JPH09120456A - Method and device for image processing, and copying machine, printer, and scanner using same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the image processor which can detect a medium containing metallic fiber through image processing without using any metal sensor, etc., uses a simple recognition system, and exerts no influence on picture quality, etc. SOLUTION: Picked up multilevel image data are binarized by a binarization part 15 to generate a binary image, which is supplied to a thin line extraction part 17 to extract a thin line image. In concrete, a thin line removal part 20 removes thin lines present in the binary image and a subtraction processing part 21 finds difference between the binary image after removal and the binary image before processing supplied to the thin line extraction part to generate thin line image data. The thin line data which are thus obtained include data based upon metallic fiber, so a feature quantity decision part 19 judges whether or not the thin line data has a specific feature quantity.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and apparatus, a copying machine, a printer and a scanner using the same, and more specifically, a metal in a medium (paper or the like) of a read original to be detected. It is suitable for detecting the inclusion of fibers.

[0002]

2. Description of the Related Art Image processing apparatuses such as copying machines and printers are
It is convenient because you can easily output many copies (printed matter) with the same contents as the original, but on the other hand, by copying documents that should not be leaked to the outside, such as confidential documents, documents with the same contents can be output externally. There is a risk that they will be taken out. If the original (original) itself remains, it becomes difficult to monitor and detect whether or not a copy has been made and whether or not the copy has been taken out. Therefore, there is a demand to prevent copying of important documents such as confidential documents. Especially today when trade secrets are receiving attention.

Therefore, conventionally, an image relating to a specific pattern is registered in advance in a copying machine, it is judged whether or not the specific pattern is included in the input image, and it is recognized that the specific pattern exists. In this case, a method has been developed which employs a copy prohibition means such as painting the entire surface in black for output, or stopping the copy process itself. Therefore,
If a confidential document is created using a sheet on which the above-mentioned specific pattern is printed, it is impossible to copy the confidential document in principle.

On the other hand, as the quality of the current full-color copying machine is improved, the original image and the copied image are becoming indistinguishable from the naked eye. Therefore, copying is allowed, but for the purpose of clearly indicating that the copy is not the original, if a specific pattern is present in the input image, overlay the logo such as "copy" with light letters. Some of them are adapted to output a copy in a closed state.

In any case, it is necessary to detect that a specific pattern is included in the input image. As a feature extracting means for such detection, binarization or the like based on the peripheral contrast in the image is performed. A method of performing a process and extracting a characteristic part of an image is adopted.

[0006]

In the above-mentioned conventional apparatus, since the specific pattern existing in the original is exclusively recognized by the image processing, for example, the specific pattern portion existing in the original is tampered with. Once processed,
Although the original document should be detected, it may be erroneously recognized as not a detection target. As a countermeasure against this, conventionally, the development of an algorithm that makes it difficult to understand the existence of a specific pattern, or that can accurately determine the presence of some tampering has been actively carried out.

Further, in principle, a copying machine or the like is supposed to make a copy as the original without the above-mentioned copy prohibition processing and processing. Therefore, the degree of coincidence (threshold of criterion) when extracting / recognizing the particular pattern
Is inevitably higher. Then, for example, it is possible to determine that the pattern is not the specific pattern only by performing a modification such as adding a line around the specific pattern, and there is a drawback in that it is vulnerable to tampering.

On the other hand, apart from the direction of development of the present invention, metal fibers are preliminarily woven into a medium (usually paper) forming an original, and the presence or absence of the metal fiber is judged in parallel with the reading of original image data. However, when the metal fiber is detected, it is attempted to judge that the object is a prohibited object such as copying.

In that case, according to the general idea of the invention,
It is conceivable to use a dedicated metal sensor that uses magnetic force and electromagnetic waves, and determine whether or not the document is a detection target based on the output of the metal sensor.

However, such a method requires a metal sensor in addition to the image reading device, which not only complicates the device, but also causes the magnetic force and / or radio waves emitted at the time of metal detection to affect the human body, image quality, and surrounding radio waves. This is not preferable because it may affect broadcasting and communication devices.

