JP4420858B2 - Method and computer program for providing automatic language switching - Google Patents

Description

  The present invention relates to image rendering. In particular, the present invention relates to a method and a computer program for providing automatic language switching capable of switching languages by accurately recognizing the language format of an input image data stream when there is no language switch for switching languages.

  Currently, image devices and many raster-only devices that support only a single language format, such as raster data specific to the device's print engine, are commercially available. In such a device, the image data is printed by a print data interpreter without a prior check. If the input is in the correct language format, all input is recognized and the job is output. If some or all of the input is not recognized in the correct language format, the job will not be output. FIG. 1 illustrates a single language supported printer technology.

  Multi-language supported image device technology requires an explicit language switch preceding the image data. In multi-language supported image device technology, some command or data is read into the image device in a known or predetermined format that specifies the language format of subsequent image data. For example, there are image devices that support explicit language switching by supporting the printer job language (PJL) command @PJL ENTER LANGUAGE = <language>. If the specified language is supported by the device, the image data is based on the corresponding interpreter. If it is not supported, the input is rejected and the job is not output. FIG. 2 illustrates a multi-language supported printer technology.

  The use of an explicit language switch may result in different types of image data generators not meeting the single specification for building an image job, but otherwise generating image data compatible with that device Limited to various input environments. For example, an image device manufacturer might supply a device driver for an environment that has an explicit switching specification, but that device generates image data without using an explicit language switch. It can be used in other environments where it may be.

  US Pat. No. 5,960,113 discloses the following technique. That is, the first byte sample of the input stream is pre-read and sent to the language recognition process. For each supported language, language-specific recognition tests such as looking for special signatures in byte samples are provided. The technology applies each language recognition test in a predetermined order until the language format is recognized. If the language is thus recognized, the input stream is processed by the corresponding language interpreter.

  The technique described above still has a limited element in that it may not necessarily be efficient in language recognition since the application order of language recognition devices is determined in advance. For example, in some environments, the majority of input to the device may be language A, but the associated language recognizer that tests this may be last. Thus, in such an environment, resources are first consumed for languages that may not be recognized. FIG. 3 illustrates this type of technique.

  Another technique is disclosed in US Pat. No. 5,467,432. In this technique, the first byte sample of the input stream is pre-read. The read byte samples are compared to a language recognizer group (eg, one for each supported language). Each language recognizer looks for a mode decision code that is suggestive but does not necessarily determine that it is part of the language. For each mode decision code, a likelihood probability is assigned that is an indicator of the language. The process is completed by summing the probabilities of all decision codes found in the sample by language. The process then selects the language with the highest total probability.

  This technique is also limited in that it only estimates the correct language and tends to select the wrong language. Second, the performance of the technique is inefficient in that all languages must be tested before the selection is performed. FIG. 4 illustrates this type of technique.

  Thus, while there are technologies currently used to provide language switching, there are still challenges. Therefore, augmenting or replacing current technology with other technologies would improve the technology.

  The present invention relates to an image rendering process. In particular, the present invention relates to a method and computer system product that provides automatic language switching that can accurately recognize the language format of an input image data stream even in the absence of a language switch.

  When the multilingual support type language device receives image data without displaying the language format, the image device samples the image data to determine the language format. The present invention uses an efficient method for accurately recognizing the language format (eg, PDL, other languages, etc.) of the image data stream in the absence of an explicit language switch (eg, automatic language switching). Runs on an image rendering system.

  In at least some embodiments, the sniffer process is initiated when the imaging device receives an input image data stream (eg, a print job). The sniffer process analyzes the first sample of the input stream to determine the language format (eg, PDL or other language). The sniffer does this by enumerating a set of language recognizers implemented as callback functions. Each language recognizer has a sample size associated with it. The sniffer process pre-reads the sample size used by each ordered language recognizer and invokes the corresponding callback function with this byte sample. The ordering continues until the language recognizer recognizes recognition of the language format or until all language recognizers in the language recognizer set are selected.

