JP2011137908A - Image processing device, image processing method, and map providing system - Google Patents

Abstract

To improve character readability in an image including a character image representing a character.
An image processing apparatus for processing image data representing an image including a character image representing a character includes a character region identifying means for identifying a character region that is a region including a character image in the image, and at least one of the identified character regions. For one, replacement means for replacing the image data of the character region with image data obtained by interpolation based on image data around the character region, and the character represented in the replaced character region with a character size Rearrangement means for enlarging and rearranging the image on the image.
[Selection] Figure 3

Description

  The present invention relates to an image processing technique for processing image data representing an image including a character image representing a character.

  In recent years, services that can acquire desired map information from a map information providing site on the Internet have become widespread. When using such a service, for example, the user receives map data from a map information providing site, displays it on a display device (for example, a display monitor connected to a computer), and confirms the contents. The map data is output to the printer to obtain a map image. When a map image is obtained in this way, a map image that displays more detailed information over a wide range is desired. Patent Document 1 below discloses a technique for printing a map image having a larger scale than a map image having a smaller scale displayed on the monitor in the same geographical range as the map image displayed on the monitor. With this technique, a map image showing detailed information in a wide range can be obtained.

  However, in the technique of Patent Document 1, although the amount of information is increased, the size of characters displayed on the map image is reduced, which causes a problem that it is difficult to read the characters. Moreover, since the characters on the map image are generally provided integrally with other images as character images representing the characters, data processing such as simply enlarging and displaying only the character data cannot be performed. It was. Such a problem is not limited to a map image, but is a problem common to images including character images representing characters. For example, when an image including a character image representing a character is reduced and output, the character size becomes small and the character may be difficult to read.

JP 2008-281840 A JP 2002-358002 A JP 2009-147562 A

  In view of at least a part of the above-described problems, the problem to be solved by the present invention is to improve the readability of characters in an image including a character image representing characters.

  The present invention has been made to solve the above-described problems, and can be realized as the following forms or application examples.

Application Example 1 An image processing apparatus that processes image data representing an image including a character image representing a character,
A character region identifying means for identifying a character region which is a region including the character image in the image;
For at least one of the identified character regions, replacement means for replacing image data of the character region with image data obtained by interpolation based on image data around the character region;
An image processing apparatus comprising: a rearrangement unit configured to rearrange the characters represented in the replaced character area in the image with a character size enlarged.

  The image processing apparatus having such a configuration identifies a character region in the image and replaces at least one image data of the character region by interpolating based on surrounding image data. Then, the characters represented in the replaced character area are enlarged and rearranged in the image. Accordingly, it is possible to improve the readability of the characters in the image by displaying the characters existing as images larger than the image including the character image representing the characters. In addition, since the characters are rearranged after the character region is replaced by interpolation, it is possible to suppress a decrease in the value of the image.

Application Example 2 The image processing apparatus according to Application Example 1, wherein the image is a map image.
The image processing apparatus having such a configuration can be suitably applied to a map image. Since map images generally include character images representing characters, that is, map images are not separated from image information and character information, so processing that simply enlarges and displays only character information However, according to the configuration of this application example, it is possible to display characters in an enlarged manner and improve readability suitably. In addition, since the characters are rearranged after the character region is replaced by interpolation, it is possible to suppress a decrease in the value as a map.

Application Example 3 The image processing apparatus according to Application Example 2, wherein the rearrangement unit performs rearrangement of the characters while avoiding overlapping of the rearranged characters.
In the image processing apparatus having such a configuration, characters to be rearranged are not superimposed and displayed. Therefore, it becomes easy to read the rearranged characters.

[Application Example 4] The image processing apparatus according to Application Example 2 or Application Example 3, further comprising mark identifying means for identifying a target mark represented in the map image, wherein the relocation means identifies the identified An image processing apparatus that avoids superimposition of a target mark and the character to be rearranged and rearranges the character.
In the image processing apparatus having such a configuration, the target mark is not hidden by the rearranged characters. Therefore, the function as a map does not deteriorate.

Application Example 5 The image processing apparatus according to Application Example 4, further comprising a determination unit that determines the importance of the character represented in the identified character area for each character area, and the rearrangement unit The image processing apparatus that changes the contents of the rearrangement according to the determined degree of importance.
Since the image processing apparatus having such a configuration changes the content of the rearrangement according to the degree of importance of the character represented in the character area, it is possible to rearrange the characters more flexibly.

Application Example 6 The image processing apparatus according to Application Example 5, wherein the rearrangement unit preferentially enlarges and rearranges characters having a relatively high degree of importance determined.
In the image processing apparatus having such a configuration, since characters with high importance are preferentially enlarged, priority is given to readability of important characters, and convenience is improved.

Application Example 7 In the image processing apparatus according to Application Example 5 or Application Example 6, the replacement unit performs the replacement for all of the identified character areas, and the rearrangement unit determines the importance determined An image processing apparatus that rearranges only characters having a relatively high degree.
The image processing apparatus having such a configuration replaces all identified character areas and rearranges only important characters. Therefore, an easy-to-view map image can be obtained while suppressing a decrease in the information presentation function of the map image.

Application Example 8 The image processing apparatus according to Application Example 4, further including a determination unit that determines the importance of the character represented in the identified character area for each character area, and the replacement unit includes An image processing apparatus that performs the replacement only on a character area in which a character having a relatively high degree of importance is represented.
The image processing apparatus having such a configuration replaces only a character area in which an important character is represented. That is, important characters are erased, enlarged and rearranged, while other characters are held as character images of the original size. Therefore, it is possible to obtain an easy-to-see map image for important information without degrading the information presentation function of the map image.

