JP2012008838A - Print document conversion device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent a situation where, due to a request of image processing to a hardware image processing device by a software RIP module, the acquisition of a processing result gets delayed more than software conversion processing means itself executes the image processing.SOLUTION: A processing execution control unit 230 refers to image processing device management information 232 when a request of image processing arrives from software RIP modules 210A-C, and estimates a time required until the completion of the image processing relating to the request for each of an image processing device 220 and image processing software 212 of the RIP modules 210 of the request sources. In this estimation, the time until the processing completion is estimated based on the performance ratio of the image processing device 220 and the image processing software 212, the number of requests or the number of pixels of each request in a queue of the image processing device 220, and the like. Then, the earlier one to complete the image processing is instructed to perform the image processing relating to the request.

Description

  The present invention relates to a print document conversion apparatus and program.

  Print data written in a page description language (hereinafter referred to as PDL; PDL is an abbreviation of Page Description Language) such as PostScript (registered trademark) or PDF (Portable Data Format: ISO 32000-1) is RIP (Raster Image Processor). ) Is converted into a bitmap image (also called a raster image) by a conversion module called) and printed by a printer. Since the conversion module needs to perform a language processing called PDL interpretation, it is generally realized by software.

  However, in the processing performed by the RIP, a dedicated hardware image processing circuit is used rather than software processing by a general-purpose computer such as image processing (for example, color space conversion, rotation, enlargement / reduction) for a bitmap image object. There are also processes that can be executed at a higher speed by using. Therefore, a system has been proposed in which a software RIP module requests image processing from a hardware image processing circuit.

  The computer disclosed in Patent Document 1 has an ability to inquire whether an external image processing circuit has an ability to perform image processing of the drawing element and to perform image processing when performing image processing of the drawing element of the print document. Only in this case, the image processing circuit is requested to perform image processing. If not, the computer itself processes the drawing element by software.

  The system disclosed in Japanese Patent Application Laid-Open No. 2004-228561 generates graphics intermediate language codes for hardware and software by distributing graphic drawing processing for hardware and software when converting a print document into an intermediate language (display list). . Then, the hardware code is supplied to the hardware circuit and the software code is supplied to the software to execute the processes in parallel.

Japanese Patent Laid-Open No. 2002-042121 JP 2006-007496 A

  By the way, with recent multi-core processors, it is conceivable to increase the processing speed by executing software for RIP in parallel by a plurality of processor cores. Here, in a configuration in which the number of hardware image processing devices is smaller than the number of RIP software executed in parallel, even if some of the RIP software being executed in parallel requests image processing from the image processing device. If all image processing devices are already processing other requests, the requests are not processed immediately. Depending on the request processing waiting status of each image processing apparatus, the processing result may be obtained earlier when the RIP software itself performs the image processing than when the image processing apparatus performs the image processing.

  In the present invention, when the software conversion processing means that executes RIP by software requests image processing from the hardware image processing apparatus, the processing result is more than that when the software conversion processing means itself executes the image processing. The purpose is to prevent the situation where it is delayed.

  The invention according to claim 1 is a hardware image processing apparatus that executes specific image processing, and software conversion processing for converting print document data described in a page description language into page image data in a bitmap image format. A plurality of software conversion means for performing a request for processing by the hardware image processing device for the specific image processing of the software conversion processing, and for performing the image processing in response to the request When receiving an instruction to be performed by the hardware conversion means itself, the software conversion means for executing the image processing, and when receiving the request from the software conversion means, the hardware image processing apparatus and the request source If the specific image processing related to the request is processed, the processing result can be obtained quickly. Towards that a printed document conversion apparatus comprising: a control means for performing control to execute the image processing, the.

  According to a second aspect of the present invention, the control means is configured based on performance information of the hardware image processing apparatus and the requesting software conversion means, and a load status of the hardware image processing apparatus. It is determined which of the image processing apparatus and the requesting software conversion means can execute the specific image processing can obtain the processing result of the image processing earlier, and can be obtained earlier. The print document conversion apparatus according to claim 1, wherein control for instructing the image processing to be executed is performed.

  According to a third aspect of the present invention, there is a case where the control unit causes the hardware image processing circuit to execute the specific image processing and a case where the requesting software conversion unit executes the specific image processing. The print document conversion apparatus according to claim 1, wherein when the time until the processing result is obtained is equal, the software conversion unit of the request source executes the image processing.