  Ordering uses a dynamic heuristics-based approach and selects the order in which the language recognizers are tried. A heuristic-based approach involves storing the cumulative number of times a language has been recognized. When a language is recognized, the cumulative number of times of the corresponding language recognition device is incremented. First (eg, initial power up), the language recognizers are ordered in a random order. If the language recognition is successful at least once, the language recognition devices are ordered in priority order according to the cumulative number of times the language has been recognized (for example, devices with increased number of language recognitions move higher). If two or more language recognizers have the same cumulative value, the priority order within the same group is still random.

  Thus, the device becomes more efficient with respect to the particular environment in which it is operating, since the average number of language recognizers applied before the language is recognized is reduced.

  These and other features and advantages of the invention will be set forth or will become more fully apparent in the description that follows and in the appended claims. The above features and advantages can be realized and attained by means of the devices and combinations particularly pointed out in the appended claims. Furthermore, the features and advantages of the present invention may be learned by practice of the present invention or will become apparent from the description provided hereinafter.

  For the purpose of achieving the above method and other features and advantages of the present invention, the present invention will be described more specifically with reference to specific embodiments thereof illustrated in the accompanying drawings. With the understanding that the drawings depict only typical embodiments of the invention and therefore should not be construed as limiting the scope of the invention, the accompanying drawings will be used to illustrate the invention. More specifically and in detail.

  The present invention relates to image rendering. In particular, the present invention relates to a method and a computer program for accurately recognizing the language format of an input image data stream even when no language switch exists.

  When the multilingual support type language device receives image data without displaying the language format, the image device samples the image data to determine the language format. This process is commonly referred to as automatic language switching. Embodiments of the present invention use an efficient method for accurately recognizing the language format (eg, PDL or other language) of the image data stream in the absence of an explicit language switch (eg, automatic language switching). An image rendering system is disclosed.

  The following disclosure of the present invention is divided into two subheadings: “Exemplary Operating Environment” and “Provision of Language Format Recognition”. The use of the above subheadings is for the convenience of the reader only and should not be construed as limiting in any way.

(Exemplary Operating Environment)
FIG. 5 and the corresponding discussion are intended to outline a suitable operating environment in which the present invention may be implemented. Those skilled in the art will recognize that the present invention may be implemented with one or more computing devices and in various system configurations, including networked configurations.

  Embodiments of the invention include one or more computer-readable recording media, each recording medium including or configured to include data or computer-executable instructions for manipulating data on the medium. The Computer-executable instructions are those associated with a general purpose computer capable of performing a variety of different functions that can be accessed by data structures, objects, programs, routines, or processing systems, or a limited number Other program modules, such as those associated with a dedicated computer capable of performing these functions. Computer-executable instructions, which cause a processing system to perform a specific function or group of functions, are examples of program code means for implementing the method steps disclosed herein. In addition, the special sequence of executable instructions provides examples of corresponding actions that may be used to implement such steps. Examples of computer readable recording media include random-access memory (“RAM”), read only memory (“ROM”), programmable read only memory (“PROM”), erasable programmable read only memory ( “EPROM”), electrically erasable programmable read-only memory (“EEPROM”), compact disk read-only memory (“CD-ROM”), or data or execution that can be accessed by a processing system Including any other device or component capable of supplying possible instructions.

  With reference to FIG. 5, an exemplary system for implementing the invention includes a computing device 10 which may be a general purpose or special purpose computer. For example, the computer device 10 may be a personal computer, notebook computer, personal digital assistant (“PDA”) or other handheld device, workstation, minicomputer, mainframe, supercomputer, multiprocessor system, It may be a network computer, a processor-based consumer electronic device, or the like.

  The computing device 10 includes a system bus 12 that can be configured to connect its various components and allows data to be exchanged between two or more components. The system bus 12 may include one of a variety of bus structures including a memory bus or memory controller, a peripheral bus, or a local bus using any of a variety of bus architectures. . Typical components connected by system bus 12 include processing system 14 and memory 16. Other components may include one or more mass storage device interfaces 18, input interfaces 20, output interfaces 22 and / or network interfaces 24, each of which will be discussed below.

  The processing system 14 includes one or more processors, such as a central processor, and optionally one or more other processors designed to perform special functions or tasks. Instructions supplied on a computer-readable medium such as memory 16, magnetic hard disk, removable magnetic disk, magnetic cassette, optical disk, or instructions from a communication connection that can also be considered a computer-readable medium It is the processing system 14 that typically does.

