JP2010026986A - Image forming apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that optimum processing for character display is not performed because an applicable part is not drawn as a character by performing outline conversion of a text drawing instruction into a path drawing instruction in the conventional manner. <P>SOLUTION: By extracting a character path group from the path drawing instruction and calculating the character likeness (character likelihood) of the character path group, a character attribute is restored. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image forming apparatus and an image forming method for processing document data including a path drawing command.

  Submission processing using electronic data is widely used. In this case, the data creator decides the arrangement of images and characters on his own computer and transmits it to the data printer via a network or the like. The data printer prints based on this.

  However, in this case, if there is a difference in environment between the data creator and the printer, an undesired printing result may occur. A typical example is the presence or absence of a font. If the electronic data includes a font that the printer does not have, another alternative font similar to that font must be used. However, this may result in a print result different from what the creator intended.

  One means for avoiding this is outline processing for replacing a font that the printer does not have with an image drawing command for drawing a path. In the original electronic data, character drawing is performed by drawing attribute information such as a character drawing position and size and a text drawing command for identifying the character to be drawn and actually drawing. Outline processing is to convert a text drawing command into a pass point setting command and a pass painting command using the drawing attribute information and the drawing command.

  As a result, even if the print data uses a font that the printer does not have, a result close to the result intended by the creator can be obtained by processing as a path drawing command.

In addition, there is an apparatus that extracts an edge of a text part when performing outline processing and performs image processing for emphasizing the edge part (Patent Document 1). Starting with this, conventionally, a trapping process or the like has generally been added when making an outline.
Japanese Patent Laid-Open No. 10-294863

  However, conventionally, there has been a problem that by converting a text drawing command to a path drawing command, the corresponding portion is not drawn as a character, and optimal processing for character display is not performed. This optimal processing includes line width adjustment based on hinting information, use of a dither pattern for characters, and the like.

  Japanese Patent Application Laid-Open No. 2004-133830 extracts an edge of a text portion when creating an outline and performs image processing for emphasizing the edge portion. Starting with this, conventionally, a trapping process or the like has generally been added when making an outline. However, there are cases where document data to which correction processing has not been applied is submitted when making an outline. Also, the correction process may not be appropriate for the device. Therefore, it is necessary to consider means for correcting the data that has been outlined.

In order to solve the problems, an image forming apparatus, an image forming method, and a program for executing the image forming method according to the present invention include:
Means for performing a grouping process based on the characteristics of the path drawing command and generating a path group;
Means for determining the likelihood that the path is a character based on the characteristics of the path group;
Means for executing the same drawing process as the character drawing for the path group having a high likelihood.

  As a means for obtaining the character likelihood based on the characteristics of the path group, there may be a means for recognizing that the likelihood is high when the surrounding path group is located in a straight line.

  As a means for obtaining the character likelihood based on the characteristics of the path group, the aspect ratio of the boundary group of the path group is approximately 1, or the likelihood based on the aspect ratio appropriate for identifying the character as recorded and held by an external database or the like. There may be means for determining the degree.

  As a means for obtaining the character likelihood based on the characteristics of the path group, there is a means for obtaining the boundary box frequency of the path group and recognizing that the high-frequency boundary box path group has a high likelihood. good.

  As means for obtaining the character likelihood based on the characteristics of the path group, there is means for recognizing that the likelihood is high when the color in which the path group is drawn is the same as that of the adjacent path group. Also good.

  As a means for obtaining the character likelihood based on the characteristics of the path group, a path drawing command is executed for those having a certain likelihood or less, and character recognition processing is further performed on the path drawing command to determine the likelihood. You may have a means.

  You may have a means for calculating | requiring likelihood combining at least 2 or more of the means for calculating | requiring the character likelihood shown above.

  According to the first aspect of the present invention, it is possible to determine whether or not the path drawing command being processed is derived from the outlined character drawing command when performing the image forming process. . In the case of an outline character drawing command, an optimum process can be used by performing the same drawing setting as the character drawing.

  According to the invention of claim 2, it is possible to further increase the degree of character discrimination for data including text data written in horizontal writing / vertical writing.

  According to the invention of claim 3, it is possible to further increase the character recognition degree for character glyphs having different aspect ratios for each character, such as kanji, hangul, and alphabet, whose vertical and horizontal widths for each character are almost constant.

  According to the invention of claim 4, it is possible to further increase the degree of character identification for data in which characters of the same size are repeatedly used.

  According to the invention of claim 5, it is possible to further increase the degree of character identification for data in which the color in which characters are drawn is uniform.