  According to a fourth aspect of the present invention, the hardware image processing device includes a plurality of unit circuits capable of executing the specific image processing, and the control means is a plurality of unit circuits of the hardware image processing device. If it is possible to obtain the processing result of the specific image processing earlier than the requesting software conversion processing means, the processing result of the specific image processing is obtained from the requesting software conversion processing means. 3. The print document conversion apparatus according to claim 2, wherein the unit circuit obtained earlier is caused to execute the image processing, and if not, the request source software conversion processing unit is caused to execute the image processing. It is.

  The invention according to claim 5 is a plurality of software conversion means for performing a software conversion process for converting print document data described in a page description language into page image data in a bitmap image format. The specific image processing of the software conversion processing is requested to be processed by a hardware image processing apparatus that executes the specific image processing, and the software conversion means itself performs the image processing in response to the request. Software conversion means for executing the image processing when receiving an instruction to perform the image processing, and when receiving the request from the software conversion means, the hardware image processing apparatus and the software conversion of the request source If the specific image processing related to the request is processed, the image can be obtained quickly. Control means for performing control to execute the processing which is a program to function as a.

  According to the invention according to claim 1 or 5, when the software conversion processing means requests the hardware image processing apparatus to perform image processing, the processing result is more than when the software conversion processing means itself executes the image processing. Can be prevented from being delayed.

  According to the second aspect of the present invention, the software conversion processing means and the hardware are obtained using the performance information of the hardware image processing apparatus and the requesting software conversion means and the load status of the hardware image processing apparatus. It can be determined which of the image processing apparatuses can obtain the processing result of the requested image processing earlier.

  According to the third aspect of the present invention, it is possible to reduce the load on the hardware image processing circuit while avoiding delay in the execution of image processing.

  According to the invention of claim 4, even in a configuration in which there are a plurality of unit circuits that execute specific image processing, when the unit circuits are caused to execute the image processing, the software conversion processing means itself executes it. In addition, it is possible to prevent the processing result from being delayed.

It is a figure which shows an example of the hardware constitutions of the printing system by which the apparatus of embodiment is mounted. It is a figure which shows another example of the hardware constitutions of the printing system by which the apparatus of embodiment is mounted. 1 is a functional block diagram illustrating an example of a print data conversion apparatus according to an embodiment. It is a flowchart which shows an example of the process sequence of a process execution control part. It is a functional block diagram which shows another example of the print data converter of embodiment. It is a figure which shows an example of image processing apparatus management information. It is a figure which shows another example of image processing apparatus management information. 6 is a flowchart illustrating an example of a processing procedure of a process execution control unit in the print data conversion apparatus of FIG. 5. It is a figure which shows another example of image processing apparatus management information.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In each figure, the same or similar components are denoted by the same reference numerals.

  First, an example of a system in which the print data conversion apparatus according to the embodiment is mounted will be described with reference to FIG. This system includes a print data conversion apparatus 100, a host computer 140, a print control apparatus 160, and a printer engine 170.

  The print data conversion apparatus 100 print data such as a bitmap image that can be handled by the printer engine 170 using print document data described in a page description language (hereinafter referred to as PDL, which is an acronym for Page Description Language). It is a device to convert to. The host computer 140 sends print document data to the print data conversion apparatus 100 via a network such as a LAN (local area network) 130 and instructs printing of the print document data.

  The print data conversion apparatus 100 includes a controller 110. The controller 110 includes communication control units 112 and 120, a multi CPU 114, a local memory 116, and a secondary storage device 118 that are communicably connected to each other via a local bus 122. The communication control unit 112 is a device that performs data communication with the host computer 140 via the LAN 130. For example, the communication control unit 112 receives a print instruction and print document data from the host computer 140, and exchanges information for print control with the host computer 140. The multi CPU 114 is a processor package having a plurality of CPU cores. The local memory 116 is primary storage used by each CPU core of the multi-CPU 114 as a work area, for example, and is configured by, for example, a random access memory. The secondary storage device 118 is a relatively large-capacity nonvolatile storage device such as a hard disk device or an EEPROM (Electrically Erasable Programmable Read Only Memory), and a program executed by the multi CPU 114 or print document data received from the host computer 140. Memorize etc. A control program describing the processing contents of various functional modules such as the RIP module 210 described later is stored in the secondary storage device 118, for example, and executed by the multi CPU 114. Some of the control programs can be executed in parallel by a plurality of CPU cores of the multi CPU 114. The multi CPU 114 converts the print document data into print image data for each page by executing a control program. The communication control unit 120 is a device that performs data communication with the print control apparatus 160 via a network called an in-system network 150. In this example, the print data conversion apparatus 100 and the print control apparatus 160 (and the printer engine 170) are connected to each other via an intra-system network 150 to constitute one printing system. The print image data of each page generated by the processing of the multi CPU 114 is transmitted to the print control apparatus 160 via the communication control unit 120 and the system internal network 150.