  Memory 16 may contain or be configured to include data or instructions for manipulating data on the medium and can be accessed by processing system 14 via system bus 12. One or more computer-readable media are included. The memory 16 may include, for example, a ROM 28 used to store information permanently and / or a RAM 30 used to temporarily store information. ROM 28 may include a basic input / output system (“BIOS”) having one or more routines used to establish communication, such as during startup of computer device 10. The RAM 30 may include one or more program modules, such as one or more operating systems, application programs, and / or program data.

  One or more mass storage device interfaces 18 may be used to connect one or more mass storage devices 26 to the system bus 12. Mass storage device 26 can be incorporated into or a peripheral device of computer device 10 and allows computer device 10 to hold large amounts of data. Optionally, one or more of the mass storage devices 26 can be removed from the computer device 10. Examples of mass storage devices include hard disk drives, magnetic disk drives, tape drives, and optical disk drives. The mass storage device 26 can read from and / or write to a magnetic hard disk, a removable magnetic disk, a magnetic cassette, an optical disk or another computer readable medium. Mass storage device 26 and its corresponding computer-readable medium may include one or more program modules, such as an operating system, one or more application programs, other program modules or program data. Provides non-volatile storage of certain data and / or executable instructions. Such executable instructions are examples of program code means for implementing the method steps disclosed herein.

  The one or more input interfaces 20 can be used to allow a user to input data and / or instructions to the computer device 10 via one or more corresponding input devices 32. Examples of such input devices include keyboards and mice, trackballs, light pens, styluses, or other pointing devices, microphones, joysticks, game pads, satellite dishes, scanners, camcorders, digital cameras and these Alternative input devices such as those similar to are included. Similarly, examples of input interface 20 that can be used to connect input device 32 to system bus 12 include serial ports, parallel ports, game ports, universal serial bus (“USB”). Firewire (IEEE 1394) or another interface is included.

  One or more output interfaces 22 may be used to connect one or more output devices 34 to the system bus 12. Examples of output devices include monitors or display screens, speakers, printers and the like. A particular output device 34 may be integrated into the computer device 10 or it may be a peripheral device. Examples of output interfaces include video adapters, audio adapters, parallel ports, and the like.

  The one or more network interfaces 24 may include wired and / or wireless links with one or more other local or remote computer devices that the computer device 10 has shown as the computer device 36. Information can be exchanged via a network 38. Examples of network interfaces include a network adapter for connecting to a local area network (“LAN”) or modem, a wireless link or other adapter for connecting to a wide area network (“WAN”) such as the Internet. Is included. The network interface 24 may be incorporated into the computer device 10 or may be a peripheral device thereof. In a networked system, accessible program modules or portions thereof can be stored in a remote memory storage device. Further, in networked systems, computing device 10 may participate in a distributed computing environment where functions or tasks are performed by multiple networked computer systems.

  Accordingly, those skilled in the art will recognize that embodiments of the present invention may be implemented in a variety of different environments having many types of system configurations, although FIG. It presents a representative networked configuration that could be used as FIG. 6 illustrates an embodiment that includes one client, three imaging devices, and one optional print server connected to the network, but alternative embodiments include two or more clients, two or less. Image devices, four or more image devices, no servers, and / or two or more servers connected to a network. Yet another embodiment of the present invention provides a local, networked or peer-to-peer image in which one or more computer devices are connected to multiple image devices for rendering an imaging job. Including computerized environment. Some embodiments include a wireless network environment or include an environment where the network is a wide area network such as the Internet.

  The exemplary system of FIG. 6 includes a computing device, shown as client 40, connected to a plurality of imaging devices 50-54 through network 56. Image devices 50-54 in FIG. 6 can be any type of image device that may be used to render an imaging job. Examples include printers, multifunctional peripherals (“MFPs”), scanners, facsimile devices, electronic whiteboards, display monitors, tablet PCs, document servers, and others. Although the image devices 50-54 are connected to the network 56, embodiments of the present invention may be image devices that are locally connected to computer devices, image devices configured in a peer-to-peer imaging environment, or Includes the use of imaging devices configured in a wireless network environment.