Application Example 9 In the image processing apparatus according to any one of Application Example 5 to Application Example 8, the determination unit includes a unit that determines a character size of a character represented in the identified character region, An image processing apparatus that determines that a character having a larger character size has a higher degree of importance.
The image processing apparatus having such a configuration determines that the degree of importance is higher as the character size of the character represented in the character area is larger. In a map image, important information such as a municipality name is generally displayed in relatively large characters, and according to such a configuration, the degree of importance can be accurately determined.

[Application Example 10] The image processing apparatus according to any one of Application Example 5 to Application Example 9, wherein the determination unit determines whether or not the identified target mark is related to the identified character area. Means for determining that the character represented in the character area having the relevance to the target mark is more important than the character represented in the character area having no relevance to the target mark. Image processing device.
The image processing apparatus having such a configuration determines that the character represented in the character region having a relationship with the target mark is highly important. In the map image, the target mark is an important mark, so with such a configuration, the degree of importance can be accurately determined.

Application Example 11 In the image processing device according to any one of Application Example 5 to Application Example 10, the determination unit receives a point on the map image designated by a user of the image processing device, Means for determining whether or not there is a relationship between the received point and the identified character region, and the character represented in the character region having the relationship with the received point is related to the received point. An image processing apparatus that determines that the degree of importance is higher than that of a character represented in a character area that is not included.
The image processing apparatus having such a configuration determines that a character represented in a character region having a relationship with a point on a map image designated by a user has a high degree of importance. Accordingly, it is possible to accurately determine the degree of importance by appropriately reflecting the user's intention.

Application Example 12 In the image processing apparatus according to claim 4, further comprising means for determining whether or not the identified target mark is associated with the identified character area, and the rearrangement means includes: An image processing apparatus for rearranging characters represented in the character area determined to have no relevance to the target mark with the same position as the center position of the character area as a center position.
Since the image processing apparatus having such a configuration rearranges the characters in the character area not related to the target mark so that the center position of the characters does not change, the position on the map image indicated by the characters is erroneously recognized. There is no.

[Application Example 13] The image processing apparatus according to any one of Application Example 1 to Application Example 12, further comprising means for storing character data representing a predetermined character, wherein the character area identification means is a pattern The character represented by the character image is identified by matching to identify the character region, and the rearrangement unit performs the rearrangement using the character data corresponding to the character specified by the pattern matching. Image processing device.
Since the image processing apparatus having such a configuration displays characters specified by pattern matching using character data, the image quality of the characters does not deteriorate due to the rearrangement of characters. As a result, it is possible to provide a map image in which characters are easy to read.

Application Example 14 When the relevance ratio in the pattern matching is a predetermined value or less, the rearrangement unit uses data obtained by binarizing the image data of the identified character area instead of the character data. An image processing apparatus according to Application Example 13 in which the rearrangement is performed.
Since the image processing apparatus having such a configuration performs rearrangement using the binarized data of the image data of the identified character area when the precision in pattern matching is equal to or less than a predetermined value, the image processing apparatus erroneously performs pattern matching. It can suppress that the specified character is displayed. In addition, since the rearrangement is performed using the binarized data of the image data of the character area, the character readability can be sufficiently ensured.

[Application Example 15] When the relevance ratio in the pattern matching is equal to or less than a predetermined value, the rearrangement unit uses the enlarged image of the identified character area instead of the character data. An image processing apparatus according to Application Example 13 in which
Since the image processing apparatus having such a configuration performs rearrangement using an image obtained by enlarging the identified character area when the matching rate in pattern matching is equal to or less than a predetermined value, a character incorrectly specified by pattern matching is displayed. It can be suppressed. In addition, since the original image is enlarged and rearranged, characters faithful to the original image can be displayed.

In addition to the configuration as an image processing apparatus, the present invention can also be realized as an image processing method of Application Example 16, a map providing system of Application Example 17, an image processing program, a storage medium storing the program, and the like. .
Application Example 16 An image processing method for processing image data representing an image including a character image representing a character, the character region being a region including the character image in the image is identified, and the character region of the identified character region For at least one, the image data of the character area is replaced with image data obtained by interpolation based on image data around the character area, and the character represented by the replaced character area is set to a character size. An image processing method for enlarging and rearranging the image.
Application Example 17 A map providing system for providing a map image, wherein an acquisition unit that acquires image data representing a map image and a character region that is a region including a character image representing a character in the map image are identified. A character area identifying means; and a replacement means for replacing image data of the character area with image data obtained by interpolation based on image data around the character area for at least one of the identified character areas; A map providing system comprising: a rearrangement unit that rearranges a character represented in a replaced character region on the map image by enlarging a character size; and an output unit that outputs the rearranged map image.

2 is an explanatory diagram showing a system configuration of a map providing system 20. FIG. 2 is an explanatory diagram showing a schematic configuration of a computer 30. FIG. It is a flowchart which shows the flow of a readability improvement process. It is a flowchart which shows the detail of the character recognition process in a readability improvement process. It is explanatory drawing which shows the specific example of the image used as the process target of the character recognition process in a readability improvement process. It is explanatory drawing which shows a mode that the position of the character was detected from the image of a process target. It is a flowchart which shows the detail of the character erasure | elimination process in a readability improvement process. It is explanatory drawing which shows a mode that the character deletion process was performed with respect to the image of a process target. It is a flowchart which shows the detail of the character rearrangement process in a readability improvement process. It is a flowchart which shows the detail of the arrangement | positioning coordinate determination process in a character rearrangement process. It is explanatory drawing which shows the specific example of an arrangement | positioning coordinate determination process. It is explanatory drawing which shows the specific example of an arrangement | positioning coordinate determination process. It is explanatory drawing which shows the specific example of an arrangement | positioning coordinate determination process. It is explanatory drawing which shows the specific example of an arrangement | positioning coordinate determination process. It is explanatory drawing which shows the result of having performed the readability improvement process with respect to the process target image.

