JP2001014425A - Method and device for processing picked-up image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for processing picked-up image with which all handwriting characters or the like can be easily and exactly inputted without limiting the colors of characters. SOLUTION: This device is provided with a scanner 2 for picking up the image of a handwriting character with a non-fluorescent material on a paper 1, which is coated with an ultraviolet fluorescent material, while using visible light, a scanner 3 for picking up the image of the paper 1 while using ultraviolet light, an extracting means 4 for extracting the character area of the image picked up with ultraviolet light by the scanner 3, a discriminating means 5 for extracting the character area part, which is extracted by the extracting means 4, as a character part out of the image picked up by the scanner 2 while using visible light and a synthesizing means 6 for synthesizing the image of the character part extracted by the discriminating means 5 with the other image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for processing a picked-up image by picking up a handwritten character or the like and extracting a character portion from the picked-up image.

[0002]

2. Description of the Related Art Conventionally, when creating postcards or the like in digital image processing, when inputting handwritten (handwritten) characters, write with a mouse or the like, or handwritten paper brought in from the outside with a flatbed scanner. The image is captured and captured, and characters in the image are cut out and used by using image editing processing.

[0003]

However, characters written with a mouse or the like do not always have the same quality in terms of definition as characters written with a pen. Further, it is troublesome to cut out and edit an image captured by a flatbed scanner, and this is also a problem in terms of work efficiency.

Further, in the latter case, if a chroma key, which is a well-known image synthesizing method, is applied to a character obtained by cutting out a captured image, there is a limitation on colors that can be used for the character. That is, the chroma key extracts only a specific hue component from the hue components of an image to create a key signal,
Although a plurality of images are synthesized by the key signal, a color character having the same hue component as the hue component used in the key signal cannot be used at this time. For example, a character drawn in the same color as or close to the ground color of paper
It is difficult to determine the character portion.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art, and an image processing method and an image processing method capable of easily and accurately inputting any handwritten characters without limiting the color of characters. It is intended to provide a device.

[0006]

To achieve the above object, a first aspect of the present invention is a method for processing a picked-up image by picking up a handwritten character and extracting a character portion from the picked-up image. A paper coated with a fluorescent material that emits fluorescent light and a paper on which characters are handwritten with a non-fluorescent material was imaged using visible light and the above-mentioned radiated light, and was imaged using the above-mentioned radiated light. A character image region is extracted from the image, and the extracted character region portion is extracted as a character portion from the image captured using the visible light as a character image processing method.

According to this method, a sheet of paper coated with a fluorescent material that emits fluorescent light by a predetermined radiated light, on which a character is handwritten with a non-fluorescent material, is formed using visible light and the radiated light. Character regions are extracted from the images captured using the above-mentioned radiation light. The outside of the extracted character region is a portion where the fluorescent material has reacted to the emitted light, and by using this as a chroma key signal, the extracted portion is extracted from the image captured using the visible light. The character area is extracted as a character part. Therefore, only the written characters can be captured as image data without regard to the color of the written characters. As a result, input of all handwritten characters and the like can be easily and accurately performed without limiting the color of characters.

Further, if the extraction of the character region from the image picked up by using the above-mentioned radiation is performed based on the brightness of the image (claim 2), no handwritten characters are written on the paper. The part emits fluorescence from the fluorescent material because it is not masked by the non-fluorescent material, but the part where handwritten characters are written does not emit fluorescence because it is masked by the non-fluorescent material, and the brightness of both Are clearly different. Therefore, a character region can be accurately extracted from such a luminance distribution.

Further, if the extracted character portion image is combined with another image (claim 3), the captured character portion image can be enlarged or reduced in the same manner as ordinary image data. It can be freely reduced, rotated, etc., and pasted at a desired location in the image.

Further, if the image picked up by using the visible light identifies the color of a non-fluorescent material (claim 4),
Characters of a desired color can be captured as image data.

Further, if the handwritten characters include a handwritten illustration (claim 5), a desired handwritten document can be created.

