KR20080055710A - Information processing device and method of controlling the same - Google Patents

Abstract

An information processing device and a controlling method thereof are provided to generate a document image facilitating cutting characters correctly when a document image is generated, and facilitate cutting the characters normally when a post-process such as character recognition is performed. A mask pattern storing part(203) stores a plurality of different mask patterns. A sentence information input unit(201) receives sentence information(200). A character pattern generator(202) generates a corresponding character pattern based on a character code form a sentence included in the received sentence information. A mask pattern selector(204) selects the different mask pattern from the mask pattern selected by the previous character code of a target character code among the mask patterns stored in the mask pattern storing part. A combiner(205) combines a character pattern corresponding to the target character pattern and the selected mask pattern. The mask pattern storing part stores two mask patterns, and the mask pattern selector selects one of two mask patterns alternatively whenever the character pattern corresponding to the size of one character is generated.

Description

Information processing device and control method of information processing device {INFORMATION PROCESSING DEVICE AND METHOD OF CONTROLLING THE SAME}

The present invention relates to a technique for embedding watermark information in an electronic document.

In the optical character recognition device (OCR), characters are cut out in units of characters from an optically read document image, and character recognition is performed for each cut out character. However, in the character cutting process, it is difficult to accurately recognize characters in units of characters, and various techniques have been proposed so far.

For example, Japanese Patent Application Laid-open No. Hei 5-128299 (Document 1) cuts characters in a document image one by one in sequence from an input document image, performs feature extraction and similarity calculation, and performs a character with low similarity. A technique of changing the character clipping of an image and repeating feature extraction and similarity calculation is disclosed (see Figs. 15A to 15E).

In addition, Japanese Patent Laid-Open No. Hei 9-237316 (Document 2) cuts a character out of a read string, and based on the ratio of the horizontal width dx and the vertical width dy of each cut character, 2 to each cut character. A technique for determining whether or not a character abnormality is included is disclosed (see Fig. 16).

The technique disclosed in the document 1 is preferable at the point which can reduce the misperception by the character cutting.

In addition, in the technique disclosed in Document 2, when it is determined that two or more characters are included in each cut character, and if the cut character cannot be recognized, at least two character character recognition results are obtained. The correction symbol and the space are displayed on the screen of the display unit. For this reason, it is preferable at the point which can shorten the time required for correction of a character recognition result.

However, in Document 1, since the similarity calculation is repeated one by one for the characters in the document image, there is a problem that the processing time is unnecessarily taken.

In addition, in Document 2, it is difficult to cut out characters when half-width characters and full-width characters are mixed (see Fig. 17). In addition, a character composed of sides and rooms may be cut out into two characters.

Therefore, an object of the present invention is to generate a document image which makes it easy to cut out characters when generating a document image. In addition, in the case of performing post-processing such as character recognition, the character is easily cut off.

In the present invention, an information processing apparatus for generating document image data based on sentence information composed of character codes, comprising: storage means for storing a plurality of different mask patterns, input means for inputting sentence information, and input thereof Character pattern generating means for generating a corresponding character pattern based on a character code constituting a sentence in the sentence information input by the means, and when generating a character pattern corresponding to the character code of interest, Selecting means for selecting a mask pattern different from the mask pattern selected by the previous character code from the storage means, a character pattern corresponding to the character code of interest and a combining means for combining the mask pattern selected by the selection means; An information processing apparatus is provided, which is provided.

Moreover, in this invention, input means is a control method provided with the memory means which stores several mask patterns different from each other, and produces | generates document image data based on the sentence information comprised from a character code. A character pattern generation step of generating a character pattern corresponding to the character pattern generation means based on an input step of inputting information, a character code constituting a sentence in the sentence information by the input step, and the character pattern generation step In the case of generating a character pattern corresponding to the character code of interest in the step of selecting, wherein the selection means selects from the storage means a mask pattern different from the mask pattern selected by the character code preceding one of the character code of interest; The character pattern corresponding to the character code of interest and the mask pattern selected by the selection means. The information processing method comprising: a synthesizing step is provided of synthesizing.