  The image processing device 125 is a device including a hardware image processing circuit dedicated to image processing, such as a graphic accelerator. This hardware image processing circuit has a circuit configuration for executing one or more of various bitmap image processing such as color space conversion, image enlargement / reduction, image rotation, halftone processing, and the like. . For example, the image processing apparatus 125 may be connected to a slot provided on the local bus 122 of the controller 110, for example, as an expansion board.

  The RIP processing program (RIP modules 210A to 210C described later) executed by the multi CPU 114 can request the image processing device 125 to perform image processing and the like, receive the processing result, and use it for creating page image data. However, as will be described later, the RIP modules 210A to 210C themselves can execute the same image processing by software. For this reason, in this embodiment, it is controlled whether the image processing apparatus 125 executes the image processing or the RIP modules 210A to 210C themselves. This control will be described in detail later.

  The printer engine 170 is printing hardware that prints an image represented by print image data on paper using a color material such as ink or toner. The print control device 160 is a device that controls the printer engine 170 and includes a communication control unit 162, a page control unit 164, and an engine control unit 166. The communication control unit 162 communicates with the print data conversion apparatus 100 via the in-system network 150 to receive print image data of each page and exchange various control information necessary for print control. The page control unit 164 accumulates the received print image data of each page and sends it to the engine control unit 166 according to the print order. The engine control unit 166 supplies the received print image data of the page to the printer engine 170. The printer engine 170 prints the received print image data on paper.

  FIG. 2 shows another example of a system in which the apparatus of the embodiment is implemented. In the example of FIG. 1, the image processing device 125 is provided in the print data conversion device 100, whereas in the example of FIG. 2, an image processing circuit 182 similar to the hardware image processing circuit provided in the image processing device 125. Is provided outside the print data conversion apparatus 100. The image processing apparatus 180 is connected to a network 186 to which a communication device 127 included in the print data conversion apparatus 100 is connected. An image processing request is received from the print data conversion apparatus 100 via the network 186, and image processing is performed. Or return the result. The network 186 may be the same network as the intra-system network 150 (or LAN 130), and in this case, the communication device 127 may be the same as the communication control unit 120 (or 112). The image processing apparatus 180 includes a control unit 184 that performs communication via the network 186 and manages requests from the print data conversion apparatus 100. The control unit 184 monitors the load status of the image processing circuit 182, that is, the status of requests currently processed by the image processing circuit 182 and requests waiting for processing (collectively, requests in the queue), and print data. A function of notifying the conversion device 100 may be provided.

  Next, an example of a functional module configuration of the print data conversion apparatus 100 according to the embodiment will be described with reference to FIG. The functional module configuration of the print data conversion apparatus 100 illustrated in FIG. 3 is realized when the multi CPU 114 of the controller 110 executes a control program except for the image processing apparatus 220.

  In FIG. 3, the print document receiving unit 202 receives print document data from the host computer 140. The print document storage unit 204 stores PDL print document data received from the host computer 140. For example, the print document data may be data described in page-independent PDL (for example, PDF) (but not limited to this). “Page independence” means that information describing print image data of a page is represented only by PDL data for the page and does not depend on PDL data for other pages. The print document storage unit 204 may store a plurality of print document data together with information about the print order, for example. A print document selection unit 206 selects a print processing target from print document data stored in the print document storage unit 204. For example, the next print document data is selected as a print target every time the print process being executed is completed according to the print order information.

  The page allocation control unit 208 allocates different pages of the print document data to the RIP modules 210A to 210C (hereinafter collectively referred to as the RIP module 210 when it is not necessary to distinguish them). The method of assignment is not particularly limited. For example, a method of fixedly assigning in order from the first page according to the number of the RIP module 210 may be used, or a method of dynamically assigning pages in order from the RIP module 210 that has completed the page conversion processing may be used. This page allocation may be performed by notifying the RIP module 210 from the page allocation control unit 208, for example, the page number to be allocated.

  The RIP module 210 converts the data of the page portion designated by the page assignment control unit 208 out of the print document data to be printed selected by the print document selection unit 206 into print image data. That is, the RIP module 210 interprets the PDL data of the designated page in order from the top, and draws (rasterizes) it on the page memory, thereby generating print image data of the page. In the case of print document data in the PDF format, it includes index information of PDL data of each page included in the document. Each RIP module 210 refers to the index information in the print document data stored in the print document storage unit 204, and acquires the PDL data of the page corresponding to the page number designated from the page allocation control unit 208. That's fine.

