JP2001157045A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor capable of preventing forgery, without hindering copying of an ordinary original. SOLUTION: This processor is provided with a correspondence degree detecting part 6 for detecting the degree as to whether image data corresponds to an image formation inhibiting article, a color tone changing part 7 for changing the color tone of the image data, when the correspondence degree is high, and an image-forming part 2 for forming an image on a recording medium, based on this image data where the color tone is changed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to forming an image on a recording medium based on image data, and more particularly to an image processing apparatus capable of preventing forgery without hindering normal copying of an original.

[0002]

2. Description of the Related Art In recent years, counterfeiting of banknotes and securities using a color copying machine (including a facsimile machine) has occurred, which has become a social problem. In order to prevent such a case, it is desired that the color copying machine be provided with a forgery prohibition function. This forgery prohibition function is that the copying machine automatically recognizes the image of the image forming prohibited object and attempts to copy the image so that the image cannot be copied. As an image recognition technique for realizing the forgery prohibition function, for example, there is a technique in which image data of an image forming prohibited property is stored in advance, and image data read from a document at the time of copying is checked by pattern matching. Also, a method of extracting image features from image data has been considered in order to solve the huge amount of stored data, which is a drawback of pattern matching. For example, there is a method in which a density histogram for each color is extracted for the entire image data, and the similarity between a histogram pattern of a read image and a histogram pattern obtained from an image forming prohibited object is compared. In the case of comparison using the density histogram for each color, the inclination of the image has no effect, and this also compensates for the drawback of pattern matching.

[0003] As an operation of recording in the forgery prohibition function, if image data read as a result of collation of an image pattern or a histogram corresponds to an image formation prohibition property, the image data is blacked out when forming an image. Is known. As a result, an image is recorded on the output sheet for the time being, but the recorded image is an image that can be distinguished from an image formation prohibited object.

[0004]

However, when performing a comparison determination, it is not possible to determine that an image read from a banknote can be formed due to contamination of the banknote or the like. For example, if the image patterns and histograms are similar even if they do not completely match, it is better to determine that image formation is impossible. The wider this comparison, the less likely the forgery will be successful. However, there may be a case where an image read from a non-banknote document happens to be similar to a banknote. When the range of comparison is widened, the number of cases where it is determined that image formation is not possible even for a document other than a bill increases. If an image is not formed even though a normal document which is not an image forming prohibited property is read, the operator is confused.

It is an object of the present invention to solve the above-mentioned problems and to provide an image processing apparatus capable of preventing forgery without hindering copying of a normal document.

[0006]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention provides an appropriateness detecting unit for detecting the degree of whether image data corresponds to an image forming prohibited property, and an image detecting means for determining whether the image data is high. A color tone changing unit for changing the color tone of the data,
And an image forming unit that forms an image on a recording medium based on the image data whose color tone has been changed.

[0007] The color tone changing section may change the color tone by changing a characteristic color in the image data to another color.

[0008] An image formation prohibiting section for prohibiting image formation when the degree of applicability is extremely high may be provided.

[0009]

An embodiment of the present invention will be described below in detail with reference to the accompanying drawings.

As shown in FIG. 1, an image processing apparatus according to the present invention comprises a reading unit 1 for reading image data of a document,
Image forming unit 2 that forms an image on paper based on image data
A histogram extracting unit 3 for extracting a density histogram for each color from the read image data, a histogram storage unit 4 for storing a density histogram for each color of the image of the image forming prohibited object in advance, and detecting the size of the document from the image data. Comparison unit 5 that compares the size of the image forming prohibited object with the size
And a histogram stored in the histogram storage unit 4 and a histogram extracted from the image data to determine whether or not the image data corresponds to an image formation prohibited property. A color tone changing unit 7 that changes the color tone of the image data when it is higher than a preset change threshold, and an image formation prohibiting unit 8 that prohibits image formation when the relevance is higher than a preset prohibition threshold. Prepare.

[0011] Hereinafter, an image formation prohibited property is a bill.

The reading unit 1 is, for example, a scanner that causes a line image sensor to run on a document placed on a glass plate. The document may be run on a stationary image sensor. Scanner scans R
(Red), G (green), and B (blue) are read and digitized.

The image forming section 2 is, for example, an ink jet printer for sub-scanning ink nozzles on a running paper, but need not be an ink jet printer as long as it can form a color image. Ink jet printers have Y (yellow), M stored in the image memory.
The ejection control of the ink nozzles of each color is performed according to four types of signals (magenta), C (cyan), and K (black). Although not shown, the image processing apparatus is provided with means for converting image data of RGB signals read from a document into image data of YMCK signals.

The histogram extracting section 3 extracts a density histogram for each color by counting the number of pixels for each density value for each color from a set of R, G, B density values of each pixel constituting the image data. It is. For example, if the concentration is 2
Assuming that the values are digitized in 56 steps, the density histogram of one color is represented by 256 numerical values.

The histogram storage unit 4 stores color density histograms extracted in advance for all bills.

The size comparison unit 5 obtains a Y signal in an XYZ color space based on the CIE (International Commission on Illumination) 1931 color system based on the image data of the RGB signals. , A color area is cut out, and the size (eg, area) of the document is obtained. The size of the manuscript is compared with the size of the banknote stored in advance, and if the sizes are the same, the appropriateness check is performed. If the sizes are different, the appropriateness check can be omitted.

