contents of the input data. In this embodiment, the case where the processing result of the image processing unit 130 is stored in the storage unit 110 will be described. However, the processing result may be transmitted to another device via a communication network or the like. Of course, an image corresponding to the processing result may be formed and output on a recording material such as printing paper or an OHP (Over Head Projector) sheet.

  The storage unit 110 is, for example, a hard disk, and stores data delivered from the image processing unit 130. The image reading unit 120 scans and reads an image formed on the recording material as described above in units of lines, and reads image data corresponding to the read images in a predetermined data unit (in this embodiment, for one line). Data) one by one to the image processing unit 130. Hereinafter, the image data for one line is referred to as “unit data”. In this embodiment, a case where image data for one line is used as the unit data will be described. However, it is needless to say that image data for a plurality of lines or image data for one page may be used as the unit data. .

  The image processing unit 130 encodes and compresses the unit data delivered from the image reading unit 120 according to a predetermined compression algorithm, and then writes the encoded unit data into the storage unit 110. The image processing unit 130 performs processing for receiving the unit data from the image reading unit 120, compression processing, and writing processing to the storage unit 110 by the pipeline method described above. Hereinafter, the image processing unit 130 will be described in detail with reference to FIG.

  FIG. 2 is a block diagram illustrating an example of a hardware configuration of the image processing unit 130. As shown in FIG. 2, the image processing unit 130 includes a memory 210, a control unit 220, a data input module 230, a data compression module 240, and a data writing module 250.

  The memory 210 is, for example, a RAM (Random Access Memory), and is used as a work area when executing the pipeline processing. Although details will be described later, in the memory 210, a buffer used for data exchange between the modules is secured.

  As shown in FIG. 2, the control unit 220 includes a pipeline control unit 220a, a buffer allocation unit 220b, and a buffer access unit 220c, and implements the above-described pipeline processing. Each time the unit data is delivered from the image reading unit 120, the pipeline control unit 220a performs drive control of the three modules, and as shown in FIG. 3, the data input module 230, the data compression module 240, the data The writing modules 250 are driven in this order.

  The buffer allocation unit 220b allocates a buffer having a predetermined size to the module before the module is driven by the pipeline control unit 220a. More specifically, the buffer allocation unit 220b secures a storage area of a predetermined size in the memory 210, for example, by updating the storage contents of the management table (see FIG. 7) described above. A storage area is allocated as a buffer. In this embodiment, the buffer allocating unit 220b allocates the buffer A having the same size as the data size of the unit data to the data input module 230, while the buffer having a size corresponding to the average compression rate of the data compression module 240. Let B be assigned to the data compression module. Here, the compression rate is a 100-percentage of the data size of the compression result with respect to the data size of the data input to the data compression module 240. The compression rate is obtained for a plurality of pieces of data having different contents, and the average of the compression rates is obtained. By calculating, the above average compression rate can be determined. For example, when the average compression rate is 50%, the buffer allocation unit 220b allocates the buffer B with a size that is half the data size of the unit data. In the present embodiment, a case has been described in which a buffer having a predetermined size is allocated to each module before each module is driven. However, the buffer allocation unit 220b determines the size of the buffer to be allocated to each module. Needless to say, the buffer allocation unit 220b may allocate the buffer having the notified size.

  Then, the buffer access means 220c writes data as a processing result of the module to a buffer assigned to the module or reads data from the buffer in response to a request from each module. The buffer access means 220c performs processing characteristic of the present invention when writing to the buffer. Specifically, as shown in FIG. 4, the buffer access means 220c, for example, if the data delivered from each module does not fit in the buffer assigned to that module, for example, the management table (FIG. 7), the data is written after the buffer is expanded by updating the stored contents.

  The data input module 230 receives unit data delivered from the image reading unit 120 and writes the unit data to the buffer A by the buffer access means 220c. The data compression module 240 reads the unit data stored in the buffer A using the buffer access unit 220c, compresses it according to a predetermined compression algorithm, and writes the compression result to the buffer B by the buffer access unit 220c. Is. The data writing module 250 reads the data stored in the buffer B (that is, compressed unit data) by the buffer access means 220c and writes it to the storage unit 110. In the present embodiment, the case where each module forming the pipeline is mounted by hardware has been described. However, it goes without saying that each module may be mounted by software.