Moreover, in this invention, the information processing apparatus which produces | generates document image data based on the sentence information comprised from a character code, Comprising: Storage means which stores several different mask patterns, Input means which inputs sentence information, On the basis of the character code constituting the sentence in the sentence information input by the input means, character pattern generating means for generating a corresponding character pattern, and whether a distribution region of independent significant pixels of the character pattern of interest is composed of two or more; A judging means for judging whether or not, and if it is judged by the judging means that the distribution region of the independent significant pixels of the attention character pattern is composed of one, the first outputting the attention character pattern as part of the page image; By the output means and the said determination means, the distribution area | region of independent significant pixel of the character pattern of interest was comprised by two or more. If it is determined, selecting means for selecting a mask pattern different from the mask pattern selected by the character pattern preceding one of the character patterns of interest from the storage means, the character pattern of interest, and the mask pattern selected by the selection means. There is provided an information processing apparatus comprising: second output means for synthesizing and outputting the synthesized character pattern as part of a page image.

Moreover, in this invention, as a control method of the information processing apparatus provided with the memory means which stores several mask patterns different from each other, and based on the sentence information comprised from the character code which comprises a sentence, A character pattern generation process in which the character pattern generation means generates a corresponding character pattern in order of an input step of inputting sentence information by the input means, and a character code constituting a sentence in the sentence information input by the input step; The judging means determines whether or not the distribution region of the independent significant pixels of the attention character pattern is composed of two or more, and the distribution region of the attention significant character pattern and the independent significance pixels is one by the determination process. If it is determined that it is configured, the first output means outputs the character pattern of interest as part of the page image. If it is determined by the step and the determination step that the distribution area of the independent significant pixel of the character pattern of interest is composed of two or more, the selection means selects a mask pattern different from the mask pattern previously selected from the storage means. The selection step and the second output means comprise a second output step of synthesizing the character pattern of interest and the mask pattern selected in the selection step, and outputting the synthesized character pattern as part of the page image. A control method of a processing apparatus is provided.

As described above, according to the present invention, when generating a document image, it is possible to generate a document image which makes it easy to cut out characters correctly. In addition, when post-processing such as character recognition is performed, character truncation can be easily performed.

Additional features of the present invention are described in the following detailed description (see accompanying drawings).

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention, and both the foregoing general description and the following detailed description describe the principles of the invention.

EMBODIMENT OF THE INVENTION Hereinafter, embodiment which concerns on this invention is described in detail according to an accompanying drawing.

<First Embodiment>

BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the electrical structure of the mask pattern synthesizing apparatus and character cutting apparatus which concerns on one Embodiment of this invention.

The computer 101 is a general-purpose information processing apparatus such as a personal computer, which is generally spread, and can read and edit or store an image from the scanner 117. In addition, the image read by the scanner 117 can be printed from the printer 116. In addition, various instructions from the user are performed by input operations from the mouse 113, the keyboard 114, and the touch panel 119, which are pointing devices. Inside the computer 101, each block described later is connected to the bus 107 to receive various data.

The CPU 102 can control the operation of each block inside the computer 101 or execute a computer program stored therein.

The main memory device (consisting of RAM) 103 is a device that temporarily stores a computer program or image data of a processing target for processing performed by the CPU 102.

The CPU 102 functions as a mask pattern synthesizing device or a character cutting device by executing a computer program stored in the main memory device 103 by an input instruction from the keyboard 114 or the like.

The hard disk (HDD) 104 is a device capable of storing a program or image data transmitted to the main memory device 103 or the like, or storing image data after processing.

The scanner interface (I / F) 115 is an I / F that is connected to the scanner 117 and can input image data read by the scanner 117.

The printer interface 108 is an I / F which is connected to the printer 116 and can transmit the image data to be printed to the printer 116.

The CD drive 109 is a device capable of reading or writing data stored in a CD (CD-R / CD-RW), which is one of external storage media.

The DVD drive 110 is a device capable of reading or writing data stored in a DVD, which is one of external storage media.

The FDD drive 111 is a device capable of reading or writing data stored in FDD, which is one of external storage media.

In addition, when a program for image editing or a printer driver is stored in a CD, DVD, FDD or the like, the program is installed in the hard disk 104 and transferred to the main memory 103 as necessary.

The interface (I / F) 112 is an I / F connected to receive input instructions from the mouse 113 and the keyboard 114.

The interface (I / F) 118 is an I / F connected to receive an input instruction from the touch panel 119.

The monitor 106 is a display device that can display the extraction processing result and the processing of the mask pattern.

The video controller 105 is an apparatus for transmitting display data to the monitor 106.

FIG. 2 is a diagram illustrating a configuration of a mask pattern synthesizing apparatus according to an embodiment of the present invention, and FIG. 3 is a flowchart illustrating an operation procedure of the mask pattern synthesizing apparatus according to an embodiment of the present invention.

