JP2011175309A - Image processing apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a nonrealistic image including characters such as a signature so as not to be displayed in the background. <P>SOLUTION: First filtering is applied to recorded photographic image data, using a specific art filter (S11). After parameters of the art filter are changed (S12), second filtering is applied to signature or other image data, using the changed parameter of the art filter (S13). The filtered signature or other image data is combined with the filtered photographic image data to generate data of a nonrealistic image including characters such as a signature (S15, S16). In the nonrealistic image with the generated characters, the character part and the background image part can be visually distinguished. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing technique for generating a non-realistic image including characters from an original image.

  Conventionally, for example, Patent Document 1 described below describes a technique for converting a photographed image photographed by a digital camera or the like and recorded as image data into a picture-like image or the like. In general, a photorealistic image such as a photographed image (photograph) is called a photorealistic image, and a non-photographic image such as a picture-like image or a computer graphic image is called a non-photorealistic image. Yes.

  On the other hand, when a photographed image is converted into a non-realistic image such as a picture-like image and is viewed as a new work, adding characters such as a signature to the photographed image in advance makes it more suitable for viewing. A non-realistic image can be obtained. A technique for adding characters to an image is known in Patent Document 2 below.

JP 2006-31688 A Japanese Patent Laid-Open No. 11-187292

  However, when a photographed image to which a character such as a signature is added is converted into a non-realistic image, the sign becomes difficult to read depending on the content of the effect given to the converted non-realistic image. That is, there is a problem that the sign is buried in the background image portion.

  The present invention has been made in view of such conventional problems, and an object thereof is to provide a non-realistic image in which characters are not buried in a background portion as a non-realistic image including characters such as a signature.

  In the image processing apparatus according to the first aspect of the present invention, a first conversion means for converting an original image into a non-realistic image having a first visual effect, and the first image conversion Second conversion means for converting a character to be combined with the non-realistic image converted by the means into a character with an effect having a second visual effect different in content from the first visual effect; A non-realistic image converted by the first image converting means, and a combining means for synthesizing the character with effect converted by the second converting means and generating a non-realistic image including the character. It is characterized by.

  In the program according to the first aspect of the present invention, the computer converts the original image into a non-realistic image having a first visual effect, and the first conversion means. A second conversion for converting a character to be combined with the non-photorealistic image converted by the image conversion means into a character with an effect having a second visual effect different from the first visual effect. And a non-realistic image converted by the first image conversion means, and a non-realistic image converted by the second conversion means, and a non-realistic image including the character is generated as a synthesizing means It is made to function.

  According to the present invention, a non-realistic image in which characters are not buried in the background portion can be provided as a non-realistic image including characters such as a signature.

It is a schematic block diagram which shows the hardware constitutions of the digital photo frame concerning this invention. It is a functional block diagram which shows the principal part of the function of a digital photo frame. It is a flowchart which shows the process sequence of CPU in an effect mode. (A) is a figure which shows an example of the non-realistic image containing a character, (b) is a schematic diagram corresponded to the principal part enlarged view of (a). It is a figure corresponding to Drawing 4 (a) showing other examples of a non-realistic picture containing a character.

  Hereinafter, an embodiment of the present invention will be described. The present embodiment mainly relates to a digital photo frame (hereinafter referred to as DPF) having a function of displaying an image photographed by a digital camera or the like and recorded as image data, that is, a photograph.

  FIG. 1 is a schematic block diagram showing a hardware configuration of a DPF 1 to which the present invention is applied. The DPF 1 has a configuration in which the entire system is controlled by a CPU (Central Processing Unit) 101. Connected to the CPU 101 are a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a media controller 104, an operation key group 201, and a display 301.

  The ROM 102 is a memory in which a plurality of types of programs and data for causing the CPU 101 to control the entire system are stored. The programs stored in the ROM 102 include a display control program 102a that causes the CPU 101 to function as first conversion means, second conversion means, composition means, and character area detection means in the present embodiment.