  According to the invention of claim 6, when it is difficult to determine whether or not it is a character, the character is actually drawn and all data is processed by determining whether or not it is a character by character recognition processing. Character identification processing can be executed in a shorter time.

  According to the invention of claim 7, character identification processing based on integrated judgment can be performed by combining the above inventions.

  Next, details of the present invention will be described in accordance with the description of the embodiments.

<Computer configuration>
FIG. 1 shows an example of a configuration diagram of a computer of a system according to an embodiment of the present invention. Reference numeral 101 denotes a CPU, that is, a central processing unit, which controls the entire apparatus, performs arithmetic processing, and the like. Reference numeral 103 denotes a ROM, that is, a read-only memory, which stores a system activation program, a basic I / O program, character pattern data (font data) for converting a character code into a bit pattern, and the like.

  Reference numeral 102 denotes a RAM, that is, a random access memory, which temporarily stores data used for calculation by the CPU 101, calculation results, character pattern data strings converted from character codes, graphic data, image data, and the like for display.

  Reference numeral 106 denotes KBDC, that is, a keyboard control unit. Reference numeral 107 denotes KB, that is, key input data (character code or control code) is received from the keyboard and transmitted to the CPU 101. Reference numeral 104 denotes a DSPC, that is, a display control unit, which reads a character pattern data string stored in the RAM 102 and transfers it to the DSP 105. Reference numeral 105 denotes a DSP, that is, a display device, which receives a character pattern data string, graphic data, and image data from the DSPC 104 and displays them on a display screen.

  Reference numeral 108 denotes DSKC, that is, a disk control unit, which controls access to the external storage device 109. Reference numeral 109 denotes a hard disk device. HD stores character pattern data (font data) and character development processing programs that read font data and convert them to bitmap data, graphic development processing programs that process graphic data, image data processing programs that process image data, etc. is doing.

  Reference numeral 110 denotes a NIC, that is, a network interface card, which is connected to the network 201 and exchanges information with other devices. Reference numeral 201 denotes a network, which means a computer network such as an intranet in which a large number of computers are connected using a network interface.

<Printer configuration example>
FIG. 2 is a block diagram illustrating a configuration of main parts of the multifunction machine according to the present embodiment.

  The controller unit 2000 connects a scanner 2070 as an image input device and a printer 2095 as an image output device, and performs control for realizing a copy function for printing out image data read by the scanner 2070 by the printer 2095. At the same time, the controller unit 2000 performs control for inputting / outputting image information and device information by connecting to the LAN 3000 and the public line 3001 (WAN).

  Specifically, the controller unit 2000 includes a CPU 2001, and the CPU 2001 starts up an operation system (OS) by a boot program stored in the ROM 2003. Various processes are executed by executing application programs stored in an HDD (Hard Disk Drive) 2004 on this OS. A RAM 2002 is used as a work area of the CPU 2001. The RAM 2002 provides an image memory area for temporarily storing image data together with a work area. The HDD 2004 stores image data together with the application program.

  A ROM 2003 and a RAM 2002 are connected to the CPU 2001 via a system bus 2007. In addition, an operation unit I / F (operation unit interface) 2006, a network I / F (network interface) 2010, a modem 2050, and an image bus I / F (image interface) 2005 are also connected to the CPU 2001.

  The operation unit I / F 2006 is an interface with the operation unit 2012 having a touch panel, and outputs image data to be displayed on the operation unit 2012 to the operation unit 2012. In addition, the operation unit I / F 2006 transmits information input by the user in the operation unit 2012 to the CPU 2001.

  Next, the network I / F 2010 is connected to the LAN 3000 and inputs / outputs information to / from each device on the LAN 3000 via the LAN 3000. The modem 2050 is connected to the public line 3001 and inputs / outputs information via the public line 3001.

  An image bus I / F 2005 is a bus bridge for connecting a system bus 2007 and an image bus 2008 that transfers image data at high speed and converting the data structure. The image bus 2008 is configured from a PCI bus or IEEE1394. On the image bus 2008, a RIP 2060, a device I / F 2020, a scanner image processing unit 2080, a printer image processing unit 2090, an image rotation unit 2030, a thumbnail creation unit 2035, and an image compression unit 2040 are provided. The RIP 2060 is a raster image processor and is a processor that develops a PDL code into a bitmap image. A scanner 2070 and a printer 2095 are connected to the device I / F 2020, and the device I / F 2020 performs synchronous / asynchronous conversion of image data. A scanner image processing unit 2080 corrects, processes, and edits input image data. A printer image processing unit 2090 performs printer correction, resolution conversion, and the like on print output image data. The image rotation unit 2030 rotates image data. The image compression unit 2040 compresses multi-valued image data into JPEG data and binary image data into data such as JBIG, MMR, and MH, and performs the decompression process.