The sentence information 200 is composed of a character code (for example, an ASCII code) and is input from the sentence information input unit 201 (step S300). The sentence information input unit 201 is a keyboard 114, but a data file stored in a computer readable storage medium may be used.

In addition, the sentence information 200 may be electronic data such as an application program-specific format or a text format. Moreover, the electronic data created using the document editing application program may be used. The sentence information 200 may be stored in each computer readable storage medium connected to the hard disk 104, the CD drive 109, the DVD drive 110, the FDD drive 111, or the like.

Next, in the character pattern generation unit 202, the character code is read in order of the character codes constituting the sentence in the sentence information 200 input by the sentence information input unit 201, and the corresponding character pattern is read. (Step S301).

The mask pattern storage unit 203 stores two different mask patterns. In the present embodiment, for example, "0" and "1" are stored. Next, the mask pattern selection unit 204 alternately selects a mask pattern different from the mask pattern previously selected from these two mask patterns (step S302). In the case of this embodiment, since two mask patterns are memorize | stored in the mask pattern memory | storage part 203, in the mask pattern selection part 204, two mask patterns are produced every time the character pattern for one character is produced | generated. Alternately select one side of.

Next, the combining unit 205 synthesizes the character pattern corresponding to the character code of interest and the mask pattern selected by the mask pattern selecting unit 204 (step S303). The character pattern is generated in step S301, and the mask pattern is selected from the mask pattern storage unit 203 in step S302. In addition, the synthesis | combination procedure of a character pattern and a mask pattern is mentioned later.

Next, it is determined whether or not the character obtained by combining the character pattern and the mask pattern is the final character of the character pattern (step S304). As a result, when it is determined that it is not the last character, the process returns to step S301 and the same process is repeated. On the other hand, when it is judged that it is the last character, the synthesized character pattern is output from the printer 116 as a part of the page image 206 (step S305).

In addition, in this embodiment, although the output means uses the printer 116, you may store in the hard disk 104 as image data. It may also be configured to transmit to other terminals via a network.

With such a configuration, document image data is generated based on sentence information composed of character codes.

4 is a diagram for explaining a procedure of synthesizing a character pattern and a mask pattern in the mask pattern synthesizing apparatus according to one embodiment of the present invention.

The character 401 is the character "イ" with which the mask pattern is not synthesize | combined. By forming the characters 401 in different dot patterns, the characters 402 and 403 having different mask patterns, respectively, are obtained. The dot patterns represented by the pattern 0 404 and the pattern 1 405 form letters 402 and 403 having different information, respectively.

Next, the process of cutting out each character image for the post-processing of a character image unit from the page image 206 output by the information processing apparatus is demonstrated.

FIG. 5 is a diagram illustrating a configuration of a character slicing device according to an embodiment of the present invention, and FIG. 6 is a flowchart illustrating an operation procedure of the character slicing device according to one embodiment of the present invention.

First, a page image 206 obtained by combining a mask pattern is input from the page image input unit 501 (step S600).

Next, the histogram generator 502 projects the input page image 206 in the vertical and horizontal directions to create a histogram indicating the distribution of the pixel of interest (step S601).

Next, the mask pattern temporary extraction unit 503 extracts the area of the pixel of interest as a temporary character image area based on the histogram created by the histogram generator 502 (step S602).

Next, the mask pattern discriminating unit 504 determines which mask pattern is stored in the mask pattern storage unit 505 with the temporary character image extracted by the mask pattern temporary extracting unit 503 (step S603). .

Next, the mask pattern determination unit 506 determines whether the type of the mask pattern of the temporary character image of interest determined by the mask pattern determination unit 504 is the same as the type of the mask pattern synthesized in the temporary character image at the previous position. It is judged whether or not (step S604).

Next, when the mask pattern determination unit 506 determines that the types of mask patterns are the same, the mask pattern control unit 507 displays an image of a region including the temporary character image of interest and the temporary character image at the immediately preceding position. Is updated as a new temporary character image (step S605). Thereafter, output to the post-processing (step S606). On the other hand, when the mask pattern determination unit 506 determines that the type of mask pattern is different, the temporary character image at the position immediately before the temporary character image of interest is output to the post-processing as a cut-out image of the normal character image. (Step S606).

In addition, the page image 206 may input the printed matter from the scanner 117, and may bitmap it. Moreover, the electronic data created using the document editing application program may be used. In addition, the format may be unique to an application program stored in each computer readable storage medium connected to the hard disk 104, the CD drive 109, the DVD drive 110, the FDD drive 111, or the like. Further, the electronic data such as a text format may be converted into bitmaps by image processing software or the like.