  Further, the data stored in the ROM 102 includes character pattern data indicating characteristics such as various character shapes used when the CPU 101 performs character recognition processing on arbitrary image data. . The character recognition process is a process for recognizing a character area where a character exists from an image represented by image data.

  The RAM 103 is specifically an SDRAM (Synchronous dynamic random-access memory) or the like.

  The media controller 104 is an input / output interface that controls input / output of data between the CPU 101 and a recording medium 40 that is detachably mounted in a memory card slot provided in a main body (not shown) of the DPF 1.

  A plurality of image data representing images (photographs) acquired by photographing with a digital camera or the like are recorded on the recording medium 40. The image data is compressed by, for example, JPEG (Joint Photographic Expert Group) method, and is stored in the recording medium 40 as a still image file together with additional information such as the shooting date and time. In the following description, an image acquired by photographing is referred to as a photographed image, and image data representing the photographed image is referred to as photographed image data.

  The operation key group 201 includes a power switch and a plurality of operation switches. The CPU 101 detects various instructions from the user by predetermined key operations by constantly scanning the operation state of the operation keys and the like in the operation key group 201.

  The display 301 drives the color liquid crystal display panel with backlight and the color liquid crystal display panel according to display data such as image data supplied from the CPU 101, and displays the image represented by the image data on the color liquid crystal display panel. And a driving circuit to be displayed.

  FIG. 2 is a functional block diagram showing the main part of the functions of the DPF 1. The DPF 1 includes a control unit 10, an operation unit 20, and a display unit 30. The control unit 10 includes a character area detection unit 11, a first conversion unit 12, a second conversion unit 13, an image composition unit 14, a display control unit 15, and a storage unit 16.

  The operation unit 20 is a functional part that detects various instructions from the user to the DPF 1, and is realized by the operation key group 201 shown in FIG.

  The display unit 30 displays a photographed image represented by the photographed image data recorded on the recording medium 40, an image represented by other image data, and various types required when the user operates the DPF 1. Display the operation screen. The display unit 30 is realized by the display 301 illustrated in FIG.

  The character area detection unit 11 performs a character area detection process on arbitrary captured image data recorded on the recording medium 40. The character area detection process is a process of detecting a character area in which a character is expressed in a captured image represented by image data, and acquiring position information indicating the position of the detected character area in the image. The character area detection unit 11 is realized by the CPU 101 shown in FIG.

  The first conversion unit 12 performs a specific filter process on arbitrary captured image data recorded on the recording medium 40, thereby obtaining a captured image (original image) represented by the captured image data to be processed. Convert to a non-realistic image with a specific visual effect for viewing purposes. The specific visual effect for the purpose of appreciation given to the photographed image by the first conversion unit 12 is the first visual effect in the present embodiment. The specific visual effect is, for example, a style of a specific painting such as a watercolor painting, an oil painting, a pencil drawing, or a pastel drawing.

  There are a plurality of types of filter processing that the first conversion unit 12 can apply to any photographed image data, and the first conversion unit 12 provides a plurality of types of visual effects to the converted non-realistic image. be able to. The first conversion unit 12 is realized by the CPU 101 illustrated in FIG.

  The second conversion unit 13 is data representing a signature image data recorded on the recording medium 40 or a character region image (hereinafter referred to as a character region image) detected by the character region detection unit 11 from an arbitrary image. In addition, by applying a specific filtering process, the sign image and the character area image are converted into a non-realistic image having a specific visual effect.

  However, the filtering process performed by the second conversion unit 13 is different from the filtering process performed by the first conversion unit 12, and the second conversion unit 13 performs a specific visual process applied to the signature image and the character area image. The effect is different from the visual effect that the first converter 12 gives to the original image. The specific visual effect that the second conversion unit 12 gives to the sign image and the character area image is the second visual effect in the present embodiment.

  In addition, the sign image and the character area image after the filter processing by the second conversion unit 13 are characters with effects in the present embodiment. The second conversion unit 13 is realized by the CPU 101 shown in FIG.