  FIG. 3 is a diagram schematically showing a hardware configuration of the reader unit 2070 and the printer unit 2095 of the copier of FIG.

  The reader unit 2070 and the printer unit 2095 are integrally configured as shown in FIG. The reader unit 2070 includes a document feeding unit 9250. The document feeding unit 9250 feeds documents one by one from the top onto the platen glass 9211. Each time the reading operation of each original is completed, the original is discharged from the platen glass 9211 to a discharge tray (not shown). When the document is fed onto the platen glass 9211, the reader unit 2070 turns on the lamp 9212 and starts moving the moving unit 9213. By the movement of the moving unit 9213, the scanning on the original on the platen glass 9211 is performed. During this scanning, the reflected light from the original is guided to a CCD image sensor (hereinafter referred to as CCD) 9218 through each mirror 9214, 9215, 9216 and lens 9217, and the image on the original is connected to the imaging surface of the CCD 9218. Imaged. The CCD 9218 converts an image formed on the imaging surface into an electric signal, and the electric signal is input to the control device 9110 after being subjected to predetermined processing.

  The printer unit 2095 has a laser driver 9321, and the laser driver 9321 drives the laser light emitting unit 9322 based on the image data input from the control device 9110. Accordingly, laser light corresponding to the image data is emitted from the laser light emitting unit 9322, and this laser light is irradiated onto the photosensitive drum 9323 while being scanned. An electrostatic latent image is formed on the photosensitive drum 9323 by the irradiated laser light, and the electrostatic latent image is visualized as a toner image by the toner supplied from the developing device 9324. In synchronization with the irradiation timing of the laser light, the recording paper is fed from the cassettes 9131 and 9132 between the photosensitive drum 9323 and the transfer unit 9325 through the conveyance path, and the toner image on the photosensitive drum 9323 is transferred to the transfer unit 9325. Is transferred onto the recording paper fed.

  The recording paper onto which the toner image has been transferred is sent to a fixing roller pair (heating roller and pressure roller) 9326 via a conveyor belt, and the fixing roller pair 9326 heats the recording paper to transfer the toner image on the recording paper onto the recording paper. To settle. The recording paper that has passed through the fixing roller pair 9326 is discharged to a paper discharge unit 9330 by a paper discharge roller pair 9327. The paper discharge unit 9330 includes a sheet processing apparatus capable of performing post-processing such as sorting and stapling. When the duplex recording mode is set, after the recording sheet is conveyed to the discharge roller pair 9327, the rotation direction of the discharge roller pair 9327 is reversed and guided to the refeed conveyance path 9329 by the flapper 9328. . The recording paper guided to the refeed conveyance path 9329 is fed again between the photosensitive drum 9323 and the transfer unit 9325 at the timing described above, and the toner image is transferred to the back surface of the recording paper.

  Example 1 describes a preferred example to which the present invention is applied.

  The basic flow of the PDL process is composed of the following steps: process start 111, data reception 112, document structure grasp 113, process page selection 114, print command execution 115, process continuation determination 116, and process end 117.

<Data reception 112>
This is a process for receiving document data used for printing from outside or inside.

  It is assumed that the document data is submitted to the printer from another computer located on the network from the LAN 3000 by meeting the network I / F 2010. In the present invention, since the data format of the submitted document data is different from the essence, it is not limited, but can be applied to, for example, PDF ((registered trademark) Portable Document Format).

  The timing of printing the data submitted to the printer may include immediate response when data is received, or temporary retention and start based on user support. The present invention can be applied to both.

  For example, data input to the printer is input from the network I / F 2010 and stored in the external storage device 109. On the other hand, it is assumed that printing is performed when the user issues a print instruction from the operation unit 2012 or the like.

  Alternatively, a processing setting for processing document data may be received.

<Document structure grasp 113>
The received document data may describe the structure of the document data. By analyzing this, it is a process for grasping the number of pages to be processed.

<Process page selection 114>
This is a process for selecting data indicating a page to be processed. This is a process for specifying page data to be processed based on the document structure grasp 113 and executing the print command shown below.

<Print command execution 115>
This is a step of forming image data for the page selected in the processing page selection 114. Details will be described later.