The present invention has been made in view of the above background. An object of the present invention is to solve the above problems and to provide a medium containing a specific fiber (metal fiber) without using a metal sensor or the like. To provide an image processing method and apparatus which can be detected by image processing, have a simple recognition system and a high recognition rate, and have no influence on the human body, image quality, and the like, and a copier, printer, and scanner using the same. is there.

[0012]

In order to achieve the above-mentioned object, in the image processing method according to the present invention, predetermined image processing is performed on image data obtained by picking up an image of a document to be processed, Image information corresponding to a specific fiber (metal fiber in the embodiment) included in the original is extracted, and image data being processed is based on the specific original including the specific fiber based on the extracted image information. Whether or not it is a thing is determined (Claim 1).

It is preferable that the original is irradiated with light and an image formed on the basis of the reflected light from the specific fiber has a line of a predetermined width, and the predetermined width existing in the image data is used. Is to extract the image information by extracting the line (Claim 2).

More preferably, the determination as to whether or not the document is a specific document containing the specific fiber is made based on at least one of the length, density and distribution of lines existing in the image data. (Claim 3).

In addition to the determination processing based on the image processing of the specific fiber, the extraction / determination processing of the specific pattern existing in the original document is performed, and whether the final specific original document is determined based on the determination results of both of them. It is more preferable to make a judgment (claim 4).

Further, in the image processing apparatus according to the present invention, predetermined image processing is performed on image data obtained by picking up an image of a document to be processed, and the image data being processed is based on a specific document. In an image processing device that determines whether or not there is image information corresponding to a specific fiber included in the document, based on an extraction unit that extracts from the image data and a line extraction image output from the extraction unit. And a determination unit that determines whether or not the image data being processed is based on a specific document containing a specific fiber (claim 5).

The extracting means includes binarizing means for binarizing the multivalued image data to obtain a binary image, and line extracting means for extracting line image data of a predetermined width from the generated binary image. (Claim 6).

The extracting means includes a line removing means for removing a line having a predetermined width existing in the binary image, a binary image after the removal generated by the line removing means, and the line extracting means. The present invention can be configured by a processing means for generating a line image data having a predetermined width by taking a difference from the given binary image before the processing (claim 7).

Further, the judging means judges whether or not the length of the line existing in the given line image data is a predetermined length. It is determined whether the density of lines existing in the given line image data is within a certain range. The presence distribution area having a predetermined width existing in the given line image data is compared with the reference presence distribution area registered in advance, and it is determined whether or not the degree of coincidence is high. Say
It can be configured to include at least one of (Claim 8).

More preferably, a recognition means for recognizing a specific pattern existing in the image data to be processed is further provided, and predetermined judgment control processing is performed based on the recognition result of the recognition means and the recognition result based on the specific fiber. It is to provide a control means for controlling (claim 9).

According to each of the above-mentioned configurations, the specific fiber is present as a line of a predetermined width (referred to as "fine line" in the embodiment) in the image data obtained by imaging. Therefore, a line with a predetermined width existing in the image data is extracted, and it is determined whether or not the extracted line data is based on a specific fiber (feature data (line length, density, distribution, etc.) obtained from it. It is possible to determine whether or not the specific fiber is included by image processing. As described above, since the specific fiber can be extracted by image processing, a metal sensor or the like is not used, and accurate determination can be performed without affecting the human body or image quality.

Further, in the copying machine, scanner, and printer according to the present invention, any one of the above image recognition devices is mounted, and the image data output from the image reading means of the copying machine is used as the image converting means. The image recognition device may be input in parallel, or the input / output signals to the control means of the scanner or the printer may be input in parallel to the image recognition device.
~ 12).

Then, based on the image data obtained by the predetermined scan, it is determined whether or not the image data is a detection object such as a copy-prohibited object. A control signal was sent to control predetermined processing such as copying, image reading, and output (printout).

Further, when an attempt is made to copy a bill or the like using a copying machine equipped with the image processing apparatus of the present invention, or to read or print out an original using a scanner or printer (claims 10 to 12). ), When a specific pattern existing in the document to be processed is detected, a copy stop command or the like is issued, and the same item is not copied, read, or printed out.