7A to 7D are diagrams for specifically describing character cutting processing according to an embodiment of the present invention.

First, projection is acquired from the input image character "大 石靜" shown in FIG. 7A, and a character is cut out. As a result, four temporary character images called "大石 靑 爭" shown in FIG. 7B are obtained.

Next, the mask pattern "0100" shown in FIG. 7C is extracted from these four temporary character images. Since "0" and "1" are synthesized alternately with the input image character, it can be seen that when the type of mask pattern of the temporary character image of interest and the temporary character image of the previous position is the same, it is an error of character truncation. have. Therefore, in that case, the image of the area | region containing the temporary character image of interest and the temporary character image in the immediate position is updated as a new temporary character image.

As shown in Fig. 7C, since the third &quot; 0 &quot; and the fourth &quot; 0 &quot; are the same kind of mask patterns, both are merged and corrected. If the character cut is corrected in such a procedure, the character cut is in the state "大 石靜" shown in Fig. 7D.

Thereby, the error of character cutting can be corrected by using a mask pattern.

In addition, in this embodiment, the mask pattern memory | storage part 203 memorize | stores several mask patterns, and the synthesis | combination part 205 synthesize | combined the mask pattern selected from the mask pattern selection part 204, and a character pattern. However, the present invention is not limited to this embodiment, but for example, all patterns obtained by combining a character pattern and a mask pattern (at least two kinds) are stored in the storage unit, and the pattern conforms to the character code from these patterns. You can also select.

<Modification 1>

As a modification 1 of the first embodiment, further devising was made so that the mask pattern was synthesized only by the character pattern in which an error would easily occur in the character cutting.

8 is a flowchart for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

In addition, since an apparatus is the same as that of 1st Embodiment, description is abbreviate | omitted. In addition, in the process of this modification, the common part with 1st Embodiment mentioned above is attached | subjected with the reference number of a common step, and description is abbreviate | omitted.

First, in the character pattern generation unit 202, a character pattern corresponding to the input character code is generated (step S301). The processing up to this point does not differ from the first embodiment.

Next, it is judged whether or not the character pattern of interest consists of a distribution area of two or more independent significant pixels (step S801). In this embodiment, the projection histograms in the horizontal and vertical directions of the pixel of interest forming the character pattern of interest are created, and based on the histogram, it is determined whether there are two or more independent distribution areas of the pixel of interest.

As a result, when it is determined that the character pattern of interest is composed of a distribution area of two or more independent significant pixels, a mask pattern different from the mask pattern previously selected is selected from the mask pattern storage unit (step S302). Thereafter, the selected mask pattern and the character pattern are synthesized (step S303), and the processing proceeds to the next process.

On the other hand, when it determines with the character pattern of interest not being comprised by the distribution area of two or more independent pixel of interest, the process of step S302 and step S303 is skipped and it progresses to the next process. Thereby, the page image in which the mask pattern was synthesize | combined only by the character pattern of which the character pattern of interest consists of the distribution area | region of two or more independent significance pixels is produced | generated.

Further, when correcting the character cutout, if the same mask pattern is extracted, the distribution area of each independent significant pixel may be merged.

These processes will be described concretely with the assumption that the sentence information shown in Figs. 9A to 9E is input.

9A to 9E are diagrams specifically illustrating the character cutting process according to the embodiment of the present invention.

It is assumed that "LARILERRO" shown in FIG. 9A is input as sentence information. 9B shows that different mask patterns "1" and "0" are synthesized in "L" and "L", respectively. The mask pattern is not synthesized with other characters, and is represented by "_". The mask pattern is synthesize | combined in the state of "_10__" as a whole.

When the character is cut out of the character string in which the mask pattern is synthesized in this way, it is divided into seven blocks as shown in Fig. 9C. Next, FIG. 9D extracts a mask pattern "_1100__" from seven blocks. Since the same mask pattern ("1" and "1", "0" and "0") were extracted from the adjacent character blocks, they are merged. As a result, since the mask pattern becomes "_10__", as shown in FIG. 9E, the character cutout can be corrected with "La Lilorero".

As a result, in the first modified example, the mask pattern is synthesized by only a part of the character patterns, so that the character image quality can be maintained, and the page image generation and character cropping correction processing can be performed at high speed.

<Modification 2>

As a modified example 2 of the first embodiment, the present invention has been devised so as to select a character at which a character truncation error is likely to occur at high speed.