  The RIP module 210 is a software module that interprets print document data described in PDL and converts it into a bitmap (raster) image that can be handled by the printer engine 170. For example, when the RIP module 210 detects a drawing command for drawing an image object such as a font, an image, and a form while interpreting the PDL data of the assigned page in order from the top, the RIP module 210 executes the drawing command. In accordance with the command, the pixel value is written in the storage area of each pixel secured on the local memory 116. In addition, the RIP module 210 performs image processing such as color space conversion, rotation, enlargement / reduction, halftone processing, etc., on the bitmap image formed in this way, according to commands and settings in the PDL data. Has the function to execute. This function is shown as image processing software (abbreviated as SW in the figure) 212A to C in the figure (hereinafter collectively referred to as image processing software 212 when it is not necessary to distinguish). The image processing software 212 is a program that describes one or more of the exemplified color space conversion, rotation, enlargement / reduction, halftone processing, or other various bitmap image processing not shown. .

  Each RIP module 210A-C including the image processing software 212A-C is a device that executes the above-described processing contents by causing each CPU core of the multi-CPU 114 to execute a program group describing the above-described processing contents. Realized. Each of the RIP modules 210A to 210C is executed by the CPU core as a process or a thread. In a typical example, one CPU core executes one RIP module 210. However, the present invention is not limited to this, and one CPU core may execute a plurality of RIP modules 210. In FIG. 3, three RIP modules 210 are executed, but this is only an example.

  In this way, each RIP module 210 can execute image processing with its own image processing software 212. However, since the image processing device 220, which is hardware dedicated to image processing, is usually faster in processing, as a basic operation, if it is necessary to perform image processing in the RIP processing, Request image processing. However, as will be described later, in some cases, the processing may be completed earlier when the requesting RIP module 210 itself performs the image processing than when the hardware image processing device 220 executes the image processing. Therefore, in this embodiment, the request from the RIP module 210 is analyzed by the process execution control unit 230, and the request is processed by a person who can obtain the processing result of the request earlier (details will be described later).

  Thus, in this embodiment, a plurality of pages are RIP processed in parallel by the plurality of RIP modules 210. When the conversion processing of PDL data for one page is completed, each RIP module 210 supplies the print image data of the page formed on the page image memory to the printer engine 170 via the print control device 160.

  Note that the print document selection unit 206, the page allocation control unit 208, and a process execution control unit 230, which will be described later, are executed by any of the CPU cores in the multi CPU 114.

  The image processing device 220 is a device that executes the same image processing as the image processing software 212 by a hardware image processing circuit, and corresponds to the image processing device 125 shown in FIG. 1 or the image processing device 180 shown in FIG. To do. In the example of FIG. 1, it is assumed that the image processing apparatus 220 includes only one hardware image processing circuit capable of executing the same image processing as the image processing software 212.

  The processing execution control unit 230 causes the image processing apparatus 220 to execute the image processing in response to a request for image processing from the image processing apparatus 220 issued by the RIP module 210, or the image processing of the RIP module 210 itself that issued the request. Controls whether the software 212 is executed. In this control, an image processing result that is obtained earlier is selected between the case where it is executed by the image processing software 212 of the RIP module 210 itself and the case where it is executed by the image processing apparatus 220.

  Since the image processing apparatus 220 is a hardware process, the processing speed itself is faster than that of the image processing software 212. However, if there is another request in the queue of the image processing apparatus 220, the new arrival request is made after the processing of the first arrival request. It takes time to process the first request and the new request itself. For this reason, depending on the load status of the image processing device 220, that is, the status of the queue of the image processing device 220, it may occur that the image processing software 212 itself can perform image processing earlier. Therefore, in the present embodiment, the situation of the queue of the image processing apparatus 220 is also taken into consideration, and the case where the processing by the requesting image processing software 212 itself is earlier is determined.

  The queue is a first-in first-out data structure for managing the processing order of requests. For example, identification information for identifying each request is arranged in the order of arrival. In this specification, for the sake of brevity, it is assumed that the request currently being processed by the image processing apparatus 220 is at the head of the queue. That is, the queue includes a request that is currently being processed and a request that is waiting to be processed by the image processing apparatus 220. A request for image processing from each RIP module 210 to the image processing apparatus 220 is added to the tail of the queue associated with the image processing apparatus 220. Then, when the processing for the request that is being executed (this is the head of the current queue) is completed, the image processing apparatus 220 removes the request from the head of the queue and executes the image processing related to the request that has been newly headed. To do. In this way, the method in which the request currently being processed is included at the head of the queue is illustrated by doing so by combining the request currently being processed and the request waiting to be processed as “ This is because it can be collectively referred to as “a group of requests”. That is, this is merely for convenience of explanation, and does not substantially affect the technical contents of the embodiment. In the case of using a queue that does not include a request that is currently being processed, it is only necessary to express the same thing as a “request being processed and a group of requests waiting to be processed”. There is no need to change the control itself.