The relevance detector 6 detects the relevance P using the density histogram for each color of each banknote and the density histogram for each color of the image data. For example, the relevance P can be calculated as a ratio of the number of pixels for each density value of each color, the absolute value of the difference is summed up for all the density values, and the sum sum occupies the total pixel number S. The relevance P is determined for each of the three colors R, G, and B. As an example for one color, as shown in FIG. 2, the number of pixels for each density value n of the banknote stored in the histogram storage unit 4 is m, and the number of pixels of the image data of the corresponding density value is a. Then, the total sum T of the absolute values of the differences in the number of pixels for each density value n is

[0018]

(Equation 1)

And the relevance P is

[0020]

(Equation 2)

## EQU1 ## The relevance P indicates the degree of whether or not the read image data indicated by the broken line in FIG. 2 and the histogram waveform 21 corresponds to the image forming prohibited article (banknote) indicated by the solid line and the histogram waveform 22.

The color tone changing section 7 operates when it cannot be determined whether or not the original from which the image data is based is a bill. That is, for all three colors of R, G, and B, if the relevance P is lower than the inhibition threshold T _A but higher than the change threshold T _B , the color tone of the image data is changed. Change threshold T _B, for example, to 80%. Here, as a method of changing the color tone, the characteristic color in the image data is replaced with another color. The characteristic color is
When there is a banknote having a high degree of matching P with respect to the density histogram for each color of the image data, this is a color that the banknote has characteristically, specifically, a color having a large number of pixels of a specific density value. .

The image formation prohibiting section 8 is activated when there is a high possibility that the document from which the image data is based is a bill. That is, for all three colors R, G, and B, if the corresponding degree P is higher than the prohibition threshold T _A , image formation is prohibited. The prohibition threshold value T _A is, for example, 95%.

The operation of the image processing apparatus will be described with reference to FIG.

First, when an operator sets a document on the reading section 1 and operates to start reading, the reading section 1 operates to read three kinds of signals of R, G and B. The histogram extraction unit 3 extracts R, G, and B color density histograms. The size comparing unit 5 converts the RGB signals into the XYZ color system, creates a histogram of the Y signals in the XYZ color system, and calculates the area of the document. The area of the document is compared with the area of the banknote stored in advance. If the areas are the same, the process proceeds to the check of the relevance. If the area is different,
It is determined that the document is not a bill, and the process proceeds to normal copy processing.

In the checking of the degree of relevance, a check of image formation prohibition is performed first, and then a check of color tone change is performed.
Therefore, R, G, respectively, for Color density histogram of B obtaining the appropriate degree by the above equation, but sequentially compares with the corresponding degree P obtained prohibited threshold T _A, the relevant degree P is If it is lower than the prohibition threshold value T _A , the comparison for other colors is omitted, and the process immediately proceeds to the color tone change check. All three colors Conditional degree P is higher than the prohibition threshold value T _A, prohibits copying process.

[0027] In checking the color changes sequentially, compares with the corresponding degree P obtained and change the threshold value T _B, if applicable degree P is lower than the change threshold value T _B, the comparison for other colors Immediately after omitting, the process proceeds to the normal copy process. All three colors Conditional degree P is higher than the change threshold T _B, the process proceeds to discoloration copy process. In normal copy processing, an image is formed on paper based on the original image data. In the color change copying process, an image is formed on paper based on the image data whose color tone has been changed by the color tone changing unit.

According to the above operation, when the original is a bill, it is expected that the relevance P is 95% or more even in consideration of the contamination of the bill. Conversely, if the relevance P is 95% or more, it can be determined that the document is a bill. Therefore, when the degree of applicability P is 95% or more, copying is not allowed to prevent forgery. When the relevance P is between 80% and 95%, it cannot be determined whether or not the original is a bill. At this time, the characteristic color of the bill is changed to another color and copied. If the original is a bill, a copy output in a color tone that can be easily distinguished from an actual bill is obtained, so that the purpose of forgery is not achieved. If the original is not a bill, the color tone is different from that of the original, but since the image shape is maintained, a copy can be obtained for the time being.

In the above embodiment, the image forming prohibited property is a banknote. However, the present invention can be widely applied to securities, stock certificates, and the like.

Although the image data is read from the original by the reading unit, the present invention can be applied to image data obtained by image communication and image data synthesized by a computer.

Further, the density histogram for each color is extracted as the feature amount of the image, and the degree of relevance P is obtained from the density histogram for each color. The present invention can be implemented by detecting the corresponding degree from the comparison of the matching result and the feature amount.

[0032]

The present invention exhibits the following excellent effects.

(1) In the case of image data having a high degree of relevance because it happens to be similar to an image-forbidden property, an image can be formed for the time being, although the color is changed,
Usually, there is no big trouble in copying the original.

(2) In the case of image data having a high degree of relevance in an attempt to copy an image-forbidden property, the image is discolored and formed, so that the purpose of forgery cannot be achieved.

[Brief description of the drawings]

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

FIG. 2 is a diagram of a density histogram for explaining the degree of matching in the present invention.

FIG. 3 is an operation flowchart of the image processing apparatus of FIG. 1;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Reading part 2 Image formation part 3 Histogram extraction part 4 Histogram storage part 5 Size comparison part 6 Relevance detection part 7 Color tone change part 8 Image formation prohibition part

Claims (3)

[Claims] 1. A matching level detecting unit for detecting whether image data corresponds to an image forming prohibited property, a color changing unit for changing a color tone of the image data when the matching level is high, and a color changing unit for changing the color tone. An image forming unit for forming an image on a recording medium based on the obtained image data. 2. The image processing apparatus according to claim 1, wherein the color tone changing unit changes the color tone by changing a characteristic color in the image data to another color. 3. The image processing apparatus according to claim 1, further comprising an image formation prohibition unit that prohibits image formation when the degree of relevance is extremely high.