10 is a flowchart for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

In addition, since an apparatus is the same as that of 1st Embodiment, description is abbreviate | omitted. In addition, about the part in common with 1st Embodiment in the process of this modification, the reference numeral of a common step is attached | subjected, and description is abbreviate | omitted.

In this modified example, it has a dictionary storage part which stores dictionary information which shows the character code which shows the character which has two or more independent distribution area | regions of a significant pixel.

In the character pattern generation unit 202, a character pattern corresponding to the input character code is generated (step S301). The processing up to this point does not differ from the first embodiment.

It is detected whether or not the input character code is registered in the preliminary storage unit (step S1001).

If the inputted character code is registered in the dictionary storage unit, a mask pattern for combining with the character pattern is selected (step S302).

Next, the selected mask pattern is combined with the character pattern (step S303), and the processing proceeds to the next process.

On the other hand, when the input character code is not registered in the preliminary storage unit, the processing of step S302 and step S303 is omitted, and the processing proceeds to the next processing.

Modification 2 is common in synthesizing a mask pattern only with characters having two or more independent distribution regions of the pixel of interest, but the processing can be performed at high speed because the dictionary data registered in the preliminary storage unit is used.

<2nd embodiment>

In the present embodiment, in order to accurately recognize the code, further devising has been made.

11 is a flowchart for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

In addition, since an apparatus is the same as that of 1st Embodiment, description is abbreviate | omitted. In addition, in the process of this embodiment, about common part with the modification 2 of 1st Embodiment, the common step is attached | subjected with the reference number of a description, and description is abbreviate | omitted.

In the character pattern generation unit 202, a character pattern corresponding to the input character code is generated (step S301). The processing up to this point does not differ from the first embodiment.

Next, in the modification 2 mentioned above, it is determined whether the input character code is registered in the prior storage part (step S1001). However, in this embodiment, it is determined whether the character immediately after the character pattern of interest immediately before step S1001 is a sign (step S1101).

As a result, in the case of a code | symbol, the process of step S1001 is abbreviate | omitted and a specific mask pattern is selected from the mask pattern storage part 203 (step S302). For example, "2" different from the above-mentioned mask pattern "0" and "1" is selected. As a method of synthesizing a mask pattern, a method capable of synthesizing two or more bits in one character is used. If the next character is not a sign, the same process as in Modification 2 of the first embodiment is performed.

It is a figure which shows the structure of the character cutting device which concerns on one Embodiment of this invention.

In addition, about the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

First, the mask pattern discriminating unit 504 identifies which mask pattern stored in the mask pattern storage unit 505 has been extracted from the character image of interest. The processing up to this point is common to the first embodiment.

When it is determined that the specific mask pattern is synthesized, the code recognition unit 1201 recognizes that immediately after the character image of interest is the code, and updates as a new character image. Subsequent processing is the same as that of 1st Embodiment.

The flow of the character truncation process is the same as that of the modification 2 of the first embodiment, but is additional to the process of step S604.

13 is a flowchart for explaining the detailed operation procedure of step S604 according to the embodiment of the present invention.

First, the mask pattern determination unit 504 determines whether or not the extracted mask pattern is a specific mask pattern (step S604a). For example, "2" different from the mask pattern "0" "1" mentioned above can be made into the specific mask pattern.

When it is determined in step S604a that it is determined as the specific mask pattern, character recognition of the next character is performed (step S604b). Next, in the code recognition unit 1201, it is determined whether the character recognized character is a code (step S604c). If the character recognized character is a sign, it is determined that the correct character truncation is completed, and the process ends. If it is not a sign, an error point designation for character truncation is performed as in the second modification of the first embodiment (step S604f).

On the other hand, when it is judged that it is not a specific mask pattern, a mask pattern is extracted from the next character of the character image of interest (step S604d), and the mask pattern and the mask pattern extracted from the character image of interest are compared (step S604e). If these mask patterns differ from each other, the process ends. In the same case, an error point designation for character truncation is performed (step S604f), and the processing ends.

Such a process will be described in detail, assuming a case where an arbitrary character image is input.

14A to 14E are views for explaining in detail the character cutting process according to the embodiment of the present invention.

First, as shown in FIG. 14A, a character pattern called "Surure. Next" is generated.

Next, as shown in FIG. 14B, a mask pattern called "_2_01" is embedded in the character pattern. As a result, character cutting is performed as shown in Fig. 14C.

14D shows the mask pattern "_201" extracted from each pixel of significance. Since this mask pattern contains "2" which is a specific mask pattern in the 2nd character, it is judged that the code is included, and the process which separates a character and a code is performed.