Further, the emitted light may be light having a wavelength outside visible light (claim 6), and particularly, the light outside visible light may be ultraviolet light (claim 6). 7),
Its handling is simple.

According to a further aspect of the present invention, there is provided an image pickup apparatus for picking up a handwritten character and extracting a character portion from the picked-up image, wherein the sheet is coated with a fluorescent material which emits fluorescent light by a predetermined radiation. A first image pickup means for picking up an image of a sheet on which characters are handwritten with a non-fluorescent material using visible light, and a second image pickup means for picking up an image of the sheet using the radiation light.
And an extracting unit for extracting a character region from an image captured by the second imaging unit, and a character region extracted by the extracting unit from an image captured by the first imaging unit And a determination means for extracting the minute as a character portion. The image capturing apparatus may further include a combining unit that combines the image of the character portion extracted by the determining unit with another image, or the first imaging unit may include a color that can identify the color of the non-fluorescent material. It may be a scanner (claim 10). With these device configurations, the above-described method of the present invention can be realized.

[0014]

FIG. 1 is a schematic configuration diagram showing an embodiment of a captured image processing apparatus according to the present embodiment.

As shown in FIG. 1, an image processing apparatus according to the present embodiment (hereinafter referred to as “the present apparatus”) captures a handwritten character and extracts a character portion from the captured image. A scanner (first image pickup unit) 2 for picking up an image by receiving reflected light when irradiating visible light with a character handwritten with a non-fluorescent material on a sheet 1 coated with an ultraviolet fluorescent material, and a sheet A scanner (second imaging means) 3 that performs imaging by receiving fluorescent light emitted from an ultraviolet fluorescent material when the ultraviolet light is irradiated on the scanner 1, and an image captured by the scanner 3 based on ultraviolet light. Extraction means 4 for extracting a character area, and a portion of the image picked up by the scanner 2 based on the visible light, which is within the area of the handwritten character extracted by the extraction means 4 (hereinafter referred to as a character area). As And judging means 5 issuing Ri, the image of the retrieved portions by determining means 5 as a character and a combining unit 6 for combining with another image.

Here, a fluorescent whitening agent such as a stilbene dye can be used as the ultraviolet fluorescent material, and a non-fluorescent material such as a commercially available marker having a property of hardly transmitting ultraviolet light can be used. Can be used.

In the figure, 10 is a scanner 2 and a scanner 3
This is a flatbed scanner that integrally includes The scanner 2 has a line sensor composed of photoelectric conversion elements such as CCDs arranged in an array.
This is a color scanner in which the colors of non-fluorescent materials can be distinguished by providing three primary color filters of (green) and B (blue). The scanner 3 has a line sensor similar to that of the scanner 2 and is provided with a band-pass filter that allows passage of fluorescent light emitted from a phosphor.

Reference numeral 11 denotes a fluorescent lamp as an irradiation light source that emits both visible light and ultraviolet light.
Reference numeral 0 denotes an image taken by manually or mechanically scanning the paper 1 set on the flat flat bed 10a using visible light and ultraviolet light emitted from the fluorescent lamp 11. The scanners 2 and 3 may be area sensors in which CCDs are arranged in a matrix, instead of the line sensors. In that case, scanning becomes unnecessary.

The picked-up image data is taken into the data processing device 12, digitized by the preprocessing means 13, temporarily stored in the RAM 14, read out at a predetermined timing, and stored in the ROM 15 in advance. A series of image processing is performed according to a prepared predetermined program. That is, the above-mentioned extraction means 4, determination means 5, synthesis means 6, and preprocessing means 13
5 comprises an executable program installed in the data processing device 12, and its execution is controlled by the CPU 12a under the control of the input device 16, such as a keyboard, and C.
It can be performed interactively using an output device 17 such as an RT. For this reason, the RAM 14
The ROM 15 has a sufficient capacity to store both the R, G, and B image data and the fluorescent light image data, and the ROM 15 has a sufficient capacity to store the program. Note that the execution of the program may not be performed interactively, but may be performed automatically.