[0025]

1 shows an example of an embodiment in which an image processing apparatus according to the present invention is installed in a copying machine according to the present invention. As shown in FIG. The image reading unit 1 reads a document and transfers the read image data (RGB signals) to the image forming unit (printer) 2. The image forming unit 2
Then, the given RGB signal is decomposed into each component of magenta, cyan, yellow and black which are ink colors, and a laser beam is applied to a predetermined position of the photosensitive drum based on the obtained YMC signal. A copy process is performed on the printout and the printout is made.

Then, the image data from the image reading section 1 is input in parallel with the image forming section 2 of the main body of the copying machine to the recognition section 4 constituting the image processing apparatus according to the present invention. The recognition unit 4 recognizes whether or not the given image data includes metal fibers representing a document whose copying is prohibited, and gives the recognition result to the copying machine control unit 3. There is.

Based on the obtained recognition result, the copying machine control section 3 outputs an image output control signal to the image forming section 2 if the read original is a copy-inhibited original, and a normal image is not output. To do so. Here, the processing for preventing the normal image from being output includes not only the interruption of the printout but also the case of degrading the image quality (painting black, reducing, etc.). The copying machine control unit 3 also controls the operations of the image reading unit 1 and the image forming unit 2 when a normal copying process is performed.

Next, each part will be described in detail. First, as shown in FIG. 2, the image reading unit 1 irradiates a document 6 placed on a document table 5 with light from a light source 7. The reflected light from the original 6 is reflected by the mirror 8 and guided to the lens 9. The lens 9 forms an image on the CCD 10, and the CCD 10 converts the formed image into an electric signal and outputs it. Further, since the electric signal of the CCD 10 is a minute signal, the CCD output is amplified by the amplifier 11.
It is input to the A / D converter 12 after being amplified by. In this way, digital image signals for the R, G and B signals are obtained.

The light source and the mirror are incorporated into one unit, and move (scan) on the document at a constant speed. The resolution in the sub-scanning direction can be changed by changing the moving speed.

As shown in FIG. 3, the recognizing unit 4 obtains a binarizing unit 15 for binarizing a given digital image signal and a binary image output from the binarizing unit 15, Whether or not the original document being processed is a detection target based on the thin line extraction unit 17 that extracts the thin line corresponding to the metal fiber existing in the binary image and the thin line image data output from the thin line extraction unit 17 And a feature determination unit 19 for determining.

More specifically, the binarization unit 15 includes an edge enhancement processing unit 15a and a floating binarization processing unit 15b. That is, since the metal fibers woven (mixed) into the paper are extremely thin, the scattered light on the metal surface is emphasized by the edge emphasizing process, and the data is converted into data in which thin lines are likely to emerge. Then, specifically, it can be configured by using the differential filter shown in FIG.

Further, in this example, in order to remove the influence of the background density, the binarization threshold value is changed according to the background density, and the optimum state (separation between the fine line image based on the metal fiber and the background) is performed. The thin line portion can be extracted as binary data. Specifically, as shown in FIG. 5, peripheral pixels (indicated by hatching in the figure) existing in a predetermined area (5 × 5 local area in this example) around the pixel of interest (indicated by cross hatching in the figure). ), The average value (average intensity) is obtained, and the threshold value is determined based on the average intensity of the surrounding pixels.

[0033]

(Equation 1) Then, if the threshold value for the pixel of interest is determined as described above, it is determined as a black pixel if the intensity of the pixel of interest is less than or equal to the determined threshold value according to the following determination formula, and the threshold value is determined. If it is larger than the value, it is determined to be a white pixel, and a binary image is generated based on the determination.

[0034]

(Equation 2) In the generated binary image, at least a portion corresponding to the metal fiber becomes a black pixel and forms a thin line image. The filter shape shown in FIG. 4 and the area of the peripheral pixel for obtaining the average shown in FIG. 5 are appropriately set based on the width of the metal fiber and the resolution of the image reading unit.