As a result, as shown in Fig. 14E, "RU" and "." Are separated into two character blocks. In this embodiment, the specific mask pattern (for example, "2") was embedded in the character before a symbol, and the method of separating a character and a symbol was shown. In the same manner as in this method, in the case where there is an image other than a character in the periphery of the character image of interest, a specific mask pattern is synthesized to the character image of interest. For example, when a table line or the like is disposed at the periphery of the character image of interest, the character and the table line can be separated by synthesizing a specific mask pattern (for example, "3") to the character image.

As a result, the characters and symbols in the table can be cut out accurately.

<Other embodiments>

In addition, even if the present invention is applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), the present invention may be applied to a device (for example, a copier, a facsimile device, etc.) consisting of one device. You may apply.

Moreover, it is achieved by supplying the storage medium which stored the computer program which implement | achieves the function of the above-mentioned embodiment to an apparatus, and the computer (or CPU, MPU) of the apparatus executes the computer program stored in the computer readable storage medium.

In this case, since the computer program itself read out from the computer readable storage medium realizes the functions of the above-described embodiments, the computer readable storage medium storing the computer program constitutes the present invention.

Moreover, since the operating system (OS) etc. which run on a computer perform an actual process based on the computer program read by the computer, the case where the function of the above-mentioned embodiment is implemented by the process is also included.

In addition, when a computer program stored in a computer readable storage medium is written in a memory provided in a function expansion card or a function expansion unit connected to the computer, and then the CPU or the like performs the actual processing based on the instruction of the computer program. There is. In such a case as well, the functions of the above-described embodiments can be realized.

When the present invention is applied to a computer readable storage medium, the computer program corresponding to the above-described flowchart is stored in the computer readable storage medium.

Additional advantages and modifications can be readily made by those skilled in the art. Accordingly, the invention in its broadest sense is not limited to the description and representative embodiments illustrated and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

BRIEF DESCRIPTION OF THE DRAWINGS It is a figure which shows the electrical structure of the mask pattern synthesizing apparatus and character cutting apparatus which concerns on one Embodiment of this invention.

2 is a diagram illustrating a configuration of a mask pattern synthesizing apparatus according to an embodiment of the present invention.

3 is a flowchart for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

4 is a diagram illustrating a procedure of synthesizing a mask pattern in the mask pattern synthesizing apparatus according to one embodiment of the present invention.

5 is a diagram illustrating a configuration of a character cutting device according to an embodiment of the present invention.

6 is a flowchart for explaining the operation procedure of the character cutting device according to the embodiment of the present invention.

7A to 7D are views for concretely explaining a character cutting process according to an embodiment of the present invention.

8 is a flowchart for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

9A to 9E are views for explaining in detail the character cutting process according to the embodiment of the present invention.

10 and 11 are flowcharts for explaining the operation procedure of the mask pattern synthesizing apparatus according to the embodiment of the present invention.

12 is a diagram illustrating a configuration of a character cutting device according to an embodiment of the present invention.

Fig. 13 is a flowchart for explaining a detailed operational procedure of step S604 according to the embodiment of the present invention.

14A to 14E are views for explaining in detail the character cutting process according to the embodiment of the present invention.

15A to 15E are diagrams for explaining conventional error correction processing for character truncation.

16 and 17 are diagrams for explaining conventional error correction processing for character truncation.

<Explanation of symbols for the main parts of the drawings>

101: computer

103: main memory device

105: video controller

106: monitor

113: mouse

114; keyboard

116: Printer

117: Scanner

119: touch panel

200: sentence information

201: sentence information input unit

202: character pattern generation unit

203: mask pattern storage unit

204: mask pattern selection unit

205: synthesis unit

206: page image

Claims (19)