  In the following description, the filter process performed by the first converter 12 is referred to as a first filter process, and the filter process performed by the second converter 13 is referred to as a second filter process. The conversion of the original image into the non-real image by the first conversion unit 12 and the conversion of the signature image and the character area image into the non-real image by the second conversion unit 13 are referred to as art conversion.

  The image synthesizing unit 14 synthesizes image data representing a sign image and a character area image after the art conversion by the second conversion unit 13 with the image data representing the photographed image after the art conversion by the first conversion unit 12. Thus, image data representing a non-realistic image including characters is generated. The image composition unit 14 is implemented by the CPU 101 illustrated in FIG.

  The display control unit 15 displays an image represented by arbitrary image data recorded on the recording medium 40 or a non-realistic image including characters represented by the image data generated by the image composition unit 14 on the display unit 30. Display. The display control unit 15 is realized by the CPU 101 shown in FIG.

  The storage unit 16 temporarily stores character region image data, which is data representing a character region image detected from an arbitrary image by the character region detection unit 11, and various image data to be displayed on the display unit 30 by the display control unit 15. To remember. The storage unit 16 is realized by the RAM 103 illustrated in FIG.

  Next, the operation of the DPF 1 having the above configuration will be described. The DPF 1 has a normal mode and an effect mode as display modes. The display mode can be appropriately switched by the user operating a predetermined operation key of the operation key group 201.

  The normal mode is a display mode in which a captured image represented by captured image data recorded on the recording medium 40 is displayed on the display 301 as it is. The effect mode is a display mode in which a captured image is converted into a non-realistic image including characters and displayed on the display 301.

  Hereinafter, a specific operation of the DPF 1 in the effect mode will be described. In the effect mode, the user selects a captured image to be displayed by a predetermined key operation.

  FIG. 3 is a flowchart showing processing executed by the control unit 10 (CPU 101) in accordance with the display control program 102a stored in the ROM 102 in the effect mode, after the captured image to be displayed is selected. is there.

  In the effect mode, after the captured image to be displayed is selected, the control unit 10 reads captured image data representing the captured image selected by the user from the recording medium 40 and temporarily stores it in the storage unit 16 (RAM 103) (step 103). S1).

  Next, the character area detection unit 11 (CPU 101) performs character area detection processing on the captured image data stored in the storage unit 16 as a processing target (step S2). In the process of step S2, the character area detection unit 11 detects a character area in which a character is expressed in the captured image represented by the captured image data, and obtains positional information indicating the position of the detected character area in the captured image. get.

  A method of detecting a character region by the character region detection unit 11 is a known method for extracting a character pattern from a character image by using, for example, structure information of a character category of a feature standard dictionary described in Japanese Patent Application Laid-Open No. 2009-259190. It is a method to which technology is applied. That is, it is a method of detecting an area having a feature similar to the character category of the feature standard dictionary as a character area.

  When the detection of the character area is successful (step S3: YES), the control unit 10 cuts out the image data of the character area detected by the character area detection unit 11 from the photographed image (step S4), Along with the position information of the character area acquired by the area detection unit 11, it is temporarily stored in the storage unit 16 separately from the captured image data (step S5). The image data cut out by the control unit 10 in step S4 is image data representing only characters that do not include the background expressed in the captured image.

  On the other hand, when the character area cannot be detected in the process of step S2 (step S3: NO), the control unit 10 displays a sign image selection screen on the display unit 30 (display 301) (step S6). In step S <b> 6, the control unit 10 reads one or a plurality of signature image data recorded in advance from the recording medium 40, and displays the signature image represented by the signature image data on the display unit 30. Then, the control unit 10 causes the user to select a desired sign image by a predetermined key operation.

  Thereafter, when a sign image is selected by the user (step S7: YES), the control unit 10 temporarily stores sign image data representing the selected sign image in the storage unit 16 (step S8).

  That is, by the processing in steps S2 to S8, the storage unit 16 temporarily stores either one of the character area image data cut out from the captured image data or the sign image data.