  In the illustrated embodiment, client 40 includes a software application 42, one or more print drivers 44, a port manager 46, a spooler 48 and a print processor 49. Optionally, a server 60 having, for example, one or more print queues 62, one or more printer drivers 64, a port manager 66, a spooler 68 and a print processor 69 is included.

  Thus, according to the illustrated embodiment of the present invention and other embodiments, the language format of the input image data stream is recognized even in the absence of an explicit language switch.

(Providing language format recognition)
When the multilingual supported language device receives image data without an explicit indication of the language format, the image device samples the image data to determine the language format. Embodiments of the present invention provide an efficient method for accurately recognizing the language format (eg, PDL or other language) of an image data stream in the absence of an explicit language switch (eg, automatic language switching). Disclosed is an image rendering system used.

  Referring now to FIG. 7, a flowchart illustrating an exemplary embodiment according to the present invention is presented. In FIG. 7, if the image device receives an input image data stream (eg, a print job) at step 70, a sniffer process is initiated. The sniffer process analyzes the first sample of the input stream and determines the language format (eg, PDL or other language) by the initial ordering (step 72) of the language recognizer set implemented as a callback function.

  In at least some embodiments, the ordering selects an order in which to try the language recognizer using a dynamic heuristic-based approach. For example, a rule-of-thumb approach in certain embodiments includes storing a cumulative value of the number of times a language has been recognized. When a language is recognized, the cumulative number of times of the corresponding language recognition device is incremented. First (eg, first power up), the language recognizers are ordered in a random order. If the language recognition is successful at least once, the language recognition devices are ranked in the priority order according to the number of times the language is recognized (for example, the order of the language recognition devices with the increased number of language recognition moves higher). If two or more language recognizers have the same cumulative value, the priority order within the same counting group is still random. Thus, the device becomes more efficient with respect to the particular environment in which it is operating, since the average number of language recognizers applied before the language is recognized is reduced.

  In another embodiment, the device maintains a separate set of rules of thumb that are applied according to environmental factors. Examples of such environmental factors include the location of the device (eg, the device moves), the locality (eg, IP address) of the user (eg, department), or the user itself (eg, specific to the user history).

  Referring again to FIG. 7, each language recognizer has a corresponding sample size. For each ordered language recognizer, the sniffer process pre-reads or pokes the corresponding sample size at step 74 and invokes the corresponding callback function with byte samples at step 76. Next, at decision block 78, a determination is made as to whether the language recognizer has recognized the language format. If it is determined at decision block 78 that the language recognizer has recognized the language format, the control process ends the ordering process, updates the selection heuristic at step 80, and starts or loads the corresponding language interpreter at step 82. , Bind the byte stream to the corresponding language interpreter at step 84 and render the job at step 86.

  Returning to decision block 78 and referencing, if it is determined that the language recognizer did not recognize the language format, execution proceeds to decision block 88 where more language recognizers are present or used in the set. Determine whether it is possible. If it is determined at decision block 88 that another language recognizer exists or is available, execution returns to step 72 to order another language recognizer, and at step 74 the corresponding sample size is pre- Read or poke and call the corresponding callback function with byte samples in step 76.

  Alternatively, if it is determined at decision block 88 that there is no other language recognition device in the set, execution proceeds to step 90 where the user is notified. Thus, as shown in FIG. 7, the ordering continues until the language recognizer recognizes the recognition of the language format or until the determination is completed on all language recognizers in the language recognizer set.

  Referring now to FIG. 8, an exemplary embodiment for providing dynamic determination of the next language recognition device to select is shown. The sniffer in FIG. 8 comprises a control process / component, a language recognizer selector component, a language recognizer, and a selection heuristic.

  The control process is the entry point to the sniffer, starting other subprocesses and making control decisions based on the results. When an image data byte stream (shown as “print job” in FIG. 8) is received, the control process orders until the language is recognized or until all language recognition devices determine the language. . The ordering is handled by the language recognizer selector.

  The language recognizer selector selects the order of the recognizers using a selection rule of thumb (for example, past consistent cumulative values). Each time a language recognizer is selected, the process returns to the control process of the recognizer and the location of the recognizer's corresponding byte size sample (eg, function callback address).