Incidentally, the paper 1 is coated with the ultraviolet fluorescent material as described above. This ultraviolet fluorescent material reacts to ultraviolet light but does not react to visible light. The fluorescent lamp 11, which is the light source of the flatbed scanner 10,
It is possible to generate a sufficient amount of ultraviolet light for the ultraviolet fluorescent material to react. In the present embodiment, a portion where the ultraviolet light fluorescent material has reacted is used as a key signal of a chroma key.

That is, the portion where the handwritten characters are not written on the sheet 1 is not masked with the non-fluorescent material, so that when irradiated with the ultraviolet light from the fluorescent lamp 11, the ultraviolet light fluorescent material applied to the portion is Although the part emits fluorescent light, the part where the handwritten characters are written on the paper 1 is masked with the non-fluorescent material, so that the ultraviolet light irradiation from the fluorescent lamp 11 is blocked, and the ultraviolet light fluorescent material applied to the part is Does not fluoresce and the brightness of both parts is clearly different. Therefore, the extracting unit 4 of the data processing device 12 accurately extracts a character region based on such a luminance distribution, and the determining unit 5 distinguishes an image (character) in the visible light region using the extracted character region. be able to. When synthesizing the image data of the character portion with other image data, the synthesizing means 6 enlarges or reduces the image data of the character portion in the same manner as normal image data.
The image can be pasted at a desired position in the image by freely performing rotation or the like.

The operation of the apparatus will be described below with reference to FIGS. That is, this operation implements the image processing method of the present invention.

FIG. 2 is a flowchart showing the operation of the apparatus according to this embodiment, and FIG. 3 is an explanatory diagram showing an example of image data according to this embodiment.

A paper 1 such as a post card coated with an ultraviolet fluorescent material is prepared in advance. The user hand-writes characters (including illustrations) on the prepared paper 1 using a non-fluorescent material such as a marker of an arbitrary color. Then, the paper 1 on which the handwritten characters are written is transferred to the flatbed scanner 10.
(See FIG. 1).
It is assumed that the sheet 1 that has just been set is a sheet 1 on which handwritten characters are written, for example, as shown in FIG. Under the above assumptions, the operation is performed as follows.

When a power switch (not shown) is turned on by a user, initial settings such as setting default values in the RAM 14 are first performed as shown in FIG. 2 (step S1).

Next, the handwritten characters on the paper 1 are scanned by the scanner 2 of the flatbed scanner 10 using visible light, and are taken into the data processing device 12 as R, G, and B image data, respectively (step S1). S2). Each of the captured R, G, and B image data is digitized by the pre-processing unit 13, for example, as shown in FIG.
Therefore, a digital image as shown in FIG. In FIG.
Although only one image is shown, three images of each color of R, G, and B are actually created. The flatbed scanner 10 is returned to the original position and prepares for the next scan. Note that the flatbed scanner 10 is provided with scanners 2 and 3
Are integrated, it is possible to execute step S2 and step S3 at the same time, but in this case, the return operation is not necessary here.

Next, as in step S2, the handwritten characters on the paper 1 are scanned by the scanner 3 of the flatbed scanner 10 using ultraviolet light, and are captured in the data processing device 12 as ultraviolet light image data (FIG. 2). Step S3). The captured ultraviolet light image data is digitized by the preprocessing means 13, and the character area scanned by the ultraviolet light in the digital image is extracted by the extraction means 4 (step S4). The character area extracted in this manner is, for example, as shown in FIG. In the figure, a white portion is a character area.

Next, the judging means 5 extracts only the portion in the character area extracted in the step S3 from the digitized image captured in the step S2 (step S5). That is, the outside of the character region is a portion where the ultraviolet fluorescent material has reacted to ultraviolet light, and by using this as a chroma key signal,
Images (characters) in the visible light region can be distinguished. As a result, a portion of the digital image in the extracted character area, which is captured with visible light, is extracted as a character. This is the image data of the extracted character portion.
The color of the original handwritten character is reproduced by superimposing the R, G, and B image data.