<Processing continuation determination 116>
This is a process for determining whether there is a page to be processed in addition to the page processed in the document structure grasp 115. If there is a page to be processed, the process returns to the process page selection 114 and a process for selecting a new process page is performed.

  If there is no page to be processed, the process ends (process end 117).

  Here, the print command execution 115 will be described in more detail with reference to FIG. The print command execution 115 includes a print processing command execution start 121, a processing command selection 122, a processing command branch 123, drawing processing 124 to 126, a path group DB registration 127, a processing continuation determination 128, and an end processing 129.

<Drawing command selection 122>
The print data includes content data describing the processing order and processing method of which drawing command. This step is a step for selecting a drawing command to be processed in the subsequent steps in accordance with the processing order.

<Drawing instruction branch 123>
This is a process for branching the process to be processed next based on the drawing command selection 122. Here, the drawing commands are classified into an image drawing command, a text drawing command, and a path drawing command.

<Improved path drawing commands 124A and 124B>
These are processes for processing the outlined text that is the core of the present invention in common. Details will be described later.

<Image drawing command 125>
This step is a step for executing an image drawing command. The image drawing command is a command for drawing image data called an image. When storing in document data, the image may be compressed using DCT compression, MR / MH compression, or the like to reduce the size. In this case, expansion is performed when the image is drawn.

<Text drawing command 126>
This step is a step for executing a text drawing command. The text drawing command is a command for drawing characters, and specifies the type of font to be used, the size to be displayed, and the character string to be displayed.

<Path group DB registration process 127>
In the drawing command branch 123, when a path drawing command is used, registration in the path group DB is performed by this process.

  The path drawing command includes a path point setting command and a path painting command. The pass point setting command is a process for designating an area to be processed in line drawing / painting. On the other hand, the path painting command is a process of performing various drawing commands for the path specified by the pass point setting command. In addition to simple one-color painting, it may be possible to specify gradation, pattern, and the like.

  Here, the path group DB holds a path point setting instruction and a path filling instruction.

<Processing continuation judgment 128>
This is a step for determining whether there is a drawing command to be processed in addition to the drawing command selected in the drawing command selection 122 and processed in the above steps. If there are still drawing commands to be drawn, the process returns to the drawing command selection 122.

If there is no drawing command to be drawn, the process ends (process end 129).
Next, further details of the improved path drawing instructions 124A and 124B will be described with reference to FIG. The improved path drawing command includes an improved path drawing command start 131, a path group DB check 132, a character candidate path group extraction 133, a character likelihood calculation 134, a character likelihood check 136, a drawing attribute setting 137 and 138, a path drawing 139, and a process continuation determination. 140 and end 141.

<Path group DB confirmation processing 132>
This is a step of determining whether or not a path group to be processed in the future is registered in the path group DB. If there is no data in the path group DB, the process ends (process end 141).

  If data exists in the path group DB, character candidate path group extraction 133 is executed.

<Character candidate path group extraction processing 133>
The path data included in the path group DB is not in character units. This step is a step for grouping these path groups in units of characters.

  First, an example of outline processing is shown in FIG. The original text drawing command 401 includes a font name to be used, a drawing position, a character code of the text to be drawn, and the like. Based on this, pass point setting commands 402 and 404 and pass painting commands 403 and 405 are converted. In this example, the character glyph is roughly divided into two (402 and 404).

  The gist of the present invention is to determine whether the path drawing command is a text drawing command using the path drawing commands 402 to 405.

  FIG. 5 shows the target path drawing command. At this point, since it is a drawing command group composed of a path setting command and a path painting command, it cannot be processed as text.

  Next, a boundary box of this path drawing command group is obtained. The result is shown in FIG. There is a case where one character is drawn by a combination of a plurality of path drawing commands and path filling commands. In this case, (1) when there is an overlap with a boundary box of another pass point setting instruction in a boundary box of a certain pass point setting instruction, these are recognized as one path group. (2) If the new boundary box combined with the adjacent path point setting command has an aspect ratio close to 1, they are recognized as one path group.

  FIG. 7 focuses on the characters “N” and “Company” in FIG. In the case of “n”, since the boundary box 702 overlaps the boundary box 701, these are recognized as one character candidate path group. On the other hand, in the case of “Company”, 703 and 704 are adjacent to each other, and the aspect ratio of the boundary box 705 combining these is close to 1. Also in this case, it is recognized as a path group of one character candidate.

  By repeating the above operations, it is possible to extract character candidate path groups from all path point setting commands as shown in FIG.