A. Example:
Examples of the present invention will be described.
A-1. System configuration:
A schematic configuration of a map providing system 20 as an embodiment of the present invention is shown in FIG. The map providing system 20 according to the present embodiment is a system for downloading and printing map information from a map information providing service connected to the Internet INT. As shown in FIG. 1, the map providing system 20 is configured by connecting a computer 30 as an image processing device and a printer PRT as an output device via a local area network LAN1. The local area network LAN1 is connected to the local area network LAN2 via the Internet INT, and a server SV that provides map information is connected to the local area network LAN2. Note that the connection means of the computer 30, the printer PRT, and the server SV is not limited to the above example, and may be connected by a communicable communication means. For example, the computer 30, the printer PRT, and the server SV may be connected to the local area network LAN1, or the computer 30 and the printer PRT may be locally connected.

A-2. Equipment configuration:
A schematic configuration of the computer 30 is shown in FIG. The computer 30 is a general-purpose personal computer in which a predetermined program is installed, and includes a CPU 40, a hard disk drive 50, a ROM 61, a RAM 62, an input mechanism 63, a display 64, and an interface 65, as shown in the figure. Connected with.

  The CPU 40 controls the entire computer 30 by developing and executing firmware and OS stored in the hard disk drive 50 and the ROM 61 on the RAM 62. The CPU 40 also functions as a character area identification unit 41, a replacement unit 42, a rearrangement unit 43, a mark identification unit 44, and a determination unit 45 by executing a program stored in the hard disk drive 50. Details of these functional units will be described later.

  The hard disk drive 50 stores a printer driver for controlling the printer PRT connected to the local area network LAN1, a WEB browser program, and the like in addition to the program for realizing the functions of the above-described functional units. Not shown). The hard disk drive 50 stores character data 51. The character data 51 is data defining the shape of each character.

  The input mechanism 63 includes a keyboard and a pointing device (here, a mouse), and a user can give an operation instruction to the computer 30 via the input mechanism 63. The display 64 is a liquid crystal display. The interface 65 is an interface for connecting to the local area network LAN 1, and the computer 30 is connected to the local area network LAN 1 via the interface 65.

  The server SV is a WEB server that provides a map information providing service via the Internet INT. Map data, which is image data representing a map image, is stored in a storage medium included in the server SV. This map data is prepared for a plurality of predetermined scales. In the map data, characters included in the map image are provided as an image. That is, the character information and the image information are not separated and are configured as an integral image.

  The printer PRT is a general-purpose inkjet printer. The format of the printer PRT is not particularly limited, and may be a laser printer, for example.

  In the map providing system 20, when a user operates the computer 30 to activate a WEB browser and designates a desired point, scale, etc., the computer 30 transmits a transmission request designating the point, scale, etc. When the server SV receives a transmission request from the computer 30, the server SV cuts out data corresponding to the request from the stored map data and returns it to the computer 30. The map data obtained in this way is displayed on the display 64 of the computer 30 and printed by the printer PRT through a user operation.

A-3. Readability improvement processing:
The readability improving process in the map providing system 20 described above will be described. The readability improving process is a process for improving the readability of characters on the map image represented by the map data acquired by the computer 30, and is executed in the computer 30 before the map image is printed by the printer PRT. The flow of the readability improving process in the present embodiment is shown in FIG. When this process is started, the CPU 40 of the computer 30 first performs a character recognition process as a process of the character area identification unit 41 (step S100). The character recognition process is a process for recognizing a character area existing on a map image represented by map data to be processed. When the character recognition process is performed, the CPU 40 performs a character erasure process as the process of the replacement unit 42 (step S200). The character erasing process is a process of replacing the recognized image data of the character area with image data obtained by interpolation based on image data around the character area, that is, a process of erasing the character display. When the character erasing process is performed, the CPU 40 performs the character rearrangement process as the process of the rearrangement unit 43 (step S300). The character rearrangement process is a process of enlarging the character represented in the replaced character area and rearranging it in the image replaced in step S200. When the character rearrangement process is performed in this manner, the readability improvement process is completed. Details of steps S100 to S300 will be described below.

A-3-1. Details of character recognition processing:
The details of the character recognition process (step S100) are shown in FIG. When the character recognition process is started, the CPU 40 first scans image data to be processed from the beginning (step S110).

  When scanning the image data, the CPU 40 detects the position of the character on the image represented by the image data (step S120). In the present embodiment, the character area including the character image is detected as a rectangular area using the peripheral distribution and the line density of the image. This method is a well-known technique as one method of OCR (Optical Character Reader), and detailed description is omitted. However, only pixels representing characters are separated from pixels constituting an image, and characters in a predetermined region are separated. This is a technique for summing the pixel values to be expressed in the horizontal and vertical directions, determining the pixel distribution, and specifying each character in a rectangular area. For example, “peripheral distribution, line density, outline Region segmentation of document images using rectangular features This is described in IEICE Transactions '86 / 8 Vol.J69-D NO.8 '. In addition, since it is common that the character represented on a map uses a specific color (for example, black and a several colors may be used) according to a map, from the image data, the character of the said color is used. By separating only data, it is easy to separate only pixels representing characters. Note that when the image data to be processed is an image read by a scanner or the like, there is a problem of reading accuracy. Therefore, it is only necessary to separate pixels whose color gradation values are within a predetermined range.

  However, even in a map image, pixels representing characters may not be sufficiently separated only by color information. For example, when pixels representing black characters are separated, black is used for a graphic representing a track. In such a case, a method of separating pixels representing characters by using the aspect ratio of the characters or the property that the characters appear continuously may be used. Such a method is described in, for example, “Text Detection from Natural Scene Images: Towards a System for Visually Impaired Persons, Nobuo Ezaki et al. International conference on pattern recognition 2004”. In a map image, since the font type and size of characters are known, such a method is particularly effective.