An information processing apparatus for generating document image data based on sentence information composed of character codes, Storage means for storing a plurality of different mask patterns; Input means for inputting the sentence information; Character pattern generation means for generating a corresponding character pattern on the basis of a character code constituting a sentence in the sentence information input by the input means; Selecting means for selecting from the storage means a mask pattern different from the mask pattern selected by the character code preceding one of the character strings of interest when generating a character pattern corresponding to the character string of interest; Synthesizing means for synthesizing a character pattern corresponding to the character code of interest and the mask pattern selected by the selection means; An information processing apparatus, comprising: a. The method of claim 1, The storage means stores two mask patterns, The selection means selects one of the two mask patterns alternately each time a character pattern for one character is generated. The method of claim 1, The storage means stores three mask patterns, The selection means selects any one of the mask patterns from the first mask pattern or the second mask pattern among the three mask patterns when the next character is not a sign every time a character pattern for one character is generated. And when the next character is a sign, a third mask pattern different from the first mask pattern and the second mask pattern is selected. The method according to any one of claims 1 to 3, The synthesizing means outputs to the printing means. As the image processing apparatus which cuts out each character image for the post-processing of a character image unit from the page image output by the information processing apparatus of Claim 1, Storage means for storing a plurality of different mask patterns; Input means for inputting page images, Creation means for projecting the input page image in the vertical and horizontal directions to create a histogram indicating the distribution of significant pixels; Temporary extraction means for extracting a region of the pixel of interest as a temporary character image region based on the created histogram; Mask pattern determination means for determining which mask pattern stored in the temporary character image area extracted by said temporary extraction means is combined with which mask pattern stored in said storage means; Determination means for determining whether the type of the mask pattern of the temporary character image of interest determined by the mask pattern determination means and the type of the mask pattern synthesized in the temporary character image at the immediately preceding position are the same; When the determination means determines that the kind of mask pattern of the temporary character image of interest and the temporary character image at the immediately preceding position is the same, the image of the area including the temporary character image of interest and the temporary character image located at the immediately preceding position is renewed. We update as temporary letter image, When the determination means determines that the types of mask patterns of the temporary text image of interest and the temporary character image at the immediately preceding position are different from each other, the temporary text image at the position immediately before the temporary text image of interest is cut out of the normal character image. Control means for outputting to the post-process as an image An image processing apparatus comprising: The method of claim 5, The post-processing output by the synthesizing means is a character recognition means. As a control method of the information processing apparatus which has the memory means which stores several mask patterns different from each other, and produces | generates document image data based on the sentence information comprised from a character code, An input step in which the input means inputs sentence information, A character pattern generation step of generating a character pattern to which a character pattern generation means corresponds based on a character code constituting a sentence in the sentence information by the input step; In the case where the character pattern corresponding to the character code of interest is generated in the character pattern generation step, the selection means selects, from the storage means, a mask pattern different from the mask pattern selected by the character code preceding one of the character code of interest; Fair, A combining step in which the combining means synthesizes the character pattern corresponding to the character code of interest and the mask pattern selected by the selection means; Control method of the information processing apparatus comprising a. The method of claim 7, wherein Two mask patterns are stored in the storage means, The selecting step is a control method for an information processing apparatus, characterized in that one of the two mask patterns is alternately selected each time a character pattern of one character is generated. The method according to claim 7 or 8, And the synthesizing step is output to printing means. Control of the image processing apparatus which is equipped with the memory means which memorize | stores several mask patterns different from each other, and cuts out each character image for the post-processing of a character image unit from the page image output by the information processing apparatus of Claim 1. As a method, An input step in which the input means inputs the page image, A creation step in which the creation means projects the input page image in the vertical and horizontal directions to create a histogram indicating the distribution of pixels of interest; A temporary extraction step, wherein the temporary extraction means extracts the area of the pixel of interest as a temporary character image area based on the created histogram; A mask pattern determination step of determining, by the mask pattern discrimination means, which mask pattern stored in the temporary means image extracted in the temporary extraction step is combined with; A judging step of the judging means judging whether or not the type of the mask pattern of the temporary character image of interest determined by the mask pattern determination step is identical to the type of the mask pattern synthesized in the temporary character image at the immediately preceding position; When the control means determines that the kind of mask pattern of the temporary character image of interest and the temporary character image of the previous position is the same by the determination step, the area including the temporary character image of interest and the temporary character image located at the immediately preceding position. To update the image as a new temporary character image, In the case where it is determined by the determination step that the type of mask pattern of the temporary character image of interest and the temporary character image at the immediately preceding position is different, the temporary character image at the position immediately before the temporary character image of interest is cut out of the normal character image. Control process output to post-processing as image Control method of an image processing apparatus comprising a. An information processing apparatus for generating document image data based on sentence information composed of character codes, Storage means for storing a plurality of different mask patterns; An input means for inputting sentence information, Character pattern generation means