The thin line extracting unit 17 extracts only the portion based on the metal fibers from the binary image (including all the image information including the metal fiber portion) generated by the binarizing unit 15. The principle of thin line extraction is to generate image data with thin lines removed from a given binary image, and take the difference between the image data (after thin line removal) and the binary image data before processing to obtain thin line image data. I am trying to generate.

In order to perform such processing, as shown in FIG.
The thin line removing unit 20 and the subtraction processing unit 21 are provided. Then, the given binary image is input to the thin line removing unit 20 and the subtraction processing unit 21 in parallel, and the output of the thin line removing unit 20 is input to the subtraction processing unit 21.

Further, the thin line removing section 20 is composed of a reduction processing section 20a and an enlargement processing section 20b connected in series. Then, the reduction processing unit 20a scans a window of n1 × n2 pixels with respect to the contrast image (binary image), and if the number of black pixels in the window is i or more, the pixel of interest (generally the center pixel of the window). Is a black pixel, and if less than i, it is a white pixel. That is, since the thin line portion has many white pixels around the thin line portion, by appropriately setting i pixels, the target pixel becomes a white pixel when the region is a thin line.

On the contrary, the enlargement processing unit 20b scans the window of m1 × m2 pixels for the binary image data reduced by the reduction processing, and if the number of black pixels existing in the window is j or more, the pixel of interest is selected. (Generally, the center pixel of the window) is a black pixel. As a result, the black pixels reduced by the reduction process are restored to the original size. However, since the thin line portion in the original image has already been removed by the reduction process, it does not return even by the enlargement process, and binary image data in which only the thin line is removed can be formed by this.

The shape size of the window, i and j, which are threshold values for binarization, are set appropriately according to the width of the thin line to be removed. Further, the size of the window used in the reduction process and the size of the window used in the enlargement process may be the same or different. Also,
n1 and n2 may be equal (window shape is square) or may be different (window shape is rectangular). Similarly, m1 and m2 may be the same (window shape is square) or different (window shape is rectangular).

As an example, when erasing a thin line having a line width of 64 μm, it is necessary to erase the line width of one pixel at 400 DPI. Therefore, as shown in FIG. 6, assuming that one square corresponds to one pixel of 400 DPI, the size and shape (n1 × n2 pixels) of the window used for the reduction processing is 3 × 3 pixels, and the threshold is set to a black pixel. If the value i is 7 or more, fine lines with a line width of 2 pixels or less are removed. That is,
In the case where the line width is 2 pixels, assuming that a line having such a line width crosses the window, 6 out of 9 pixels forming the window become black pixels, which is less than 7 and disappears. (White pixels). Therefore, in the case of the image data as shown in FIG.
Will be modified as follows.

Then, the size and shape (m1 × m2 pixels) of the window used when performing the enlargement process is set to 3 × 3 pixels, and if there is one or more black pixels in this window, the pixel of interest is black (j = By adopting 1), enlargement processing can be realized. That is, when the image data of FIG. 7B is enlarged according to the above-described principle, a binary image as shown in FIG. 7C is obtained, and the thin line portion shown in FIG. I can see that In this way, the thin line removing unit 20 removes thin lines having a line width of a certain value or less.

Although it is the width of the fine line, since the light is scattered on the surface of the metal fiber, the line width read by the CCD is larger than the actual size of the metal fiber (see FIG. 7). Therefore, it is necessary to determine the line width to be removed in consideration of this light scattering. In the example of FIG. 6, the line width of 64 μm is removed, but this is the value of the image data read by the CCD. The metal fiber does not have a diameter of 64 μm, but even a diameter of a few tenths of a few μm appears as a thick line width due to light scattering on the metal surface. Therefore, when the diameter of the metal fibers mixed in the paper is large, this can be dealt with by, for example, increasing the size of the window during the reduction process and increasing the threshold value i.

As shown in FIG. 8, the subtraction processing unit 21 has a buffer function (original image buffer 21a, thin line removed image buffer 21b) and stores binary image data (original image) before thin line removal processing in the buffer 21a. The thin line removed image provided by the thin line removing unit 20 is subtracted from the original image by the processing unit 21c. As a result, as shown in FIG. 8, only the thin line portion remains in the original image, and the thin line extracted image is generated.