  The process execution control unit 230 controls the image processing apparatus management information 232 in order to control the allocation of the image processing request outlined above to hardware (image processing apparatus 220) or software (requesting image processing software 212). refer. The image processing apparatus management information 232 is management information that serves as a judgment material for the control. The image processing device management information 232 first includes a performance ratio between the image processing device 220 and the image processing software 212 or information that is a material for calculating the performance ratio. “Performance” here is performance that represents the time required to process a certain amount of image data, in other words, the processing speed. The information used as the material for calculating the performance ratio is, for example, a numerical value representing the performance of each of the image processing device 220 and the image processing software 212. The performance ratio is calculated by taking the ratio of these numerical values. These performance ratios or information on the calculation materials are obtained by an experiment in which the image processing device 220 and the image processing software 212 process test images and measure the processing time, respectively. You can register with. Here, for the sake of simplicity, it is assumed that the performance of the image processing software 212 of each RIP module 210 is the same, but if not, the image processing software 212 and the image processing apparatus are provided for each RIP module 210. Information on the performance ratio to 220 (or information as a material for obtaining the performance ratio) may be incorporated in the image processing apparatus management information 232.

  Further, the image processing device management information 232 is used as a judgment material for the time until processing of all the requests in the queue of the image processing device 220 is completed (hereinafter, referred to as “processing time for first-come-first-served request”). Contains information. For example, there are the following three examples of information that can be used to determine the time required for processing the first-come-first-served request (however, the present invention is not limited to this).

(1) Number of requests in the queue This example is based on a model in which the image processing apparatus 220 processes any request in a certain time. Actually, the time required for the image processing apparatus 220 to process the request varies depending on various parameters such as the size (number of pixels) of the image that is the target of the image processing related to the request. A certain degree of accuracy is achieved even with the assumption that is constant for all requests. For example, when the image processing requested to the image processing apparatus 220 is limited to processing of the entire page (for example, color space conversion for the page), the number of pixels to be processed is the same for requests regarding any page. The time required for processing is also almost constant.

(2) Total number of pixels of target images in each queue The RIP module 210 performs image processing on each object in the PDL print document data if necessary to perform image processing on the object. The number of pixels of the object is the number of pixels related to the request. The total number of pixels in the queue is the sum of the number of pixels related to each request in the queue. When the image processing performed by the image processing device 220 (and the image processing software 212) is processing for a bitmap image, the time required for the image processing is roughly proportional to the number of pixels. Accordingly, the number of pixels of the object related to the request can be used as an index value of the time required for image processing of the object, and the total number of pixels related to each request in the queue is calculated as the time required for processing the first-arrival request. Can be used as an index value. In many cases, the number of pixels of the object related to the request is included in the print document data as attribute data of the object.

(3) Total estimated value of processing time required for each request target image in the queue In the above (2), the processing time required for the image processing device 220 (and the image processing software 212) is the number of pixels to be processed. Whereas (3) assumes that there are factors that are not proportional to the “number of pixels” so that they cannot be ignored. For example, in the case of image processing including enlargement / reduction of an image, the number of pixels changes, so the time required may not be proportional to the number of pixels. In addition, there may be image processing whose required time is not proportional to the number of pixels. Even in such a case, a calculation formula for estimating the time required for the processing from the number of pixels subject to image processing can be obtained by experiments or the like. Time can also be estimated. For each request placed in the queue, the required time is estimated from the number of target pixels of the request, thereby obtaining an estimated value of the total processing time required for all requests in the queue.

  Note that information regarding the processing time required for the first-arrival request in the image processing apparatus management information 232 is, for example, executed according to the execution status of the processing of the image processing apparatus 220 or the arrival status of requests from the RIP modules 210. The control unit 230 updates information as needed. For example, in the case of the above example (1), when the notification that the image processing being executed is completed is received from the image processing device 220, the processing execution control unit 230, for example, in the queue in the image processing device management information 232. Decrease the number of requests by one. In the case of the above examples (2) and (3), when the image processing completion notification is received from the image processing device 220, the number of pixels or the estimated required time of the request at the head of the queue (request that has completed the current processing) is obtained. Then, the total number of pixels in the image processing apparatus management information 232 or the total value of the estimated required time is subtracted. In addition, when a request from the RIP module 210 is added to the image processing apparatus 220, 1 is added to the number of requests in the image processing apparatus management information 232, or the total number of pixels or the estimated required time is calculated. Add the number of pixels and estimated time required for the request added this time.