for generating a corresponding character pattern on the basis of a character code constituting a sentence in the sentence information input by the input means; Judging means for judging whether or not the distribution area of the independent significant pixels of the attention character pattern is composed of two or more; A first output means for outputting the character pattern of interest as part of the page image, when it is determined by the determination means that the distribution region of the independent significant pixels of the character of interest is composed of one; In the case where it is determined by the determining means that the distribution area of the independent significant pixels of the attention character pattern is composed of two or more, the mask pattern different from the mask pattern selected by the character pattern before one of the attention character pattern is described. Selection means for selecting from storage means, Second output means for synthesizing the character pattern of interest and the mask pattern selected by the selection means, and outputting the synthesized character pattern as part of the page image; An information processing apparatus, comprising: a. The method of claim 11, The determining means creates a horizontal or vertical projection histogram of the pixels of interest forming the character pattern of interest, and judges whether or not there are two or more independent areas of distribution of significant pixels based on the histogram. An information processing apparatus. The method of claim 11, Further comprising dictionary storage means for storing dictionary information indicating a character code indicating a character having two or more independent distribution areas of the pixel of interest, The determining means determines whether or not the distribution region of the independent significant pixels of the attention character pattern is composed of two or more by detecting whether the attention character code is registered in the dictionary information. . The method of claim 11, The selection means selects a mask pattern different from the mask pattern previously selected for the character pattern of interest whose character pattern immediately following is a sign. The method of claim 11, The selection means selects a mask pattern different from the mask pattern previously selected for the character image of interest having an image other than the character in the periphery. As an image processing apparatus which cuts out each character image for post processing of a character image unit from the page image output by the information processing apparatus of Claim 11, Storage means for storing a plurality of different mask patterns; Input means for inputting page images, Creation means for projecting the input page image in the vertical and horizontal directions to create a histogram indicating the distribution of pixels of interest; Temporary extraction means for extracting a region of the pixel of interest as a temporary character image region based on the created histogram; Mask pattern discrimination means for discriminating whether or not the temporary character image in the temporary character image area is combined with any one of the mask patterns stored in the storage means, and identifying the mask pattern if synthesized; When the mask pattern discriminating means determines that the temporary text image of interest is not synthesized with any of the mask patterns stored in the storage means, the temporary text image of interest is regarded as a cutout image of the normal text image. Output to processing, When it is determined that the temporary text image of interest determined by the mask pattern discriminating means is synthesized by the mask pattern, the temporary text image synthesized by the same mask pattern is included among the temporary text images adjacent to the attentional text image. Output means for outputting a region to be processed to a post-processing as a cut-out image of a normal character image An image processing apparatus comprising: As a control method of an information processing apparatus, comprising: a storage means for storing a plurality of different mask patterns, and generating document image data on the basis of sentence information composed of character codes constituting a sentence, An input step in which the input means inputs sentence information, A character pattern generation step of generating a character pattern corresponding to the character pattern generation means in a sequence of character codes constituting a sentence in the sentence information input by the input step; A judging step, wherein the judging means judges whether or not the distribution area of the independent significant pixels of the attention character pattern is composed of two or more; In the case where it is determined by the determination step that the target character pattern is composed of one independent distribution area of the significant pixels, the first output unit outputs the target character pattern as part of the page image; In the case where it is determined by the determination step that the distribution area of the independent significant pixels of the target character pattern is composed of two or more, the selection step of selecting a mask pattern different from the mask pattern previously selected by the selection means from the storage means; , A second output step in which the second output means synthesizes the character pattern of interest and the mask pattern selected in the selection step, and outputs the synthesized character pattern as part of the page image. Control method of the information processing apparatus comprising a. Control of the image processing apparatus which is equipped with the memory means which memorize | stores several mask patterns different from each other, and cuts out each character image for the post-processing of a character image unit from the page image output by the information processing apparatus of Claim 11. As a method, An input step in which the input means inputs the page image, A creation step in which the creation means projects the input page image in the vertical and horizontal directions to create a histogram indicating the distribution of pixels of interest; A temporary extraction step, wherein the temporary extraction means extracts the area of the pixel of interest as a temporary character image area based on the created histogram; The mask pattern discriminating means judges whether or not the temporary character image extracted in the temporary extraction process is combined with any one of the mask patterns stored in the storage means, and when it is synthesized, a mask for identifying the mask pattern. Pattern discrimination process, When it is determined by the mask pattern determination step that the temporary text image of interest is not synthesized with any of the mask patterns stored in the storage means, the output means cuts out the temporary text image of interest by the normal character image. Output to post-processing as an image, When it is determined that the temporary text image of interest determined by the mask pattern discrimination process is synthesized by the mask pattern, the output means is constructed by the same mask pattern among the temporary text images adjacent to the temporary text image of interest. An output step of outputting a region containing a character image to a post-process as a cut-out image of a regular character image Control method of an image processing apparatus comprising a. A computer-readable storage medium storing a computer program for executing the process according to any one of claims 7, 8 and 17.

Images