  The control process then pre-reads or pokes the byte stream for a particular byte sample size and calls a callback function with the byte sample as an argument. The language recognizer then returns whether the language has been recognized (eg, a Boolean return result).

  If the language recognizer operates successfully, the control process ends the ordering process, updates the selection heuristics, starts a byte stream and binds it to the corresponding language interpreter.

  Referring now to FIG. 9, an exemplary pre-read function for dynamically determining the sample size by the currently selected language recognizer is shown. In FIG. 9, each time the control process orders a language recognizer, the control process sends the byte sample size corresponding to that language recognizer to the pre-read subprocess. The pre-read subprocess pokes the byte stream for its byte sample size (eg, does not consume data and the byte stream offset does not progress).

  In this embodiment, the subprocess uses a cache to improve performance. The cache is used to store byte samples. The cache is initially empty and a copy of the first pork is stored in the cache. On subsequent pokes, if the sample size is less than or equal to the cache size, the byte sample is retrieved from the cache. Otherwise, byte samples are pre-read from the byte stream and the cache is updated with new (eg, larger) byte samples.

  Referring now to FIG. 10, an exemplary embodiment for sequentially testing a dynamically selected language recognizer until it is recognized or until all language verifications are complete is shown. The control process in FIG. 10 starts the selected language recognizer by calling the corresponding callback function and sending a byte sample as an argument. If the language recognizer does not operate successfully, the control process determines whether there are more language recognizers by invoking the language recognizer selection process to order the next language recognizer. If the determination is completed without matching the language recognition device, the job is not output.

The language recognizer can use any method such as looking for a language-specific special signature (eg, magic number) to determine the presence or absence of a language within a byte sample. As an example, here is a unique byte sequence that appears at the beginning of the following languages:
Language Byte sequence PJL @PJL
PS%! PS
PDF% PDF
TIFF II * <NUL> or MM <NUL> *

  The method used for each language recognizer may vary from recognizer to recognizer, and the sniffer process itself is not aware of this.

  Referring to FIG. 11, an exemplary embodiment for starting a language interpreter and updating heuristics after recognition is shown. In FIG. 11, the order in which the language recognition devices are listed is dynamically determined based on the empirical rule count of the past history of each language recognition device in order to successfully recognize the corresponding language format. The process can use any heuristic counting method, such as an increment of the degree of successful matching.

  The language recognizers are then dynamically classified according to their rule of thumb counts and ordered from highest value to lowest value. When a plurality of language recognition devices exist in the same group, the order in the group is selected at random.

  In some embodiments, the order in which languages are tested sequentially is not predetermined by upstream processes. Instead, this is determined dynamically by the same process of matching. Illustratively, this decision is random, based on past history, or both.

  When a language is recognized, the control process loads the corresponding language interpreter and binds the byte stream (ie, print data) to the loaded language interpreter. The control process also updates the selection rule of thumb to enhance the preferential selection of the corresponding language recognizer.

  Although the method and process according to the present invention have proven particularly useful in connection with the rendering of print jobs having page description language (PDL), those skilled in the art will recognize that the method and process can be used in a variety of different applications It will be appreciated that it can be used in conjunction with a variety of different types of languages. For example, the method and process according to the present invention may be processed in whole or in part by an imaging device that captures all or part of its input in electronic formats (eg, soft copy) such as fax, scan, copy, document management process, etc Or a non-electronic format (eg hard copy) is converted to an electronic format as part of an input process (eg scan to OCR) and the input or converted language format is a natural language (not a rendering language) Or any imaging operation that is to be translated.

  At least some embodiments of the present invention include medical and / or geological image analysis devices that capture x-ray, MRI, sonar and other inputs. Further embodiments include AV devices where the input is in audio / visual format. Further, some embodiments include natural language devices whose input includes words that are spoken and / or written.

  Thus, as discussed herein, embodiments of the invention relate to image rendering. In particular, the present invention relates to a system and method for accurately recognizing the language format of an input image data stream in the absence of an obvious language switch. The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing specification. All changes that come within the meaning and range of the claims are intended to be included within the scope of the claims.