Then, the color-reproduced image data is
The image is combined with other image data by the combining means 6 (step S
6). The result is, for example, image data as shown in FIG.

As described above, in this embodiment, the characters handwritten with the non-fluorescent material on the paper 1 coated with the ultraviolet fluorescent material are scanned by the scanners 2 and 3 of the flatbed scanner 10.
Are respectively imaged using visible light and ultraviolet light, and a character region is extracted from the image captured using this ultraviolet light by the extraction means 4, and based on the extracted character region, The character portion in the captured image is extracted by the determining means 5, and the image data of the extracted character portion is combined with other image data by the combining means 6, so that the color of the written character is changed. Only the written characters can be captured as image data without any concern. Therefore, input of all handwritten characters and the like can be easily and accurately performed without limiting the color of characters.

In the above embodiment, the handwritten characters on the paper 1 coated with the ultraviolet fluorescent material are scanned by the scanner 3 of the flatbed scanner 10 with the ultraviolet light. However, the present invention is not limited to this. The phosphor may be caused to emit light in a predetermined wavelength band (predetermined emission characteristics) with radiation other than ultraviolet light, for example, X-rays, gamma rays, electron beams, or the like. Further, a phosphor containing radiation may be used as long as safety is ensured. In this case, the light source of the flatbed scanner 10 only needs to be a visible light source, and the configuration of the apparatus is simplified. further,
Instead of the flatbed scanner 10, a drum scanner or the like may be used. However, the flatbed scanner 10 is convenient if the ease of setting the paper 1 is considered.

Further, in the above-described embodiment, the handwritten characters include the handwritten illustration. However, any other types of symbols may be included, and the handwritten characters may include non-handwritten characters. .

Further, in the above embodiment, the non-fluorescent material can employ any color. For example, even if the non-fluorescent material is written in the same color as the ground color, such as a white handwritten character written on white paper 1, Character regions can be extracted from the ultraviolet light image based on the luminance distribution as described above, and characters can be extracted from the visible light image based on the character regions.

Further, in the above embodiment, the scanners 2 and 3 of the flatbed scanner 10 are provided separately. However, for example, the three primary color filters of R, G and B and the fluorescent light filter which are provided in the line sensor are used. If the filter switching means is provided, only one scanner needs to be provided, thereby simplifying the apparatus.

[0035]

As described above, according to the first aspect of the present invention,
Is a photographed image processing method for photographing a handwritten character and extracting a character portion from the photographed image, wherein the paper is coated with a fluorescent material that emits fluorescent light by predetermined radiation light, and the character is formed on the paper by a non-fluorescent material. The handwritten paper is imaged using visible light and the radiation light, respectively, and a character region is extracted from the image captured using the radiation light, and the character region is extracted from the image captured using the visible light. It is configured as a picked-up image processing method characterized by extracting the extracted character area as a character part.

According to this method, a sheet on which a fluorescent material which emits fluorescence by a predetermined radiated light is applied, and a sheet on which a character is handwritten with a non-fluorescent material is formed using visible light and the radiated light. Character regions are extracted from the images captured using the above-mentioned radiation light. The outside of the extracted character region is a portion where the fluorescent material has reacted to the emitted light, and by using this as a chroma key signal, the extracted portion is extracted from the image captured using the visible light. The character area is extracted as a character part. Therefore, only the written characters can be captured as image data without regard to the color of the written characters. As a result, input of all handwritten characters and the like can be easily and accurately performed without limiting the color of characters.

Furthermore, if the extraction of the character area from the image picked up using the above-mentioned radiation light is performed based on the luminance of the image (claim 2), no handwritten characters are written on the paper. The part emits fluorescence from the fluorescent material because it is not masked by the non-fluorescent material, but the part where handwritten characters are written does not emit fluorescence because it is masked by the non-fluorescent material, and the brightness of both Are clearly different. Therefore, a character region can be accurately extracted from such a luminance distribution.