The data transferred from the thin line removing unit 20 and the original image data may be subtracted in real time without providing the thin line removed image buffer 21b. Further, the exclusive OR process may be performed instead of the subtraction process. That is, the subtraction process here means a process of taking the difference between the two and finally leaving the thin line portion.

The fine line image data generated as described above is sent to the feature determining section 19, where copying of the fine line image is prohibited based on the feature amount such as the length, density and distribution shape of the fine line. It is determined whether the metal fibers are included in the original. The following is an example of a specific determination process based on the above three feature amounts.

* Judgment based on the length of thin line As shown in FIG. 9, a window having an appropriate size is set (a judgment is made each time while scanning), and the number of pixels from the edge of the window to the black pixel is counted. , The difference in the number of pixels between the Nth row and the N + 1th row + 1 is added from the first row to the Mth row. By taking this added value as the length of the thin line, it is judged whether this length is within the range of the specified value ± tolerance, and it is judged whether or not the thin line image is the specified metal fiber.

As an example, for example, the fiber length is 8 ±
A document with a size of 1 mm is a special document. Therefore, FIG.
If one pixel in the window is 1 mm, the length of the fine line in the image becomes 9 mm according to the arithmetic processing shown in the table below, and it can be determined that the fine line corresponds to the metal fiber included in the special document.

[0048]

[Table 1] * Judgment based on fine line density The fine line density is easily obtained by scanning a window with a size that contains several metal fibers and sequentially counting the number of black pixels contained in the window. Then, it is determined whether the obtained density is within the range of the specified value + tolerance, and it is determined whether the thin line image is the specified metal fiber.

As an example, if a window area of 14 mm square has 3 to 5 fibers with an average length of 8 mm, it is assumed that the document is a special document. Then, in the example of FIG. 10, one pixel is 1 mm. Then, since there are 8 black pixels in one metal fiber, the number of black pixels (25 pixels in the illustrated example) existing in a 14 mm square area is divided by 8 pixels corresponding to the average length. The number of thin wires (about three) corresponding to the metal fibers contained in the area can be easily obtained, and in the illustrated example, it can be determined that the document is a special document.

Strictly speaking, the number of black pixels included in the area is large even if a large number of short metal fibers are present, and as a result, the range of a special manuscript may be reached. Therefore, in order to make a more accurate judgment, it can be combined with the above-described judgment processing based on the thin line length.

That is, first, a thin line length determination process is performed, and a black pixel portion determined to be not a predetermined length (too short / too long) is removed (white pixel is made). As a result, only the pixels forming the target metal fiber remain as black pixels in the image data after the processing. Therefore, the density can be obtained by counting the number of black pixels as described above.

Separately from this, a determination is made based on the length of the thin line, a labeling process is performed on the black pixels forming a desired metal fiber, a window for the determination based on the density is set, and the window is set. Density can also be determined by counting the number of labels present within.

* Judgment based on distribution shape of fine lines Although the metal fibers contained in the paper are uniformly contained in the paper, the fact that they appear in the image data after the thin lines are removed is not printed. , It is limited to the part where the background part of the paper is exposed and the part where the metal fiber can be seen through by the printing of the light density. Therefore, by extracting the portion where the thin line exists (distribution shape of the thin line) and determining the conformity (coincidence / similarity) between the distribution shape and the distribution shape registered in advance, the read original Determines whether the document is a copy prohibited document.

As an example, for example, a manuscript (original image) is shown in FIG.
As shown in FIG. 1 (A), by performing the thin line extraction processing, a thin line extracted image as shown in FIG. It is determined whether or not a thin line is included in the range of k × k pixels in the thin line extracted image, and if it is included, the pixel of interest is determined to be a thin line. Thus, the distribution (presence / absence) of the image fine line is obtained, and the result is plotted to obtain the fine line distribution shape as shown in FIG. Then, it is determined whether or not the read document is a copy-prohibited document by pattern matching with the distribution shape (see FIG. 7D) registered in advance in the image processing apparatus (recognition unit).