  Thus, when the rectangular area is detected for each character, the CPU 40 determines whether the characters represented in these rectangular areas are continuous characters representing one place name or the like based on the distance between the rectangular areas, And whether it is an independent character. Specifically, if the distance between the rectangular areas is equal to or less than a predetermined value (for example, 3 pixels), it is determined that the characters represented in these rectangular areas are continuous, and if the distance is greater than the predetermined value, It is determined that the characters represented in the rectangular area are independent. As a result of the determination, if the characters are continuous, the rectangular areas are connected and detected as one character area. Since characters on the map image may be displayed in a plurality of stages due to restrictions on character arrangement, such a connection process is performed not only in the horizontal direction but also in the vertical direction. However, the rectangular areas may be grouped instead of being connected. Grouping here refers to handling a plurality of separated rectangular areas as a group.

  The process of step S120 will be described visually with reference to FIGS. FIG. 5 shows a specific example of an image to be processed in the character recognition process (step S100). As shown in the figure, in the processing target image, characters “AAA”, “CCCCC (line feed) CCCC (line feed) CCCC”, “DDD” (line feed) that indicate these names are associated with the target marks TM1, TM2, and TM3. ) DDDDD ”is displayed. A character “BBBB (line feed) BB” indicating the name of the intersection IS is displayed in the vicinity of the intersection IS. In this embodiment, the target mark refers to a figure, a symbol, or the like representing a predetermined facility or equipment (station, hospital, school, hotel, store, signal, etc.) on a map image. For example, the target mark TM1 is a figure representing a station. The target marks TM2 and TM3 are figures representing facilities such as stores.

  FIG. 6 shows a state in which the process of step S120 is performed on the processing target image. In this example, characters associated with the target marks TM1 to TM3 are detected as character areas RA1 to RA3, respectively. A character indicating the intersection IS is detected as a character region RA4. The character areas RA2 to RA4 are detected as character areas connected over a plurality of stages by the above-described processing.

  Thus, when the position of the character is detected as a character area, the CPU 40 detects the character size represented in the character area (step S130). Here, in step S120, the character size is specified by the width of the rectangular area detected for each character.

  When detecting the character size, the CPU 40 extracts character information (step S140). This process is a process for identifying what character, for example, “A” or “B” is displayed in the character area, and is performed using a well-known OCR technique. That is, the feature amount prepared for each character is compared with the feature amount of each character included in the character region, and each character included in the character region is specified by pattern matching.

  When the character information is extracted, the CPU 40 repeats the processes of steps S120 to S140 (step S150) until character information of all the characters included in the processing target image is extracted, and ends the character recognition process.

A-3-2. Details of character erasure processing:
When the character recognition process (step S100) is performed as described above, the CPU 40 performs a character erasure process (step S200). Details of the character erasure process are shown in FIG. When the character erasing process is started, the CPU 40 first reads the position (coordinates) on the image for one of the character areas detected in step S100 (step S210). When reading the coordinates of the character area, the CPU 40 overwrites the monochrome mask on the pixels of the read character area (step S220). The color of the monochromatic mask used here is set to a color that is not used on the map image. Usually, since the color of the component of the map image acquired from the WEB site is known, the color of the monochromatic mask may be determined for each site from which the map image is acquired, or the map image is scanned and the map is scanned. Colors that are not used on the image may be specified and used. The site for acquiring the graphic image may be acquired from a WEB browser, or may be read when the image data includes information such as the creator or the copyright holder.

  When the monochromatic mask is overwritten, the CPU 40 performs interpolation erasure of the mask for the monochromatic masked area (step S230). In this process, the image data overwritten with the monochrome mask is replaced with image data obtained by interpolation based on pixel data around the monochrome mask. In this embodiment, the interpolation is performed using the well-known Inpainting method. The inpainting method is a technique for estimating the original image of a masked area based on image data within a fixed distance around the area, and interpolating the image data from the edge for a single color mask area. is there. Since no character image exists around the mask area, the interpolated (replaced) area is restored as a map image. Such an inpainting method is suitable for restoration of a map image because parameters suitable for a diagram are used. This inlining method is described in, for example, “Y. Wexler, E. Shechtman and M. Irani“ Space-Time Video Completion ”Computer Vision and Pattern Recognition (CVPR), Washington, June 2004.”. A specific example of the image represented by the image data replaced in step S230 is shown in FIG. As shown in the figure, the character areas RA1 to RA4 (see FIG. 6) are erased and restored as a map image that does not include characters.

  In this way, if interpolation is performed after applying a single-color mask, the CPU 40 can identify the interpolation target area only by referring to the image data, so there is no need to read the coordinates of the interpolation target area one by one. Can be speeded up. However, the step S220 may be omitted, and the image data after the interpolation may be directly overwritten on the image data in the interpolation target area, or after the image data in the interpolation target area is deleted and after the interpolation The image data may be overwritten.

  When the interpolation erasure of the mask is performed, the CPU 40 repeats the processes of steps S210 to S230 for all the character areas (step S240), and ends the character erasure process.

A-3-3. Details of character relocation processing:
When the character erasing process is performed as described above, the CPU 40 performs a character rearrangement process (step S300). This processing is performed by storing the processing result of step S100 in the RAM 62 and referring to the processing result. Details of the character rearrangement process are shown in FIG. When the character rearrangement process is started, the CPU 40 first selects one of the character areas from which the character has been erased by the character erasure process. It is determined whether or not there is (step S310). The character size used for this determination is the character size detected in step S130. The determination in step S310 is to determine the degree of importance of the character represented in the character area. In map images, in general, important information such as city names is often displayed in a relatively large character size. Therefore, the fact that the character size is a predetermined value or more is important. It can be treated as a high degree.