<Character likelihood calculation 134>
This step is a step of calculating the character likelihood indicating the character likeness for the character-by-character path group determined by the character candidate path group extracting step 133.

  Hereinafter, the focused evaluation method of the character candidate path group will be described with reference to FIG.

≪Calculation of character likelihood (character linearity) ≫
When kanji, katakana and hiragana used in Japanese are used to display text, the center coordinates of the glyphs are not so shifted. Based on this, the character likelihood calculation is performed in the invention described in claim 2.

  The respective center coordinates are obtained for a series of character candidate paths (901). Further, it is determined whether these center coordinates match those of another character candidate path group. In this case, it is desirable to provide a certain tolerance. If they match, the average of these center coordinates is set as a new center coordinate. By repeating this, it is possible to extract a path group located in a straight line from the character candidate paths, and it is possible to further increase the character likelihood by increasing the character likelihood.

  However, with this method, it is difficult to discriminate against stuttering sounds such as “Yayo”, prompting sounds such as “tsu”, or characters that are slightly smaller than other characters such as “Ai ぅ e ぅ”. There is also a problem of end. In such a case, instead of the center coordinates, a method of calculating based on the coordinates of the lower end in the case of horizontal writing or the coordinates of the right end in the case of vertical writing may be used (902).

  By repeating this, the likelihood that each character candidate path group is a character can be calculated.

<Processing path group selection 135>
This shift is the actual drawing process. This step is a step of selecting a path to be drawn and a path group from the path group DB.

<Character likelihood check processing 136>
This step is a step for determining which drawing attribute is set based on the character likelihood of the path and the path group.

  As a simple method, a certain threshold value is provided, and if it is less than the threshold value, drawing attribute setting 137 as a path is performed, and if it is equal to or larger than the threshold value, drawing attribute setting 138 as a character is performed.

<Path drawing attribute setting 137>
<Text drawing attribute setting 138>
The setting for drawing as a path or text is set in the path object to be drawn next. For example, there are settings such as a dither pattern to be used and a toner application amount limit.

<Path drawing process 139>
This step is a step of performing a drawing process on the path selected in the processing path group selection step using the drawing attributes set in the drawing attribute settings 137 and 138.

<Processing continuation judgment 140>
This is a process for determining whether there is a path / path group to be processed in addition to the path / path group selected by the processing path group selection 135 and processed in the above process. If there remains a path group to be processed, the process returns to the process path group selection 135.

  If there is no path group to be processed, path group DB initialization 141 is performed.

<Path group DB initialization 141>
In this step, information included in the path group DB is initialized. At this time, a method of not initializing the parameters used may be used. End processing (End processing 141)

  As Example 2, another character likelihood calculation method that can be used in the character likelihood calculation 136 will be described. Other configurations are the same as those in the first embodiment, and are omitted here.

≪Character likelihood calculation (boundary box aspect ratio) ≫
Kanji, Katakana, and Hiragana used in Japanese sentences have a characteristic that the aspect ratio is close to 1: 1 except for a part. Also, in English, hiragana and katakana, the same characters are used repeatedly, so that it is easy to find based on the aspect ratio frequency. The invention of claim 3 pays attention to these characteristics.

  In the case of Chinese characters, if some characters such as “1”, “2”, etc. are removed, many of them have a shape that fits in a square, that is, a shape close to about 1: 1 aspect ratio. Therefore, it is possible to assume whether the character is a Chinese character based on the frequency of the aspect ratio.

  Furthermore, in the case of the alphabet, the same glyph is more reusable than kanji and the like because a sentence is formed by combining limited characters such as uppercase and lowercase letters. Therefore, if the frequency with respect to the boundary aspect ratio of the character candidate path group is derived, it is possible to select a candidate that seems to be a character therefrom.

  In particular, the present invention has a preferable effect on document data mainly composed of text.

  This example will be described with reference to FIG. The height and width are calculated for each character candidate path group. From this, the aspect ratio is obtained. Of the characters from 1001 to 1008, except for 1003 indicating “no”, if all have a height of 10, a width of 10 and a ratio of 1.0, all of these path groups are treated as a path group having a high character likelihood.

≪Character likelihood calculation (boundary box size frequency analysis) ≫
As is clear from general newspapers and weekly magazines, the text size of publication documents is often uniform. This is the invention of claim 4.

  It is possible to obtain the frequency with respect to the boundary box size of the character, and to estimate the character likelihood high for the boundary box that repeatedly appears more than a certain amount.