  Subsequently, the control unit 10 causes the display unit 30 to display an art filter selection screen (step S9). In step S <b> 6, the control unit 10 displays a selection screen that displays names (may be symbols) of a plurality of types of art filters respectively named according to a plurality of types of filter processing prepared in advance in the DPF 1. indicate. Then, the control unit 10 causes the user to select a desired art filter by a predetermined key operation.

  The art filter is a term used to indicate a plurality of types of filter processing to the user. For example, as used in Adobe Photoshop (trademark) by Adobe System, which is a commercially available software package, “ It is the same as the term indicating the type of “artistic filter” (“color pencil”, “watercolor”, etc.).

  Thereafter, when the art filter is selected by the user (step S10: YES), the first conversion unit 12 (CPU 101) stores the first filter processing corresponding to the selected art filter in the storage unit 16. Art conversion to be performed on the captured image data to be processed is performed (step S11). In step S <b> 11, the first conversion unit 12 stores, in the storage unit 16, captured image data after art conversion, that is, image data representing a non-realistic image that gives a specific visual effect to the captured image.

  The first filtering process performed by the first conversion unit 12 on the captured image data to be processed is a known technique such as the following technique described in Japanese Patent Laid-Open No. 2006-31688 described as the background art of the present invention. This is a process of giving a specific visual effect to an image represented by photographed image data by a method to which technology is applied. In the technique described in Japanese Patent Laid-Open No. 2006-31688, with respect to pixels included in image data, each pixel is grouped by using color as one of the determination criteria, and the image data is divided into a plurality of basic regions. Let the region be a unit region of expression. Further, a plurality of basic areas integrated based on the characteristics of each basic area are used as unit areas for expression. This is a technique for determining the color of a pixel included in each unit area based on the characteristics of each unit area.

  The first filter processing can also be realized by an image processing technique used for image processing in Adobe Photoshop (trademark) by Adobe System, for example.

  Next, the second conversion unit 13 (CPU 101) sets the content of the second filter process (step S12). The process of step S12 is a process of changing the set value used by the first conversion unit 12 in the first filter process to the set value used by the second conversion unit 13 in the second filter process. The setting value changed by the second conversion unit 13 is a setting value that affects the visual effect given to the non-realistic image after conversion by the first filter processing.

  The setting value changed by the second conversion unit 13 is a predetermined setting value determined in advance for each type of filter processing prepared in advance in the DPF 1, and the first conversion unit 12 sets the first conversion value to the first value. This is a specified set value corresponding to the filter process executed as the filter process. The prescribed set value is, for example, a set value that affects the thickness of the stroke, the density of the stroke, and the direction of the stroke in the filtered image. Here, the brush touch is an effect of a paint brush in a painting or the like, that is, an action of a paint brush that forms expressions such as color tone, light and darkness, and rhythm feeling generated by brush judgment, and is synonymous with touch and touch. When the setting value is a setting value that affects the thickness of the brush stroke, the density of the brush stroke, and the direction of the brush stroke in the image after the filter processing, the following value is set as the setting value.

  When the setting value is a setting value that affects the thickness of the brush stroke in the image after filtering, the changed setting value is, for example, the thickness of the brush stroke obtained by the setting value before the change. The thickness is 3 times or more, or 1/3 or less. That is, the setting value after the change is a value that makes the thickness of the brush stroke thick enough to visually recognize the difference from the thickness of the brush stroke obtained with the setting value before the change.

  In addition, when the set value is a set value that affects the density of the brush in the image after the filter processing, the set value after the change is the touch density, for example, 2 of the touch density obtained by the set value before the change. It is a value that makes the density more than double or less than half. In other words, the set value after the change is a value at which the brush density is such that the difference from the touch density obtained with the set value before the change can be visually recognized. Here, the density of the brush stroke in the image after filtering is, for example, the line density when multiple lines of the same or different lengths scattered in parallel that are recognized as hatching appear in the image after filtering. It is the number.