  As a result, if the character size is smaller than the predetermined value (step S310: NO), the CPU 40 omits steps S320 to S340 described later. On the other hand, if the character size is equal to or larger than the predetermined value (step S310: YES), the CPU 40 selects a character size having a preset ratio R (1 <R) (step S320). This ratio R is the ratio of the character size of the characters to be rearranged by the processing described later to the character size of the characters erased by the character erasing processing. In the present embodiment, the ratio R is predetermined as R = 1.5 in advance. The ratio R may be determined in advance for each scale of the map image so that the size of the characters to be rearranged is approximately the same as the characters displayed on the display 64. The scale of the map image may be acquired from the WEB browser that acquired the map image, or when the scale information is included in the image data, it may be read. In this way, it is possible to rearrange characters of an easy-to-read size regardless of the scale of the map image.

  When the character size is selected, the CPU 40 performs an arrangement coordinate determination process (step S330). The arrangement coordinate determination process is a process for determining the character arrangement coordinates and the character size when the character erased in step S200 is rearranged with the character size enlarged. Details of this processing will be described later. When the arrangement coordinates and the character size are determined, the CPU 40 rearranges the characters on the processing target image (step S340). This process reads character data corresponding to the character extracted in step S140 from the character data 51 stored in the hard disk drive 50, generates character data having the character size determined in step S330, and generates image data. Of these, the character data is superimposed on the arrangement coordinates determined in step S330, thereby rearranging the characters on the processing target image.

  When the characters are rearranged (step S340), or if the character size is smaller than the predetermined value (step S310: NO), the CPU 40 performs the above process for all the character areas detected in the character recognition process (step S100). The processes in steps S310 to S340 are repeated (step S350), and the character rearrangement process is terminated.

  Details of the above-described arrangement coordinate determination process (step S330) are shown in FIG. In the arrangement coordinate determination processing, first, the CPU 40 detects a target mark from the processing target image as processing of the mark identification unit 44 (step S331). Since the target mark is generally determined in a predetermined shape for each map image provider, for example, the target mark feature amount is stored in advance for each WEB site that provides the map image, and the map image is By performing pattern matching using the feature amount corresponding to the acquired WEB site, the target mark on the image can be detected.

  When the target mark is detected, the CPU 40 determines the relationship between the extracted target mark and the character area detected in step S100 as a process of the determination unit 45, and associates the target mark having the relationship with the character area. (Step S332). This process is a process for associating the target mark with a character area for displaying details (station name, facility name, etc.) of the target mark. Such association is also called a link. Note that the processes of steps S331 and S332 do not have to be executed as the arrangement coordinate determination process, and may be executed prior to the character rearrangement process, for example.

  In this embodiment, the detected target mark is linked to the character area closest to the target mark. However, the link method is not limited to such a configuration. For example, you may comprise as follows. First, various target marks and characters to be included in the details of the target marks (hereinafter also referred to as corresponding characters, for example, “station” characters for station target marks, and “ Small, medium, high, large, etc.). Then, a character area closest to the target target mark is searched for from among character areas including characters associated with a target mark to be linked (hereinafter also referred to as a target target mark), and the searched character area and target Link with the target mark. In this way, the accuracy of the link can be improved. Or you may select the character area linked with an object target mark from the character area of the character size more than predetermined value. Even in this case, the target mark is important information, and the character for displaying it is often displayed relatively large, so that the accuracy of the link can be improved. This step S332 can be said to be determination of the importance of the character represented in the character area. In the map image, the target mark is an important landmark to know the current position of the user and the route to the destination, so the characters associated with the target mark are considered to be relatively high in importance. Because you can.

  When the target mark and the character area are associated with each other, the CPU 40 enlarges the character represented in the character area having no link (not associated with the target mark) to the character size selected in step S320 together with the character area. (Step S333). The arrangement coordinates of the characters enlarged in step S333 are set so that the center position of the enlarged character is aligned with the center position of the character area, that is, the center position of the character before and after enlargement is not changed. To be set.

  When the character area without the link and the character are enlarged, the CPU 40 determines whether or not the enlarged character area is overlapped with another target mark or character area (step S334). As a result, if superimposition is performed (step S334: YES), the character size of a character without a link is reset to a smaller value until it is not superimposed (step S335).

  Specific examples of steps S333 to S335 are shown in FIGS. FIG. 11 shows the characters “EEEEEE” and “FFFFFF” represented in the character areas RA5 and RA6 and target marks TM5 and TM6 to be processed. The center position of the character area RA6 is C1. The character area RA5 and the target mark TM5 are linked by the above S332. The character area RA6 is not linked to any of the target marks TM5 and TM6. FIG. 12 shows a character area RA6 that has been reset to a size that does not overlap with the target mark TM6. Note that the center position of the enlarged character area RA6 remains the center position.

  On the other hand, if not superimposed (step S334: NO), the CPU 40 determines the character size of the character area with the link (associated with the target mark) while maintaining the character size of the character without the enlarged link. The character size selected in step S320 is enlarged and set (step S336). The arrangement coordinates of the characters enlarged in step S336 are set such that, for example, the center position of the enlarged characters is the center of the target mark so that the characters are enlarged from the center of the target mark toward the center of the character area. On the line connecting the center of the character area and the position farther from the target mark than the center of the character area.