  Here, it should be noted that the data size of the data generated by the data compression module 240 changes according to the contents of the unit data input to the data compression module 240. In this embodiment, since the buffer B is allocated with a size that is irrelevant to the maximum value theoretically assumed for the data size of the compression result by the data compression module 240, the compression result does not fit in the buffer B. Can happen. Even in such a case, according to the buffer access means 220c, since the compression result is written after the buffer B is expanded, no particular problem occurs. In this embodiment, the case where the pipeline processing executed by the image processing unit 130 is configured in three stages will be described. However, the pipeline processing may be configured in two stages, and more than four stages. Of course, it may be configured by processing.

[B: Operation]
Next, among the operations performed by the image reading apparatus 10 according to the present embodiment, operations that clearly show the features will be described with reference to the drawings.

(B-1: Pipeline control operation)
First, the pipeline control operation performed by the control unit 220 will be described with reference to FIG. FIG. 3 is a flowchart showing the flow of the control operation performed by the control unit 220. As shown in FIG. 3, the control unit 220 first performs an initialization process by the buffer allocation unit 220b (step SA1). Specifically, the buffer allocation unit 220b allocates the above-described buffer A to the data input module 230, and allocates the above-described buffer B to the data compression module 240.

  Thereafter, the control unit 220 sequentially drives the data input module 230, the data compression module 240, and the data writing module 250 by the pipeline control unit 220a (Step SA2 to Step SA4), and executes pipeline processing. As a result, the unit data delivered from the image reading unit 120 to the image processing unit 130 is sequentially processed by the data input module 230, the data compression module 240, and the data writing module 250 according to the data flow shown in FIG. Will be written to. When the data input module 230 and the data compression module 240 write each processing result to the buffer assigned to each, the data input module 230 and the data compression module 240 perform the writing using the buffer access unit 220c described above. Hereinafter, processing performed by the buffer access unit 220c will be described in detail with reference to FIG.

(B-2: Buffer write operation)
FIG. 4 is a flowchart showing a flow of a buffer write operation performed when the buffer access unit 220c writes data delivered from each module to a buffer assigned to the module. As shown in FIG. 4, the buffer access means 220c first compares the data size of the data delivered from each module with the free capacity of the write destination buffer, and determines whether the data fits in the buffer. It is determined whether or not (step SB1). Specifically, the buffer access means 220c determines that the data will fit if the data size of the data is less than or equal to the free capacity, and conversely determines that the data will not fit if the former is larger than the latter. .

  If the determination result in step SB1 is “Yes”, the buffer access unit 220c writes the data to the buffer (step SB2), and ends the write operation. Conversely, if the determination result in step SB1 is “No”, the buffer access means 220c first writes only the portion corresponding to the free capacity from the top of the data to the buffer (step SB3). The data size of the data is updated to the data size of the unwritten portion (step SB4).

  Next, the buffer access unit 220c expands the buffer by securing an unused storage area in the memory 210 by the data size (step SB5), and repeatedly executes the processing after step SB1. In this embodiment, when there is a shortage in the free space of the buffer, the case has been described where the buffer is expanded by allocating a new storage area by the shortage, but when there is a shortage in the free space of the buffer Of course, the data may be expanded by a predetermined data size (for example, the initial size of the buffer). Further, when allocating the new storage area, a new storage area (a portion indicated by hatching in FIG. 6) may be allocated so as to be adjacent to the existing buffer as shown in FIG. As shown in FIG. 6B, a new storage area may be allocated apart from the existing buffer. Even if a new storage area is allocated apart from the existing buffer as in the latter case, there is no problem if the buffer access means 220c is operated in the manner described below. In other words, when writing data to the buffer, the data is appropriately divided and written to the existing buffer and the new storage area, and when reading data from the buffer across multiple storage areas, the data is read from each storage area. What is necessary is just to connect the data which were appropriately connected to each module.