In the actual judgment processing, each judgment processing method described above may be appropriately selected and processed independently, or the judgment methods selected appropriately may be combined to make a comprehensive judgment. good. Further, the determination may be made using a method other than the above.

FIG. 12 shows another embodiment of the present invention. In the above-described embodiment, the image processing apparatus that recognizes only the metal fibers is described. However, in this example, it is configured by combining with a conventional specific pattern-based recognition unit that recognizes a pattern printed on a document. doing.

That is, the image signal output from the image reading unit 1 is transferred to the image forming unit 2 and the recognition unit 4 for metal fibers.
In addition to this, the recognition unit 29 for symbols is provided in parallel. Since various known algorithms can be used as the recognition algorithm in the pattern recognition unit 29, detailed description thereof will be omitted. Then, the respective recognition units 4 and 29 perform a predetermined recognition process, and the judgment result is given to the copying machine control unit 3.

Then, the copying machine control section 3 judges whether or not it is a copy prohibited document based on the given recognition result, and if it is judged as a copy prohibited document, it outputs a predetermined control command. I have to. Here, as concrete determination methods, the following three methods can be cited, but the method is not limited to these, and each determination result may be weighted to make a comprehensive determination. Well, various judgment methods can be performed.

Judgment Method 1 When the recognition result of either the metal fiber recognition unit 4 or the pattern recognition unit 29 is a copy-prohibited document, it is judged as a copy-prohibited document, and the copying machine control unit 3 makes a copy. Stop or deteriorate the output image.

Judgment method 2 When the recognition results of both the metal fiber recognition unit 4 and the pattern recognition unit 29 determine that the copy prohibition manuscript, it is judged that the copy prohibition manuscript, and the copying machine control unit 3 stops copying. Or deteriorate the output image. In this case, the probability of erroneous recognition can be reduced.

Judgment Method 3 If the degree of conformity, which is the recognition result of the pattern recognition section 29, is high, the copying machine control section 3 sends a resolution control signal to the image reading section 1 to increase the resolution. By increasing the resolution, fine metal fibers can be read more clearly by the metal recognition unit 4, and highly accurate metal fiber recognition becomes possible.

Normally, a color copying machine reads four times,
In order to complete copying, if the degree of conformity of the pattern recognition unit is high in the first reading, it is possible to perform the optical resolution reading exclusively for the metal fiber recognition between the first and second readings. . Then, the resolution switching is performed as shown in FIG.
This can be easily realized by changing the scanning pitch (sub-scanning direction) of the document.

Although each of the above-described embodiments has been described as being applied to a copying machine, the present invention is not limited to this, and for example, a color scanner, a color printer, a FAX, a communication transmission device, and various other types. Of course, it can be applied to the device.

As an example, as a scanner,
A configuration as shown in FIG. 13 can be adopted. That is, the scanner is roughly classified into an input unit 30 and a control unit 31.
And an output unit 32. And input section 3
At 0, the original is scanned with light from a light source, and the reflected light (transmitted light) obtained from the original is detected by a photoelectric conversion element such as a CCD, a photomultiplier, or a photodiode, converted into an electric signal, and sent to the control unit 31. I am supposed to send it. And the control unit 31
In the above, the electric signal from the input unit is increased or decreased, predetermined gradation correction, contour enhancement, and other image processing are performed, and the corrected signal is sent to the output unit 32.

Further, the output unit 32 is adapted to perform data conversion on the basis of the signal given from the control unit 31 and output it to a predetermined output device. In other words, when the scanner and printer are separated (only for pure reading), the image data read temporarily is stored in the storage device (output device) to send information to a separately formed printer. Therefore, a predetermined process necessary for the writing process is performed.

When the output device is a printer or the like (which is arranged in an integrated device), a predetermined electric-optical conversion is performed and a predetermined signal conversion for writing on a paper medium (photosensitive material). Perform processing. The specific configuration of each part is
Since a known general one can be used, detailed description thereof will be omitted.

Here, in the present invention, the image processing device 33 is provided, and the signal relating to the image data obtained from the input unit 30 is input to the image processing device 33 together with the control unit 31. The image processing device 33 may be any device as long as it has a function of detecting the metal fiber shown in each of the above-described embodiments by image processing.