It is a single language support printer technology according to the prior art. This is a multi-language support printer technology according to the prior art. This is an automatic language switching technology according to prior art. This is another automatic language switching technology according to the prior art. 1 illustrates an exemplary system that provides a suitable operating environment for use in connection with the present invention. 1 illustrates a representative networked system configuration associated with one embodiment of the present invention. 3 is a flowchart illustrating an exemplary embodiment according to the present invention. Next, an exemplary embodiment for dynamically determining the language recognition device to be selected will be described. Fig. 4 illustrates an exemplary embodiment for dynamically determining the sample size by the currently selected language recognizer. Fig. 4 illustrates an exemplary embodiment for sequentially testing dynamically selected language recognizers. An exemplary embodiment for starting a language interpreter and updating the heuristics after recognition is shown.

Explanation of symbols

10, 36 Computer device 12 System bus 14 Processing system 16 Memory 18 Mass storage device interface 20 Input interface 22 Output interface 24 Network interface 26 Mass storage device 28 ROM
30 RAM
32 input devices 34 output devices

Claims (25)

A method for accurately recognizing a language format of an input image data stream when there is no language switch for switching a language for forming an image,
Performing a sniffer process to determine the language format and analyzing the first sample of the input image data stream;
The sniffer process is
Dynamically ordering one or more language recognizers;
Pre-reading the sample size used in the first language recognizer of one or more language recognizers;
Using the first sample to call a callback function used in the first language recognizer;
Determining whether the first language recognizer recognizes the language format;
If the language format is recognized during the sniffer process, the input image data stream is rendered ;
When dynamically ordering the one or more language recognizers, it is applied according to environmental factors including the location of the device that renders the job represented by the image data stream and the locality of the user emitting the image data stream. A separate heuristic set is maintained, one heuristic set is selected according to the environmental factors associated with the image data stream, and the one or more language recognizers are selected using the selected heuristic set. A method characterized by ordering .   2. The method of claim 1, wherein when dynamically ordering the one or more language recognizers, the order of the one or more language recognizers is selected based on heuristics.   The method according to claim 2, wherein when selecting the order of the one or more language recognition devices based on an empirical rule, a cumulative value of the number of times the language format has been recognized is stored.   3. The method according to claim 2, wherein when the order of the one or more language recognition devices is selected based on an empirical rule, the priority order is specified in advance based on the recognition frequency of the language format. The priority order is
5. The method according to claim 4, wherein the method is specified based on either (i) the previous recognition of the language format, or (ii) the number of times the language format has been recognized. When selecting the order of the one or more language recognition devices based on heuristics,
First, order in random order,
Order in a pre-established priority order based on the frequency of recognition of the language format,
3. The method according to claim 2, wherein when the cumulative number of times of two or more language recognition devices is the same, the two or more language recognition devices that are the same are ordered in a random order.   3. The method of claim 2, wherein if the language format is recognized, the rule of thumb used for selection is updated.   2. The image data stream is initialized and bound to the one language recognition device when the language format is recognized by one language recognition device among the one or more language recognition devices. The method described.   The method of claim 1, wherein when calling the callback function, the callback function is called using the first sample as an argument. If the first language recognizer does not recognize the language format,
The sniffer process is
Pre-reading the size of the sample used in the language recognition device next to the one or more language recognition devices;
Call the callback function used in the next language recognizer using the first sample,
2. The method of claim 1, wherein the next language recognition device determines whether to recognize the language format. When rendering the input image data stream, the image data stream is rendered by an image device;
The imaging device
(I) a printer,
(Ii) a multifunction peripheral device;
(Iii) a scanner,
(Iv) a facsimile machine;
(V) electronic whiteboard,
(Vi) display monitor;
(Vii) Tablet PC,
(Viii) a document server,
(Ix) copier,
(X) a medical image analysis device,
(Xi) Geological image analysis device,
(Xii) audio / visual devices,
The method of claim 1, comprising any one of (xiii) natural language devices. A method for accurately recognizing a language format of an input image data stream when there is no language switch for switching a language for forming an image,
Receive the image data stream,
Dynamically ordering one or more language recognizers;
Pre-reading the sample size used in the first language recognizer of one or more language recognizers;
Call the callback function used in the first language recognizer using the first sample,
Determining whether the first language recognizer recognizes the language format;
If it is determined that the language format is recognized, the input image data stream is rendered ;