Further, if the extracted character portion image is combined with another image (claim 3), the captured character portion image can be enlarged or reduced in the same manner as normal image data. It can be freely reduced, rotated, etc., and pasted at a desired location in the image.

Further, if the image picked up using the visible light discriminates the color of the non-fluorescent material (claim 4),
Characters of a desired color can be captured as image data.

Further, if the handwritten characters include a handwritten illustration (claim 5), a desired handwritten document can be created.

Further, the emitted light may be light having a wavelength outside visible light (claim 6), and particularly, the light outside visible light may be ultraviolet light (claim 6). 7),
Its handling is simple.

According to an eighth aspect of the present invention, there is provided a picked-up image processing apparatus for picking up a handwritten character and extracting a character portion from the picked-up image, wherein the paper is coated with a fluorescent material which emits fluorescent light by predetermined radiation. A first image pickup means for picking up an image of a sheet on which characters are handwritten with a non-fluorescent material using visible light, and a second image pickup means for picking up an image of the sheet using the radiation light.
And an extracting unit for extracting a character region from an image captured by the second imaging unit, and a character region extracted by the extracting unit from an image captured by the first imaging unit And a determination means for extracting the minute as a character portion. The image capturing apparatus may further include a combining unit that combines the image of the character portion extracted by the determining unit with another image, or the first imaging unit may include a color that can identify the color of the non-fluorescent material. It may be a scanner (claim 10). With these device configurations, the above-described method of the present invention can be realized.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram illustrating an embodiment of a captured image processing device according to the present invention.

FIG. 2 is a flowchart showing an operation of the device in the embodiment.

FIG. 3 is an explanatory diagram showing an example of image data in the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Paper 2 Scanner (1st imaging means) 3 Scanner (2nd imaging means) 4 Extraction means 5 Judgment means 6 Synthesizing means 10 Flatbed scanner 10a Flatbed 11 Fluorescent lamp 12 Data processing device 12a CPU 13 Preprocessing means 14 RAM 15 ROM 16 input device 17 output device

Claims (10)

[Claims] An image processing method for picking up a handwritten character and extracting a character portion from the picked-up image, wherein the sheet is coated with a fluorescent material that emits fluorescent light by predetermined radiation light, The paper on which the characters were handwritten with the fluorescent material was imaged using visible light and the emitted light, respectively, and a character region was extracted from the image captured using the emitted light, and the image was imaged using the visible light. A picked-up image processing method, wherein the extracted character area portion is extracted as a character portion from an image. 2. The captured image processing method according to claim 1, wherein the extraction of the character area from the image captured using the radiation light is performed based on the brightness of the image. 3. The captured image processing method according to claim 1, wherein the image of the extracted character portion is combined with another image. 4. The captured image processing method according to claim 1, wherein the imaging performed using the visible light is for identifying a color of a non-fluorescent material. 5. The captured image processing method according to claim 1, wherein the handwritten character includes a handwritten illustration. 6. The captured image processing method according to claim 1, wherein the emitted light is light having a wavelength outside visible light. 7. The captured image processing method according to claim 6, wherein the light outside the visible light is ultraviolet light. 8. A picked-up image processing apparatus for picking up a handwritten character and extracting a character portion from the picked-up image, wherein the picked-up paper is coated with a fluorescent material which emits fluorescent light by predetermined radiation light, A first image pickup unit for picking up an image of a sheet on which characters are handwritten with a material using visible light; a second image pickup unit for picking up an image of the sheet using the radiation light;
Extracting means for extracting a character area from an image captured by the image capturing means, and determining means for extracting, as a character part, the character area extracted by the extracting means from the image captured by the first image capturing means And a captured image processing apparatus. 9. The captured image processing apparatus according to claim 8, further comprising a synthesizing means for synthesizing the image of the character portion extracted by said determining means with another image. 10. The captured image processing apparatus according to claim 8, wherein said first imaging means is a color scanner capable of identifying the color of a non-fluorescent material.