  Further, when the setting value is a setting value that influences the direction of the brush in the image after the filter processing, the setting value after the change indicates the direction of the brush, for example, the direction of the brush obtained by the setting value before the change, It is a value which makes the inclination angle of a brush different from 30 degrees to 90 degrees, for example. That is, the setting value after the change is a value that sets the direction of the brush so that the difference from the direction of the brush obtained by the setting value before the change can be visually recognized.

  Further, in the process of step S12, the second conversion unit 13 changes the setting value according to the relative size with respect to the size of the photographic image of the sign image or the character area image when changing the specified setting value. Change. For example, the second conversion unit 13 determines whether the size level of the signature image or the character area image is “large” or “small” according to a predetermined standard. Then, when the magnitude level is “small”, the second converter 13 changes the set value with a greater degree of change than when the magnitude level is “large”.

  Subsequently, the second conversion unit 13 performs art conversion for applying the second filter processing to the character area image data or the signature image data stored in the storage unit 16 (step S13). The second filter process is a filter process in which the filter process based on the same algorithm as the first filter process by the first converter 12 is performed using the setting value changed in the process of step S12. In step S13, the second conversion unit 13 stores character area image data after art conversion or sign image data, that is, image data representing a non-realistic image having a specific visual effect different from the photographed image. 16 to store. In the following description, the non-realistic image represented by the image data stored in the storage unit 16 by the second conversion unit 13 in the process of step S13 is referred to as a character with effect for convenience.

  Next, the image composition unit 14 (CPU 101) first confirms whether or not the second filter processing target in the second conversion unit 13 is a sign image (step S14). When the second filter processing target is a sign image (step S14: YES), the image composition unit 14 determines image data representing the effect-added character as a photographed image after the art conversion. Overwriting is performed on the specified position (step S15). The image synthesis unit 14 stores the synthesized image data in the storage unit 16.

  In addition, when the second filter processing target is a character area image (step S14: NO), the image composition unit 14 converts the image data representing the effect-added character into the original image in the photographed image after the art conversion. The position is overwritten and synthesized (step S16). The image synthesis unit 14 stores the synthesized image data in the storage unit 16.

  Thereafter, the display control unit 15 (CPU 101) sends the combined image data stored in the storage unit 16 in the process of step S15 or step S16 to the display unit 30 (display 301), and the display unit 30 includes characters. A non-realistic image is displayed (step S17).

  FIG. 4A shows a non-realistic image A including characters displayed on the display 301 in the process of step S17, that is, an image in which the effected character C is combined with the photographed image B after art conversion as the background. It is a figure which shows an example. FIG. 4B is an enlarged view showing the periphery of the character C with effect. FIG. 4B shows a visual effect given to the photographed image B after the art conversion by the first filter processing and a visual effect given to the character C with the effect after the art conversion by the second filter processing. It is the figure which showed the case where a difference is the direction of a brush stroke for convenience.

  FIG. 5 shows an example in which the difference between the visual effect given to the photographed image B after the art conversion and the visual effect given to the character C with the effect after the art conversion is the thickness of the brush stroke. It is a figure corresponding to FIG.4 (b) shown in figure.

  Here, the post-art-converted character C combined with the photographed image B after the art conversion is a visual image that is different from the photographed image in the character area image or the sign image cut out from the photographed image before the art conversion. This is a non-realistic image that gives an effect.

  Therefore, the character C with an effect can ensure a different atmosphere from the photographed image B after the background art conversion, and the character C with the effect and the photographed image B after the background art conversion are visually differentiated. can do. Therefore, a non-realistic image in which characters are not buried in the background portion can be provided as a non-realistic image including characters.

  Further, the second filter process is a filter process using the same algorithm as the first filter process, and is a process for changing and using the set value used in the first filter process. The first filter process Changed the setting value used in the following. That is, the setting value used in the first filter processing is changed with a greater degree of change when the size level of the sign image or the character area image is “small” than when the size level is “large”. did. Therefore, when the characters with effects are small, the difference in the atmosphere from the photographed image after the art conversion is more prominent than when the characters with effects are large, so that the characters with the effect and the art conversion are more reliably obtained. It is possible to visually differentiate from the later photographed image.