  In particular, the present invention has a favorable effect on document data whose document format is determined to some extent.

  This example will be described with reference to FIG. The height and width are calculated for each character candidate path group. Here, when the frequency of each height or width is obtained, the height is 10 px, and the width is 10 except for one. Therefore, in this case, data having a height of 10 px or a width of 10 px is handled as data having a high character likelihood.

  In the case of the data of 1003, the height is the same as the other data, but since the width is different, the character likelihood is handled lower than the other data.

≪Calculation of character likelihood (color information of path painting command) ≫
In many cases, the color of characters included in a document is unified for the sake of legibility. Therefore, paying attention to the color information of the path paint command obtained as a result of the outline, if the same color is applied, it may be that they originally belonged to the same word or sentence I can say that.

  Although the same color is described here, for example, colors such as red, blue, and green may be used to stand out in the document. Therefore, it is also an effective means to calculate the character likelihood in consideration of these colors. It is.

  In particular, the present invention has a favorable effect on document-oriented document data such as office documents.

≪Calculation of character likelihood (including character recognition processing) ≫
If the target path group is not likely to be a character, but has a character likelihood that it is difficult to determine that it is not a character, the character is actually drawn and recognized. Processing may be attempted.

  Based on this result, the character likelihood of another path group related to the path may be estimated to be high using the premise that the path group is a character.

  In this case, there is a merit that the amount of calculation is smaller than when the character recognition process is performed after executing all other drawing commands and creating a final image. Since only the character candidate path to be processed is drawn, the processing time is shortened, and noise associated with the background image does not enter, so that it is possible to improve the identification performance as to whether or not the character is a character.

≪Calculation of character likelihood (combination) ≫
The calculation methods shown above are methods that attempt to calculate a more desirable character likelihood by combining them. In addition to simple averaging, weighted averaging may be performed.

1 is a computer configuration for performing image forming processing using the image forming method of the present invention. It is a system configuration of a multifunction machine that can use the present invention. 1 is a hardware configuration of a multifunction machine that can use the present invention. It is an example of text outlining. It is the text made into the outline used for description in Example 1. FIG. It is the result of extracting the boundary box with respect to FIG. It is an example of grouping with respect to FIG. It is the result of having performed path grouping with respect to FIG. FIG. 6 is an example of character likelihood calculation based on the center line or the underline. FIG. 6 is an example of character likelihood calculation based on the size of the boundary box. 4 is a flowchart of a PDL interpreter in an embodiment of the present invention. It is a flowchart of the drawing command process in the Example of this invention. It is a flowchart of the improvement path | pass drawing command in the Example of this invention.

Explanation of symbols

401 Text drawing command before outlining
402-405 Path drawing command (path point setting command / path filling command) required by outline conversion
901 Example of character likelihood calculation based on the center line of the path group
902 Example of character likelihood calculation based on bottom line of path group

Claims (7)

In an image forming apparatus that processes a document data file including a path drawing command,
Means for performing a grouping process based on the characteristics of the path drawing command and generating a path group;
Means for determining the likelihood that the path is a character based on the characteristics of the path group;
An image forming apparatus comprising: means for executing the same drawing process as that of character drawing for a path group having a high likelihood. The image according to claim 1, further comprising means for recognizing that the likelihood is high when the path group is located in a straight line as the means for obtaining the character likelihood based on the characteristics of the path group. Forming equipment.   As a means for obtaining the character likelihood based on the characteristics of the path group, the likelihood is determined when the aspect ratio of the boundary group of the path group is approximately 1, or is close to a character aspect ratio standard known by any method. 2. The image forming apparatus according to claim 1, further comprising means for recognizing a high one. As a means for obtaining the character likelihood based on the characteristics of the path group, at least one frequency in the vertical or horizontal direction of the boundary box of the path group is obtained, and the path group of the boundary box having a high frequency has a high likelihood. The image forming apparatus according to claim 1, further comprising a recognizing unit.   The means for obtaining the character likelihood based on the characteristics of the path group includes means for recognizing that the likelihood is high when the color in which the path group is drawn is the same as that of the adjacent path group. The image forming apparatus according to claim 1. As a means for obtaining the character likelihood based on the characteristics of the path group, an image is created by executing a path drawing command for those included in a certain threshold range, and character recognition processing is performed on the image. The image forming apparatus according to claim 1, further comprising means for determining likelihood. An image forming apparatus comprising: means for obtaining a likelihood by combining at least two or more of the likelihood determining means according to any one of claims 2 to 6.

Images