  The image processing apparatus management information 232 has been described above. Next, an example of a processing procedure of the processing execution control unit 230 using this management information will be described with reference to FIG.

  The process execution control unit 230 waits for an image processing request from the RIP modules 210A to 210C (S10). When the request arrives, the image processing device management information 232 is referred to (S12), and the image relating to the request is obtained for each of the hardware (the image processing device 220) and the software (the image processing software 212 of the requesting RIP module 210). The time required to complete the process is estimated (S14). In this estimation, it is only necessary to compare whether hardware or software is completed earlier, so it is not necessary to estimate “time” in terms of minutes and seconds, and it is only necessary to estimate a numerical value that can be compared on the same scale.

  For example, in the example in which the number of requests in the queue is managed as described in (1) above as information on the time required for processing the first request, it is assumed that the time required for processing one request is constant. Therefore, the number of requests may be used as an index value for the required time. In this case, the index value of the required time until the processing of the current request is completed in the image processing apparatus 220 (hardware) is n + 1 when the number of requests in the queue is n. Here, “+1” corresponds to the time required for this request. In addition, the request that is currently being executed should have already been processed halfway, but in this example, to simplify the process, it is not possible to increase the accuracy by estimating how much it has been completed. However, it is simply estimated that the time required for one request is required. On the other hand, the time required for the requesting image processing software 212 to process the request may be estimated from the performance ratio between the image processing software 212 and the image processing apparatus 220. For example, if the performance (processing speed) of the image processing device 220 is P times that of the requesting image processing software 212 (that is, the performance ratio of the device 220 to the software 212 is P), the image processing software 212 Only the image processing related to the request is executed (there is no queue), and it takes P times as long as that of the image processing apparatus 220 for each request, so the index value of the required time is eventually P.

  When the required time for software and hardware is estimated in this way, the required times are then compared to determine which of the hardware and software can obtain the processing result first (S16). In the above example (1), the estimated time index value n + 1 (hard) and P (soft) are compared. The smaller the index value, the faster the processing result can be obtained. For example, if P is smaller, the requesting image processing software 212 is faster, and vice versa, the image processing device 220 is faster. The determination in S16 may be regarded as determining whether the number of requests in the queue is less than the performance ratio P. If the number of requests in the queue is less than P, even if this request is added to this, the index value is smaller than that of the software.

  If it is determined in S16 that the hardware is earlier, the process execution control unit 230 adds the current request to the queue (queue) of the hardware (image processing apparatus 220) (S18). Accordingly, the information related to the processing time required for the first-arrival request in the image processing apparatus management information 232 is updated. On the other hand, if it is determined in S16 that the software is earlier, the software (requesting RIP module 210) is instructed to execute image processing related to the request (S20). If the estimated index values of the required time for hardware and software are the same, image processing is assigned to software. In the case of the requesting RIP module 210, there is no penalty if the required time is the same, and on the other hand, for the other RIP modules 210, the load (the number of processing waits) of the image processing apparatus 220 is high. Since the number does not increase, the possibility that the image processing apparatus 220 can process the request issued in the future is increased.

  For example, the image processing apparatus 220 has twice the performance of the image processing software 212 of the RIP module 210, and as illustrated in FIG. 3, the image processing apparatus 220 is processing the request of the RIP module 210A, and the RIP Assume that the RIP module 210C makes a request for image processing when the request for the module 210B is waiting for processing in the queue. In this case, the number of requests in the queue is 2, and since this value is 2 or more, the request for this RIP module 210C is assigned to the RIP module 210C itself.

  The requesting RIP module 210 enters a waiting state waiting for the result of the request at the time of issuing the current request to the process execution control unit 230. When the request is assigned to the image processing device 220 by the processing execution control unit 230, when the image processing device 220 completes the image processing related to the request, the result of the image processing is returned to the requesting RIP module 210. . The requesting RIP module 210 accepts the result of the image processing and advances the RIP processing further. On the other hand, when the processing execution control unit 230 assigns the request to the requesting RIP module 210 itself, the RIP module 210 that has received the assignment instruction processes the image that is the target of the request in its image processing software 212. Let When the processing result is obtained, the RIP process is further advanced using the processing result.