When dynamically ordering the one or more language recognizers, it is applied according to environmental factors including the location of the device that renders the job represented by the image data stream and the locality of the user emitting the image data stream. A separate heuristic set is maintained, one heuristic set is selected according to the environmental factors associated with the image data stream, and the one or more language recognizers are selected using the selected heuristic set. A method characterized by ordering . If the first language recognizer does not recognize the language format,
Pre-reading the size of the sample used in the language recognition device next to the one or more language recognition devices;
Call the callback function used in the next language recognizer using the first sample,
Determining whether the next language recognition device recognizes the language format;
13. The method of claim 12, wherein if the language format is determined to be recognized, the input image data stream is rendered.   13. The method of claim 12, wherein when dynamically ordering the one or more language recognizers, the order of the one or more language recognizers is selected based on heuristics.   15. The method according to claim 14, wherein when selecting the order of the one or more language recognition devices based on heuristics, a cumulative value of the number of times the language format has been recognized is stored. The order is
16. The method of claim 15, wherein the method is specified based on either (i) a previous recognition of the language format, or (ii) a number of times the language format was recognized. When selecting the order of the one or more language recognition devices based on heuristics,
First, order in random order,
Order in a pre-specified priority order based on the frequency of recognition of the language format,
If the cumulative number of times that the language formats of two or more language recognition devices have been recognized is the same, order the two or more language recognition devices that are the same in a random order;
16. The method of claim 15, wherein heuristics are updated if the language format is recognized.   13. The image data stream is initialized and bound to the one language recognition device when the language format is recognized by one language recognition device among the one or more language recognition devices. The method described. A computer program that can be executed by a computer embodying a method of accurately recognizing the language format of an input image data stream when there is no language switch for switching the language for forming an image,
In the computer,
Receiving an image data stream;
Dynamically ordering one or more language recognizers;
Pre-reading the size of the sample used in the first language recognition device of the one or more language recognition devices;
Using a first sample to invoke a callback function for use with the first language recognizer;
Determining whether the first language recognition device recognizes the language format, and, if it is determined that the language format is recognized in the step, rendering the input image data stream I mean,
The step of dynamically ordering the one or more language recognizers is applied according to environmental factors including the location of the device that renders the job represented by the image data stream and the locality of the user emitting the image data stream. A set of heuristic rules, selecting one heuristic set according to the environmental factors associated with the image data stream, and using the selected heuristic set, the one or more language recognition devices A computer program comprising the steps of ordering . In the computer,
If it is determined that the first language recognition device does not recognize the language format, a step of pre-reading a sample size used in a language recognition device next to the one or more language recognition devices;
Calling a callback function to be used in the next language recognition device by the first sample;
Determining whether the next language recognition device recognizes the language format, and, if it is determined that the language format is recognized in the step, rendering the input image data stream The computer program according to claim 19.   20. The computer according to claim 19, wherein the step of dynamically ordering the one or more language recognition devices functions as a step of selecting an order of the one or more language recognition devices based on an empirical rule. program.   The step of selecting the order of the one or more language recognition devices based on an empirical rule functions as a step of storing a cumulative value of the number of times the language format has been recognized. The computer program described. The order is
(I) at the previous recognition of the language format, or (ii) the number of times the language format is recognized,
23. The computer program according to claim 22, wherein the computer program is specified based on any one of the above. Selecting an order of the one or more language recognition devices based on heuristics;
Ordering in random order,
Ordering in a pre-specified priority order based on the frequency of recognition of the language format;
Ordering the two or more language recognition devices in the random order when the cumulative number of times the language formats of the two or more language recognition devices have been recognized is the same, and the language format The computer program according to claim 21, wherein the computer program functions as a step of updating an empirical rule when the rule is recognized. In the computer,
If the language format in accordance with one of the language recognition device among the one or more speech recognition device is recognized, characterized in that to perform the steps of the image data streams initializes, to bind said to one language recognition device The computer program according to claim 20.

Images