  A specific example of step S336 is shown in FIG. FIG. 13 shows the character area RA5 and the target mark TM5 shown in FIG. On the straight line connecting the center position C2 of the target mark TM5 and the center position C3 of the character area RA5, the size of the vector starting from the intersection CP1 between the straight line and the character area RA5 and having the center position C3 as the end point is The distance is L1. Here, when the character “EEEEEE” is enlarged M times (M is a value larger than 1), as shown in FIG. 13B, the size of the vector is set to M times starting from the intersection CP1. That is, the position of the enlarged character area RA5 is set with the position C31 of the end point of the vector when the distance is L1 × M as the center position. Thus, if the character with a link is expanded and arrange | positioned, a superposition with the target mark TM5 and character area RA5 can be avoided, and a character can be rearranged suitably.

  When the character size of the linked character is set in this way, the CPU 40 determines whether or not the character region enlarged together with the character is superimposed on another target mark or character (step S337). As a result, if not superimposed (step S337: NO), the CPU 40 ends the arrangement coordinate determination process. On the other hand, if superimposition is performed (step S337: YES), the arrangement coordinates of the character with the link are set to a position obtained by rotating the center position of the enlarged character with the link by the angle θ with the center position of the target mark as a fulcrum. Re-execute (step S338).

  A specific example of step S338 is shown in FIG. In this example, as shown in FIG. 14, a vector having the center position C2 as the start point and the center position C31 as the end point is corrected with the end point position C32 when rotated by an angle θ as the center position of the character area RA5. . As described above, if the linked character is arranged by rotating the center position, it is possible to avoid the superimposition with other characters and appropriately rearrange the characters. When the above-described vector is rotated, if the character area RA5 enlarged by M times and the target mark TM6 overlap, the length of the vector may be increased. If superposition with other characters cannot be avoided no matter how the rotation is performed, the enlargement magnification of the character area RA5 may be reduced and the processes of steps S337 and S338 may be performed again. In addition, the above-described method can be applied to a process of storing an enlarged character in the image when the character arranged at the end of the image is enlarged.

  In this manner, the CPU 40 repeats the processes of steps S337 and S338 until the superposition can be avoided, and ends the arrangement coordinate determination process. As is clear from the above description, in the present embodiment, the size of a character with a link is set to be preferentially larger than the size of a character without a link. This is because, as described above, the important character is preferentially set larger because it is determined that the character with the link is more important than the character without the link.

  A specific example of a map image subjected to such readability improvement processing is shown in FIG. FIG. 15 shows the result of performing the readability improving process on the processing target image (see FIG. 5). As shown in the figure, it can be seen that all the characters are enlarged as compared with FIG. 5 and the readability is improved. Moreover, since each character is arrange | positioned so that it may not overlap with characters, readability is not impaired. Moreover, since each character is arrange | positioned so that it may not overlap with a target mark, the functionality as a map is not impaired.

A-4. effect:
In the map providing system 20 having such a configuration, the computer 30 identifies a character area in the map image, and replaces the character area by interpolating based on surrounding image data. Then, the characters represented in the replaced character area are enlarged and rearranged in the image. Therefore, it is possible to improve the readability of the characters in the map image output by the printer PRT by displaying the characters existing as the image larger than the map image including the character image representing the characters. In addition, since the characters are rearranged after the character regions are replaced by interpolation, it is possible to prevent the map image of the character region portion from being restored and the value as the map image from being lowered. Of course, such a configuration can be suitably applied to various images other than the map image as long as the character information and the image information are not separated.

  Further, since the computer 30 avoids superimposition of characters to be rearranged and sets the character placement position and character size, the characters to be rearranged are not superimposed and displayed, and the characters are easy to read. A map image can be provided. In addition, the computer 30 avoids superimposition of the target mark represented on the map image and the character to be rearranged, and sets the character placement position and character size. It is not hidden and the function as a map is not lowered.

  Further, the computer 30 determines the importance of the character represented in the identified character area for each character area, and changes the content of the rearrangement according to the determined degree of importance. Rearrangement is possible. For example, in the computer 30, characters having a high degree of importance are preferentially greatly enlarged, so that the readability of important characters is given priority and convenience is improved. Here, the computer 30 determines that the character with the link is more important than the character without the link. In the map image, since the target mark is an important landmark, the degree of importance can be determined with high accuracy. In addition, the computer 30 replaces all the identified character areas and rearranges only important characters. That is, only important character information is rearranged, and other characters are deleted. Therefore, an easy-to-view map image can be obtained while suppressing a decrease in the information presentation function of the map image. Here, the computer 30 determines that the degree of importance is higher as the character size of the character represented in the character area is larger. In a map image, important information such as a municipality name is generally displayed in relatively large characters, so that the degree of importance can be determined with high accuracy.

  In addition, the computer 30 rearranges the unlinked character so that the center position of the character does not change, so that the position on the map image indicated by the character is not mistaken. In addition, since the computer 30 displays the character specified by pattern matching using the character data 51, the image quality of the character is not deteriorated by the rearrangement of the character. As a result, it is possible to provide a map image in which characters are easy to read.

B. Variations:
A modification of the above embodiment will be described.
B-1. Modification 1:
In the above-described embodiment, the character erasure process is executed for all the character areas detected in step S120 to erase the characters, and the character rearrangement is performed only for characters having a character size equal to or larger than a predetermined value. Although the process is configured to be executed, the character rearrangement process may be executed for characters of all character sizes. In this way, all the characters displayed on the original map image are displayed by rearrangement, and the information presentation function of the original map image does not deteriorate.

B-2. Modification 2:
In the above-described embodiment, the character erasing process is executed on all the character areas detected in step S120 to delete the characters, and characters having a character size of a predetermined value or more are highly important. However, the determination of the importance level is not limited to the character size. For example, a character with a link is a link. Only characters with links may be rearranged because they are more important than characters without. Or in addition to the above-mentioned embodiment, it is good also as a structure which rearranges the character with a link irrespective of character size.