(Deformation etc.)
In the present embodiment, the second filter processing applied to the character region image or the sign image cut out from the photographed image before the art conversion is the same as the first filter processing. The configuration that is processing performed using setting values different from the filter processing has been described. However, the DPF 1 may employ a configuration in which the second filter process is a filter process based on an algorithm that is completely different from the first filter process. For example, the first filter process may be a process for converting a photographed image into an oil painting-like image, and the second filter process may be a process for converting the photographed image into a pencil-like image.

  The second filter process may be a process of simply changing the color of the character area image or the sign image to a color obtained by inverting the negative and positive colors with respect to the background color at the synthesized portion of the photographed image after the art conversion.

  In this embodiment, when a sign image is art-converted into a character with an effect and combined with a photographed image after the art conversion, the character with an effect is combined at a predetermined specific position of the photographed image after the art conversion. Explained when to do.

  However, the composition position of the character with effect may be automatically set according to a predetermined setting rule. For example, (a) the corner of the image, (b) the lower part of the image, (c) that it is not the attention area, and (d) a position away from the in-focus position. More than one of the multiple conditions of being. Note that the attention area (c) is, for example, a human face area detected from a captured image by a known face detection technique.

  In the present embodiment, the case where the image displayed by the DPF 1 after being converted into a non-real image is an image (photograph) obtained by taking a photograph has been described. However, in the present invention, the original image to be converted into a non-real image is not limited to an image obtained by taking a photograph, and may be, for example, a photograph read by a scanner and recorded as image data.

  In the present embodiment, the case where the present invention is applied to the DPF has been described. However, the present invention can be applied to apparatuses other than the DPF as long as they have a predetermined image processing function. it can. Other devices include, for example, a digital camera, an arbitrary portable information terminal such as a mobile phone terminal incorporating a digital camera, a general-purpose personal computer, and a printer.

1 DPF
DESCRIPTION OF SYMBOLS 10 Control part 11 Character area detection part 12 1st conversion part 13 2nd conversion part 14 Image composition part 15 Display control part 16 Storage part 20 Operation part 30 Display part 40 Recording medium 101 CPU
102 ROM
102a Display control program 103 RAM
104 Media controller 201 Operation key group 301 Display A Non-realistic image including characters B Photographed image after art conversion C Character with effect

Claims (4)

First conversion means for converting an original image into a non-realistic image having a first visual effect;
A character to be combined with the non-photorealistic image converted by the first image conversion means is converted into a character with an effect having a second visual effect different in content from the first visual effect. Two conversion means;
And a non-realistic image converted by the first image converting means, and a synthesizing means for synthesizing the character with effect converted by the second converting means and generating a non-realistic image including the character An image processing apparatus. The second conversion means includes
Converting at least one of the stroke thickness, the stroke density, and the stroke direction of the character into a stroke different from the stroke thickness, the stroke density, and the stroke direction of the non-realistic image, The image processing apparatus according to claim 1. Character area detection means for detecting a character area in which characters are expressed from the original image,
The second conversion means includes
Converting the character represented in the character region detected by the character region detecting means into a character with an effect having a second visual effect;
The synthesis means includes
3. The image processing apparatus according to claim 1, wherein the character-with-effect converted by the second conversion unit is combined with the character region detected by the character region detection unit in the non-realistic image. Computer
First conversion means for converting an original image into a non-realistic image having a first visual effect;
A character to be combined with the non-photorealistic image converted by the first image conversion means is converted into a character with an effect having a second visual effect different in content from the first visual effect. Two conversion means;
A non-realistic image converted by the first image converting means is combined with a character with an effect converted by the second converting means to function as a combining means for generating a non-realistic image including the character. A program characterized by

Images