  As described above, according to the present embodiment, when a buffer is allocated with a predetermined size and a shortage occurs in the free capacity, a storage area is secured and the buffer is expanded by the shortage. Therefore, compared to the case where a buffer is always secured at the maximum value that can be theoretically assumed, an unnecessarily large buffer is not secured, and the memory capacity required for executing pipeline processing is reduced. There is an effect that becomes possible.

[C: Modification]
Although the best mode for carrying out the present invention has been described above, it is needless to say that the embodiment may be modified as described below.

(C-1: Modification 1)
In the above-described embodiment, the case where the image processing unit according to the present invention is incorporated in an image reading device such as a scanner device has been described. However, the target in which the image processing unit is incorporated is not limited to the scanner device. Of course, it may be an image forming apparatus such as a printer or a copying machine, or may be a multifunction machine having both a scanner function, a printer function, and a copying machine function. Of course, the present invention may be applied to a single image processing apparatus such as a lip server.

(C-2: Modification 2)
In the above-described embodiment, the case where data is written to and read from the buffer using the buffer access unit 220c characteristic of the present invention has been described. However, the same module as the buffer access unit 220c is used for each module. Of course, it is also possible to add functions. Similarly, it is of course possible that the same function as the buffer allocation means 220b is given to each module. Furthermore, data that uniquely identifies each buffer (for example, the start address of each buffer), data that represents the size of the buffer, access processing (data read / write processing) for the buffer, and processing that expands the buffer Of course, an object (hereinafter referred to as a buffer object) may be generated by encapsulating a program in which the procedure is described, and the buffer object may be mounted on the image processing apparatus so as to be linked to each module.

(C-3: Modification 3)
In the above-described embodiment, the case where functions specific to the image processing apparatus according to the present invention are realized by hardware modules (pipeline control means 220a, buffer allocation means 220b, and buffer access means 220c) has been described. Of course, it may be realized. Specifically, it is of course possible to install software that realizes the same functions as the above means in a CPU (Central Processing Unit) in a computer device and to provide the same functions as those of the image processing apparatus according to the present embodiment. Further, such software may be recorded and distributed on a computer-readable recording medium such as a CD-ROM (Compact Disk-Read Only Memory), and the software is distributed via a communication network such as the Internet. Of course it is good.

1 is a diagram illustrating a configuration example of an image reading apparatus 10 according to an embodiment of the present invention. 3 is a diagram illustrating an example of a functional configuration of an image processing unit 130 of the image reading apparatus 10. FIG. 4 is a flowchart showing a flow of pipeline control operation performed by the image processing unit 130. 4 is a flowchart showing a flow of a buffer writing operation performed by the image processing unit 130. 3 is a diagram illustrating an example of a data flow in the image processing unit 130. FIG. It is a figure for demonstrating the expansion aspect of a buffer. It is a figure which shows an example of the management table which manages the allocation condition of a buffer.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Image reading apparatus, 110 ... Memory | storage part, 120 ... Image reading part, 130 ... Image processing part, 210 ... Memory, 220 ... Control part, 220a ... Pipeline control means, 220b ... Buffer allocation means, 220c ... Buffer access means , 230... Data input module, 240... Data compression module, 250.

Claims (4)

Storage means;
Control means for driving a module for image processing;
Allocating means for allocating a storage area in the storage means as a buffer to the module before the module is driven by the control means;
When the data generated by the module driven by the control means fits in the buffer, the data is written to the buffer, and when the data does not fit in the buffer, the data is expanded after the buffer is expanded. An image processing apparatus having writing means for writing. A plurality of modules that perform different image processing from each other,
The control means includes
Driving each of the plurality of modules in a predetermined order;
Of the plurality of modules, the module driven second or later reads out data generated by the module driven immediately before from the buffer assigned to the module driven immediately before and performs image processing. The image processing apparatus according to claim 1. The module is
The image processing apparatus according to claim 1, wherein image processing is performed to generate data of different data sizes according to the contents of input data. Computer equipment,
A control function for driving a module for image processing;
An allocation function for allocating a storage area in a memory of the computer device as a buffer to the module before the module is driven by the control function;
When the data generated by the module driven by the control function fits in the buffer, the data is written to the buffer, and when the data does not fit in the buffer, the data is expanded after the buffer is expanded. Program that realizes the writing function.

Images