  In the above-described embodiment, the character in the character area linked to the target mark is determined to be more important than the character in the character area that is not linked, and the degree of importance is relatively high. Although the configuration is such that the characters are preferentially rearranged, the determination of the degree of importance is not limited to the presence or absence of the link. For example, characters with a relatively large character size are Characters having a relatively large character size may be preferentially rearranged with a large character size because the importance is high. Further, the ratio R used in step S320 may be changed according to the degree of importance. This makes it easier to see important characters.

  Further, the determination of the importance may be performed based on a user designation. For example, when the computer 30 acquires a map image from the server SV, it accepts a designated point (destination, departure point, etc.) on the map image input by the user via the WEB browser, and a character area related to the designated point. It may be determined that the degree of importance of the characters shown in is high. As a configuration for receiving the designated point, for example, the coordinate value of the designated point may be acquired from a WEB browser program. Alternatively, since the designated point is usually located at the center of the acquired map image, the center position may be detected from the map image that is the processing target of the readability improving process, and the coordinate value may be acquired. . Alternatively, the user may designate a desired point on the map image displayed on the display 64 using the input mechanism 63, and the computer 30 may acquire the coordinate value. As a configuration for determining the relevance between the designated point and the character area, for example, as in the above-described embodiment, the relationship between the designated point and the character area may be determined on the image. In such a case, it may be determined that only one character area is relevant to one designated point, and a plurality of character areas arranged within a predetermined range from one designated point are relevant. You may judge that. With such a configuration, it is possible to display large information on a designated point important to the user or its surroundings, and convenience is improved.

  Further, the determination of importance is not limited to one such as character size and presence / absence of a link, and may be performed by combining a plurality of determination elements. For example, the importance may be determined by combining character size, presence / absence of a link, and user designation. In such a case, the evaluation value may be calculated by weighting the suitability for each determination element, and the mode of the character erasure process or the character rearrangement process may be changed according to the evaluation value. In this way, the readability improving process can be performed more flexibly. The criterion for determining importance may be different for each process such as a character erasure process and a character rearrangement process.

B-3. Modification 3:
In the above-described embodiment, the character erasing process is executed on all the character areas detected in step S120 to delete the characters. However, characters with a relatively high degree of importance are displayed. Character erasure processing may be executed only for the character region that has been processed. In such a case, the determination of importance can use the various determination elements shown in the second modification. In this way, only important character information can be erased and rearranged, and other character information can be retained as a character image of the original size without being enlarged. Therefore, it is possible to obtain an easy-to-see map image for important information without degrading the information presentation function of the map image. Further, in such a case, the character can be easily read by avoiding the superimposition of the character that has not been subjected to the character erasing process and the character to be rearranged.

B-4. Modification 4:
In the above-described embodiment, all characters to be rearranged are rearranged using the character data 51. However, the characters may be rearranged by other methods. For example, for a character area where the matching rate in pattern matching performed in the character recognition process is a predetermined value or less, instead of the character data 51, data obtained by binarizing the image data of the original character area Rearrangement may be performed by superimposing the data. Note that when the image data to be processed is an image read by a scanner or the like, a pixel whose color gradation value is within a predetermined range may be binarized as a pixel representing a character. Alternatively, the rearrangement may be performed by superimposing an enlarged image of the original character area instead of the character data 51. With these configurations, it is possible to suppress the display of characters incorrectly specified by pattern matching. In addition, when using the binarized data of the image data of the character area, the character readability can be sufficiently secured.

B-5. Modification 5:
In the above-described embodiment, the character placement position and the character size are determined so that the characters do not overlap with each other and the target mark and the character do not overlap, and the characters are rearranged. However, prioritizing securing a large character size may be performed so that they are superimposed. Further, when the target mark and the character are superimposed, these data may be blended so that the character is displayed translucently.

B-6. Modification 6:
In the above-described embodiment, the character placement position and the character size are determined and the characters are rearranged so that the character regions including the characters do not overlap with each other. The position where the character is arranged and the character size may be determined so that the pixel is specified and the pixels do not overlap each other.

B-7. Modification 7:
In the above-described embodiment, the readability improving process with the same content is performed in all areas of the map image. However, the content of the process may be changed for each area. For example, the number of detected character areas, the area ratio occupied by the character area, and the like are calculated for each area of a predetermined area to determine the density of the character area. The reference value of the used character size may be increased. In this way, a map image that is easier to see can be provided. Alternatively, in the case where the designated point is received as in the above-described modification 4, the reference value of the character size used in step S310 is relatively set in an area within a predetermined range from the designated point. The reference value may be relatively small in other areas. Even in this case, a map image having a large amount of information is provided around a designated point important for the user, and an easy-to-view map image can be provided in other areas. Of course, in addition to or instead of this configuration, the character size selected in step S320 may be changed for each region.

B-8. Modification 8:
In the above-described embodiment, the readability improving process is executed by the computer 30 as the image processing apparatus. However, all or part of the readability improving process may be executed by the server SV or the printer PRT. May be. That is, what is necessary is just to be performed as the map provision system 20 whole.

B-9. Modification 9:
In the above-described embodiment, the map providing system 20 has shown the configuration in which the map data that is the target of the readability improving process is acquired from the server SV. However, the configuration is not limited to this configuration. For example, the map providing system 20 may be configured to acquire map data stored in the hard disk drive 50 of the computer 30 or acquire map data from an external storage device (for example, a DVD drive) connected to the computer 30. It is good also as composition to do.