In the above, the example of (1) above (an example in which the number of requests is used as the index value of the required time) has been described. The processing is the same as in the case of (1) except for the case of 1). In the case of (2) (when the number of pixels is used as an index value for the required time), at the time of performing the process of S14, the total number of pixels N _A of processing wait requests in the queue included in the image processing device management information 232 If it is assumed that the number of requested pixels N _B has arrived this time, the time required for the image processing apparatus 220 to complete the processing of the current request is approximately proportional to N _A + N _B. Therefore, N _A + N _B is used as an index value of the required time for hardware. Meanwhile, the time the requesting RIP module 210 (the image processing software 212) is required until completion of the processing of this request, we can use the performance ratio P, which is described as P × N _B. Therefore, in S16, the magnitude relationship between N _A + N _B (hard) and P × N _B (soft) may be determined. If the former is small, the current request is assigned to hardware, and if the latter is equal to or less than the former, the current request is assigned to software. For (3), the same idea as in (2) may be used.

  Next, a modified example will be described with reference to FIGS. In the example of FIG. 3, the image processing apparatus 220 has one hardware image processing circuit, but in this modification, the image processing apparatus 220 includes two hardware image processing circuits 222 </ b> A and 222 </ b> B. Each of the image processing circuits 222A and 222B is, for example, a graphic accelerator chip, and the image processing apparatus 220 is, for example, a graphic board that has two such chips and is mounted on a connector provided on the bus of the controller 110. However, this is merely an example, and the image processing apparatus 220 may be a remote apparatus that communicates with the print data conversion apparatus 100 via a network, for example. Although three RIP modules 210 are shown in the figure, it is assumed that more RIP modules 210 are actually executed by the multi CPU 114 (for example, six or more in the setting of FIG. 6A).

  In this case, for example, the process execution control unit 230 manages each queue of the image processing circuits 222A and 222B. For example, as shown in FIG. 6A, the image processing device management information 232 stores the performance ratio with respect to SW (software) and the number of requests in a queue (queue) for each of the image processing circuits 222A and 222B. Yes. Here, it is assumed that the identification numbers of the image processing circuits 222A and 222B are “1” and “2”, respectively. The SW performance ratio is the ratio of the performance of the image processing circuit 222A or 222B to the performance (processing speed) of the image processing software 212. For example, the image processing circuit 222B with the identification number 2 is three times as large as the image processing software 212. It has shown that it has processing speed of. If the performance of the image processing circuits 222A and 222B is the same, there is no need to store them separately. The number of requests in the queue is updated according to the addition (when accepting requests) or removal (when processing is completed) of requests to the queue. The example of FIG. 6A corresponds to the above-described case (1).

  Further, in the example corresponding to the above (2), the image processing device management information 232 includes, as shown in FIG. 6B, the SW performance ratio and the queue (queue) for each of the image processing circuits 222A and 222B. A list of the number of pixels of each request is stored. In the example of FIG. 6B, the requests of the request identification numbers 2, 3, and 5 are placed in the queue of the image processing circuit 222B of the identification number 2, and the target of the image processing related to the requests 2, 3, and 5 is included. The number of pixels is N2, N3, and N5, respectively. The total number of pixels N2, N3, and N5 for each request is the total number of pixels in the queue. When the image processing circuit 222B completes the image processing related to the request 2 at the head of the queue, the pixel number N2 is deleted from the information on the number of requested pixels in the queue. In the case of (3) above, the same information may be registered and managed in the image processing apparatus management information 232.

  FIG. 7 illustrates an example of a processing procedure of the processing execution control unit 230 according to this modification. In this procedure, steps having the same processing contents as the steps in the procedure of FIG. 4 are denoted by the same reference numerals and description thereof is omitted.

  In the procedure of FIG. 7, when the processing execution control unit 230 receives an image processing request from any of the RIP modules 210, the processing execution control unit 230 refers to the image processing device management information 232 until the image processing related to the request is completed. A time index value is calculated for each of the image processing circuits 222A and 222B, and a time index value until completion when the requesting RIP module 210 executes the image processing is calculated (S14a). The method of calculating the index value of the time until completion of image processing for each of the image processing circuits 222A and 222B may be the same as in the above embodiment. Then, as a result of comparing the calculated index values for the required time, the request is assigned to the one that completes processing the earliest. For example, if it is the image processing circuit 222A that completes the processing of the request most quickly, the request is added to the end of the queue of the image processing circuit 222A (S18a). If it is the RIP module 210 of the request source that completes the processing of the request, the module 210 is instructed to execute the request (S20).