Then, the image processing apparatus 33 carries out a predetermined process based on the given image data to obtain the degree of conformity and the like, and sends the determination result (output prohibition signal) to the control unit 31. As a result, the control unit 31 stops the signal output of the output unit 32 based on the determination result.

FIG. 14 shows an example of the embodiment of the printer according to the present invention. That is, image data (electrical signal) is given to the input unit 34 directly from the scanner or via a medium such as a storage device. Then, the control unit 3
After performing a predetermined image conversion process (converting into data according to its own output mechanism) in 5, a predetermined electro-optical conversion process is performed in the output unit 36 to convert the image data provided on the photosensitive material. It is supposed to be reproduced.

By the way, in the case where the scanner is not provided with the image processing device 33 having the work stopping means for the detection processing of the metal fibers as described above, the image data is not read even if the original is a read prohibited image. Will read.

Therefore, the output signal of the control unit 35 is given to the image processing device 37 (equivalent to the above-mentioned image processing device 33), and predetermined image processing is carried out there, and at least copy prohibited objects based on image information based on metal fibers. It is determined whether or not. Then, if the image processing device 37 determines that the object is prohibited, it outputs a prohibition signal to the control unit 35. Then, the data output from the control unit 35 to the output unit 36 is stopped.

[0072]

As described above, in the image processing method and apparatus according to the present invention, attention is paid to the fact that the image data obtained by picking up an image of a specific fiber (metal fiber) becomes a line with a predetermined width. This makes it possible to recognize whether or not the medium includes the specific fiber. Therefore, it can be detected by image processing without using a metal sensor or the like, the recognition system is simple and the recognition rate is high, and the human body and the image quality are not affected.

Further, when configured as in Claims 4 and 9,
It is possible to perform more highly accurate recognition processing by using the detection processing of prohibited objects such as copying by recognizing a specific pattern that has been generally performed conventionally (in the embodiment, the symbol recognition unit).

By mounting the image processing apparatus in a copying machine, a scanner, or a printer, the output of the copy is surely prohibited for the copy prohibited articles such as bills and securities. , It is possible to copy / output an image different from an original (copy prohibited article), and to stop reading or printing of the original.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall configuration of an example of a preferred embodiment of a copying machine according to the present invention.

FIG. 2 is a diagram showing a specific configuration of the image reading unit.

FIG. 3 is a block diagram showing an internal configuration of an image recognition device used for it.

FIG. 4 is a diagram illustrating an example of a differential filter used in an edge enhancement processing unit.

FIG. 5 is a diagram illustrating a function of a floating binarization processing unit.

FIG. 6 is a diagram illustrating a function of a thin line removing unit.

FIG. 7 is a diagram illustrating a line width of a fine line image based on metal fibers.

FIG. 8 is a diagram illustrating a function of a subtraction processing unit.

FIG. 9 is a diagram illustrating a function of a determination unit.

FIG. 10 is a diagram illustrating a function of a determination unit.

FIG. 11 is a diagram illustrating a function of a determination unit.

FIG. 12 is a diagram showing another embodiment of a copying machine according to the present invention.

FIG. 13 is a diagram showing an example of a scanner according to the present invention.

FIG. 14 is a diagram showing an example of a printer according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 image reading section 2 image forming section 3 copier control section 4 recognition section 15 binarization section 17 thin line extraction section 19 feature determination section 20 thin line removal section 21 subtraction processing section 29 pattern recognition section 30 input section 31 control section 32 output section 33 image processing device 35 control unit 36 output unit 37 image processing device

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Hitoshi Nakamura, 10 Hanazono Dodocho, Ukyo-ku, Kyoto City, Kyoto Prefecture Omron Corporation (72) Inventor Masahiro Akagi 10 Hanazono-Tudocho, Ukyo-ku, Kyoto Prefecture Co., Ltd. (72) Inventor Tomoji Kagaya 10 Odoron-cho, Hanazono-cho, Ukyo-ku, Kyoto City, Kyoto Prefecture Omron Co., Ltd.