B-10. Modification 10:
In the above-described embodiment, an example in which the computer 30 is configured as an image processing apparatus has been described. However, the image processing apparatus of the present invention can also be realized as a mobile phone, a PDA (Personal Digital Assistant), or the like. Moreover, although the example which uses the output device with which the map provision system 20 is provided as a printer was shown, it is good also as a display, a projector, etc. Here, the output device may be an output device included in the image processing apparatus. For example, the map providing system 20 is realized as a car navigation system of a type that provides a map image in which character information and image information are integrated, an enlarged display system for low-sighted persons, and low-sighted persons. be able to. Of course, the present invention can be applied not only to a map image but also to various images including a character image representing a character. For example, an image in which characters of an image read by a scanner are enlarged and rearranged may be displayed by a projector or the like.

  As mentioned above, although embodiment of this invention was described, elements other than the element described in the independent claim among the components of this invention in embodiment mentioned above are additional elements, and are suitably abbreviate | omitted or combined. Is possible. In addition, the present invention is not limited to such an embodiment, and it is needless to say that the present invention can be implemented in various modes without departing from the gist of the present invention. For example, the present invention can be realized as an image processing method, an image processing program, a storage medium storing the program, and the like in addition to the configuration as an image processing apparatus and a map providing system.

20 ... Map providing system 30 ... Computer 40 ... CPU
DESCRIPTION OF SYMBOLS 41 ... Character area identification part 42 ... Replacement part 43 ... Relocation part 44 ... Mark identification part 45 ... Judgment part 50 ... Hard disk drive 51 ... Character data 61 ... ROM
62 ... RAM
63 ... Input mechanism 64 ... Display 65 ... Interface LAN1, LAN2 ... Local area network INT ... Internet SV ... Server PRT ... Printer TM1-TM6 ... Target mark IS ... Intersection RA1-RA6 ... Character area C1, C2, C3, C31, C32 ... Center position L1 ... Distance CP1 ... Intersection

Claims (17)

An image processing apparatus that processes image data representing an image including a character image representing a character,
A character region identifying means for identifying a character region which is a region including the character image in the image;
For at least one of the identified character regions, replacement means for replacing image data of the character region with image data obtained by interpolation based on image data around the character region;
An image processing apparatus comprising: a rearrangement unit configured to rearrange the characters represented in the replaced character area in the image with a character size enlarged.   The image processing apparatus according to claim 1, wherein the image is a map image.   The image processing apparatus according to claim 2, wherein the rearrangement unit performs rearrangement of the characters while avoiding overlapping of the rearranged characters. An image processing apparatus according to claim 2 or claim 3, wherein
Furthermore, it comprises a mark identifying means for identifying the target mark represented in the map image,
The image processing apparatus, wherein the rearrangement unit performs rearrangement of the character while avoiding overlapping of the identified target mark and the character to be rearranged. The image processing apparatus according to claim 4,
And a determination unit that determines the importance of the character represented in the identified character region for each character region,
The rearrangement unit changes the content of the rearrangement according to the determined degree of importance.   The image processing apparatus according to claim 5, wherein the rearrangement unit preferentially enlarges and rearranges characters having a relatively high degree of importance determined. The image processing apparatus according to claim 5 or 6,
The replacement means performs the replacement for all the identified character regions,
The rearrangement unit rearranges only characters having a relatively high degree of importance determined. The image processing apparatus according to claim 4,
And a determination unit that determines the importance of the character represented in the identified character region for each character region,
The image processing apparatus that performs the replacement only on a character region in which a character having a relatively high degree of importance is represented. An image processing apparatus according to any one of claims 5 to 8,
The determination means includes
Means for determining a character size of a character represented in the identified character region;
An image processing apparatus that determines that a character having a larger character size has a higher degree of importance. An image processing apparatus according to any one of claims 5 to 9,
The determination means includes
Means for determining whether or not the identified target mark and the identified character area are related;
An image processing apparatus that determines that a character represented in a character area having an association with the target mark has a higher degree of importance than a character represented in a character area having no association with the target mark. An image processing apparatus according to any one of claims 5 to 10,
The determination means includes
Means for receiving a point on the map image designated by a user of the image processing apparatus;
Means for determining whether or not there is a relationship between the received point and the identified character area;
Image processing that determines that the character represented in the character area having relevance to the received point is higher in importance than the character represented in the character area having no relevance to the received point apparatus. The image processing apparatus according to claim 4,
And a means for determining whether or not the identified target mark is related to the identified character area.
The rearrangement unit rearranges the character represented in the character area determined to have no relevance to the target mark with the same position as the center position of the character area as a central position. . An image processing apparatus according to any one of claims 1 to 12,
Furthermore, a means for storing character data representing a predetermined character is provided,
The character area identification means identifies the character represented by the character image by pattern matching, identifies the character area,
The rearrangement unit performs the rearrangement using the character data corresponding to the character specified by the pattern matching.   The rearrangement unit uses the binarized data of the image data of the identified character area instead of the character data when the relevance ratio in the pattern matching is a predetermined value or less. The image processing device according to claim 13.   The said rearrangement means performs the said rearrangement using the image which expanded the identified character area instead of the said character data, when the precision in the said pattern matching is below a predetermined value. The image processing apparatus described. An image processing method for processing image data representing an image including a character image representing a character,
Identifying a character region that is the region containing the character image in the image;
For at least one of the identified character regions, the image data of the character region is replaced with image data obtained by interpolation based on image data around the character region,
An image processing method for enlarging a character size and rearranging the characters represented in the replaced character area in the image. A map providing system for providing a map image,
Acquisition means for acquiring image data representing a map image;
Character area identifying means for identifying a character area that is an area including a character image representing a character in the map image;
For at least one of the identified character regions, replacement means for replacing image data of the character region with image data obtained by interpolation based on image data around the character region;
Relocation means for enlarging a character size and relocating the character represented in the replaced character area on the map image;
A map providing system comprising: output means for outputting the rearranged map image.

Images