  Next, still another modification will be described with reference to FIG. In the example of FIG. 8, an image processing apparatus 240 equipped with hardware image processing circuits 242A and 242B (for example, a graphic accelerator) is connected to the print data conversion apparatus 100 via a network such as a LAN. The image processing apparatus 240 includes an image processing control unit 244 that controls the image processing circuits 242A and 242B. The image processing control unit 244 communicates with the processing execution control unit 230 of the print data conversion apparatus 100 via the network to receive an image processing request, or request image data of image processing results of the image processing circuits 242A and 242B. To the RIP module 210. The image processing control unit 244 manages the queues of the image processing circuits 242A and 242B. In addition, the image processing control unit 244 responds to an inquiry (for example, periodically) from the processing execution control unit 230 or whenever the contents of the queues of the image processing circuits 242A and 242B change. , Information about the queues of the image processing circuits 242A and 242B (for example, a list of requests in each queue) is provided. Based on this information, the process execution control unit 230 updates information on the number of requests or the total number of pixels in the queue of each of the image processing circuits 242A and 242B in the image processing apparatus management information 232. The process execution control unit 230 refers to the management information 232 and determines whether the request from each RIP module 210 is assigned to the image processing circuit 242A or 242B or the requesting RIP module 210 as described above. To do.

  The image processing control unit 244 is realized, for example, when a CPU (not shown) of the image processing apparatus 240 executes a program describing the above-described functions.

  Note that the image processing device management information 232 is provided in the remote image processing device 240, and the processing execution control unit 230 requests the image processing control unit 244 of the image processing device 240 in the queues of the image processing circuits 242A and 242B. You may make it inquire about the number.

  The embodiments and modifications described above are merely examples for explaining the present invention, and various modifications and improvements can be made within the scope of the present invention. For example, in each of the above examples, the conversion process is assigned to the RIP module 210 in units of pages, but the unit of assignment is not limited to pages, and may be units other than pages such as bands.

  100 Print Data Conversion Device, 110 Controller, 140 Host Computer, 160 Print Control Device, 170 Printer Engine, 202 Print Document Receiving Unit, 204 Print Document Storage Unit, 206 Print Document Selection Unit, 208 Page Assignment Control Unit, 210A, 210B, 210C RIP module, 212A, 212B, 212C image processing software, 220 image processing device, 230 processing execution control unit, 232 image processing device management information.

Claims (5)

A hardware image processing device for performing specific image processing;
A plurality of software conversion means for performing software conversion processing for converting print document data described in a page description language into page image data in a bitmap image format, wherein the specific of the software conversion processing The image processing is requested by the hardware image processing apparatus, and when an instruction to perform the image processing by the software converting unit itself is received in response to the request, the software for executing the image processing is executed. Wear conversion means;
When the request is received from the software conversion means, a processing result is obtained when the specific image processing related to the request is processed among the hardware image processing apparatus and the requesting software conversion means. Control means for performing control to execute the image processing for those who can quickly obtain
A print document conversion apparatus comprising: The control means, based on the performance information of the hardware image processing apparatus and the requesting software conversion means, and the load status of the hardware image processing apparatus, the hardware image processing apparatus and the request source. It is determined which one of the software conversion means and the specific image processing to execute to obtain the image processing result earlier, and the image processing to the one determined to be obtained earlier is executed. Control to direct,
The printed document conversion apparatus according to claim 1, wherein:   The control means is a time until the processing result of the image processing is obtained when the specific image processing is executed by the hardware image processing circuit and when the requested software conversion means is executed. 2. The print document conversion apparatus according to claim 1, wherein if the two are equal, the software conversion unit of the request source executes the image processing. The hardware image processing apparatus includes a plurality of unit circuits capable of executing the specific image processing,
If the control means is one that can obtain the processing result of the specific image processing earlier than the requesting software conversion processing means in the plurality of unit circuits of the hardware image processing apparatus, The unit circuit that can obtain the processing result of the specific image processing earlier than the requesting software conversion processing unit executes the image processing, and if not, the requesting software conversion processing unit causes the image processing to be performed. To execute,
The printed document conversion apparatus according to claim 2, wherein Computer
A plurality of software conversion means for performing software conversion processing for converting print document data described in a page description language into page image data in a bitmap image format. For image processing, if a request is made by the hardware image processing apparatus that executes the specific image processing, and the software conversion means itself receives an instruction to perform the image processing in response to the request. Software conversion means for executing the image processing;
When the request is received from the software conversion means, a processing result is obtained when the specific image processing related to the request is processed among the hardware image processing apparatus and the requesting software conversion means. Control means for performing control to execute the image processing for those who can obtain the
Program to function as.

Images