Claims (12)

[Claims] 1. A predetermined image process is performed on image data obtained by picking up an image of a document to be processed, image information corresponding to a specific fiber contained in the document is extracted, and the extracted image information is extracted. An image processing method for determining whether or not the image data being processed is based on a specific original document containing a specific fiber. 2. A line of a predetermined width existing in the image data is utilized by utilizing that an image formed by irradiating the original with light and reflected light from the specific fiber has a line of a predetermined width. The image processing method according to claim 1, wherein the image information is extracted by extracting the image information. 3. The determination as to whether or not the document is a specific document containing the specific fiber is made based on at least one of the length of lines, density of existence, and distribution state in image data. Item 2. The image processing method according to Item 2. 4. The determination processing based on the image processing of the specific fiber, and the extraction / determination processing of the specific pattern existing in the document are performed, and whether the final specific document is determined based on the determination results of both of them. Any one of claims 1 to 3 in which the determination is made.
The image processing method described in the item. 5. Image processing for performing predetermined image processing on image data obtained by picking up an image of a document to be processed and determining whether or not the image data being processed is based on a specific document. In the apparatus, the image information corresponding to the specific fiber contained in the original document,
An extracting means for extracting from the image data, and a determining means for determining whether or not the image data being processed is based on a specific document containing specific fibers based on the line extraction image output from the extracting means. Image processing device equipped. 6. The binarizing unit for binarizing multivalued image data to obtain a binary image, and the line extracting unit for extracting line image data of a predetermined width from the generated binary image. The image processing apparatus according to claim 5, further comprising: 7. A line removing unit for removing a line having a predetermined width existing in the binary image by the extracting unit, and a binary image after removal generated by the line removing unit,
7. The image processing apparatus according to claim 6, further comprising a processing unit that generates a line image data having a predetermined width by calculating a difference from the unprocessed binary image given to the line extraction unit. 8. The determination means includes at least one of the following determination functions (1) to (5).
The image processing device according to claim 1. It is determined whether or not the length of the line existing in the given line image data is a predetermined length. It is determined whether the density of lines existing in the given line image data is within a certain range. The existence distribution area having a predetermined width existing in the given line image data is compared with a reference existence distribution area registered in advance,
Determine if the degree of agreement is high. 9. A control which further comprises a recognition means for recognizing a specific pattern existing in the image data to be processed, and which carries out a predetermined judgment control process based on the recognition result of the recognition means and the recognition result based on the specific fiber. The image processing apparatus according to claim 5, further comprising a unit. 10. An image reading unit for reading at least a document, an image converting unit connected to the image reading unit for converting the read image data into a signal for printing, and an output from the image converting unit. A copier provided with an image forming unit for performing a predetermined printing process, the image recognition apparatus according to any one of claims 5 to 9 is mounted, and image data output from the image reading unit is provided. The image recognition device inputs the image data in parallel with the image conversion means, and the image recognition device determines whether or not the document being copied is a detection target based on the image data obtained by a predetermined scan. A copying machine which controls the copying by sending a control signal to a predetermined processing means of the copying machine when the object is judged to be a detection object. 11. An input device for reading a document, a control device connected to the input device for performing a predetermined image conversion process on the read image data, and an output device connected to receive the output of the control device. A scanner comprising: an output unit that outputs data to the image recognition device according to any one of claims 5 to 9, wherein the image recognition device outputs the image data output from the input unit to the control unit. And input to the image recognition device in parallel with the image recognition device, and the image recognition device determines whether or not the document being copied is a detection target based on the image data obtained by a predetermined scan, and detects it. A scanner adapted to control a reading process by sending a control signal to a predetermined processing means of the scanner when it is judged as an object. 12. A printer comprising: a control unit that performs a predetermined image conversion process on given image data information; and an output unit that receives an output of the control unit and performs a predetermined print process. The image recognition device according to any one of 5 to 9 is mounted, and image data input to a printer is input to the image recognition device in parallel with the output means, and the image recognition device is provided with Based on the image data, it is determined whether the image data being processed is a detection target object, and when it is determined as a detection target object, a control signal is sent to a predetermined processing means of the printer to control the output processing. Made printer.