JP2003331337A - Device for judging color image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a color image judgment device of high precision capable of detecting hardly usual copy paper erroneously as valuable paper. <P>SOLUTION: The first judgment part 5 cuts out a local area from an image data read out from a scanner 1 to judge whether it is one belonging to the valuable paper or not, the result thereof is stored in the third judgment part, and an intermediate data and result such as a feature quantity are stored in a memory part 7. The second judgment part 8 reads out the data stored in the memory part 7 to judge the propriety as the valuable paper as to a broad area, and sends the result thereof to the third judgment part 6. The third judging part 6 conducts final judgement based on the judged results in the first and second judgment parts. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the prevention of counterfeiting of securities such as banknotes, and more particularly to a color image determination technique for determining whether or not an input color image is an image of securities.

[0002]

2. Description of the Related Art Various methods have been devised to prevent counterfeiting of color image output devices. For example, JP-A-7-4
In Japanese Patent No. 7717, image features are parameterized and compared with the features of a predetermined image (security). If it is determined that they are the same, printing is suppressed. Also, Japanese Patent Laid-Open No. 7-1232
In Japanese Patent Publication No. 49, the image is encoded, further quantized, and compared with a predetermined image (securities). If it is determined that they are the same, the data is replaced with invalid data to form an image, and the image is output.
Further, in Japanese Patent Application Laid-Open No. 8-2010, feature points are detected from an image, a predetermined amount is calculated, a feature amount is calculated, and it is determined whether or not the feature amount of a predetermined image (security certificate) is included. If this happens, suppress printing.

[0003]

However, according to the conventional method, even an ordinary manuscript that is not securities is often judged as securities if it includes some of the characteristics of the securities. In such a case, there was a problem that the copy operation and the output by the printer could not be performed. The present invention has been made in consideration of the above-mentioned circumstances, and an object of the present invention is to provide a color image determination device with high accuracy in which an original is usually not erroneously detected as securities.

[0004]

In order to solve the above-mentioned problems, the invention of claim 1 of the present invention is a color image judging apparatus for judging whether or not an image is prohibited from being copied.
First determining means for determining whether the securities are securities based on the data of the local predetermined area, and second determining means for determining whether the securities are large-scale areas larger than the local predetermined area. And determining the final judgment from the judgment results of the first and second judging means.
According to a second aspect of the present invention, in the color image determination device according to the first aspect, the first determination means calculates and determines a feature amount by using a local feature included in securities. The second determination means is characterized in that the determination is made by using the extent of distribution of the local area having a characteristic amount close to securities. According to a third aspect of the present invention, in the color image determining apparatus according to the first or second aspect, the second determining unit determines using the feature amount calculated by the first determining unit. Is characterized by. Also,
According to a fourth aspect of the present invention, in the color image determining apparatus according to any one of the first to third aspects, the first determining unit determines a feature amount by using a local feature included in securities. It is characterized in that when it is calculated and compared with a predetermined threshold value and the characteristic amount is large, it is determined as securities. The invention according to claim 5 of the present invention is the color image determination device according to any one of claims 1 to 3, wherein the second determination means has a securities size having a characteristic amount larger than a predetermined threshold value. If the number of areas is larger than a predetermined number, it is determined that the area is not securities. According to a sixth aspect of the present invention, in the color image determination device according to any one of the first to fifth aspects, the threshold value to be compared with the feature amount is a value used for the first determination means. It is characterized in that it is larger than the value used for the second judging means. Further, the invention of claim 7 of the present invention is the color image determination device according to any one of claims 1 to 6,
Each determination condition in the determination means is characterized by being set / cancelled by a serviceman.

[0005]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of a color image determining apparatus of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of a color image output device using a color image determination device of the present invention. The image read by the scanner 1 is output to the printer 4 via the image processing / color conversion unit 2. However, when the third determination unit 6 finally determines that the securities are securities, a predetermined plate is printed. (Color toner) is solid coating part 3
Then, the image is solidly painted and output to the printer 4. Regarding this printing suppression method, in addition to solid painting, a certain plate is removed (color output is not performed) or a special image (for example, the character prohibition)
Methods such as embedding are devised. The first determination unit 5 cuts out a local area from the image data read by the scanner 1 and determines whether the securities are securities. The result is sent to the third determination unit and the intermediate value of the feature amount and the like. Data and results are stored in the storage unit 7. The second determination unit 8 reads the data stored in the storage unit 7, determines whether or not the global area is securities, and sends the result to the third determination unit 6. The third determination unit 6 makes a final determination from the determination results of the first and second determination units. User I / F 9
Are for setting the thresholds of the first and second determination sections and the conditions of the third determination section. However, in order to prevent general users from making arbitrary settings, only a serviceman can set the hidden commands. In this embodiment, the scanner 1, the image processing / color conversion unit 2, the printer 4
, Etc., but specific color image (securities)
It is not directly related to the judgment of.

FIG. 2 is a block diagram showing an embodiment of the first judging section 5. Image data is input in RGB and is temporarily FIFO
Stored in 20. It is cut out from the cutout mask 21 and converted into a feature amount by the feature amount conversion unit 22. The feature amount conversion unit 22 calculates a feature amount of a predetermined area by various methods such as taking a color histogram, encoding / quantizing, recognizing a figure. For example, in order to see the degree of color mixture of a specific part of securities, the frequency distribution of each RGB color is examined, the frequency component is analyzed by orthogonal transformation, and the color tone and shape of a specific color pattern are analyzed. Or The feature amount conversion unit 22 further compares the result with the feature of the securities, and converts how similar the securities are to the matching degree (feature amount). And the comparator 2
In step 3, the characteristic amount is compared with a certain threshold value to determine whether it is securities. Further, this characteristic amount is sent to the storage unit 7 and accumulated. The above operation is performed on the entire area where the image is scanned, and is accumulated in the storage unit 7 as a feature amount for each local area as shown in FIG. 7 described later.

Next, an embodiment of the second judging section 8 will be described. FIG. 3 is a configuration diagram showing an embodiment of the second determination unit 8. In this embodiment, there are three passes, and in the upper pass, when the number of local regions whose feature amount is larger than the predetermined value (threshold value 1) in the scan region is found to be larger than the predetermined number (threshold value 3), It is judged as securities. In the lower path, if the local area in which the feature amount is larger than the predetermined value (threshold value 2) is distributed in a range wider than the securities size, it is determined that the securities are not securities. Looking at the upper path in detail, the features are cut out from the memory with a mask about the size of securities. The feature value is threshold 1
The number of regions exceeding (the same value as the threshold of the first determination unit) is measured by the comparison measuring device 30. The comparator 31 outputs 1 (valued securities) if the number is larger than the threshold value 3, and outputs 0 otherwise. The middle path outputs 1 (which is securities) when the output of the comparison and measurement device 30 is larger than 0 by the comparator 32, and outputs 0 otherwise. In the lower path, the comparative measurement device 33 measures the number of feature amount regions exceeding the threshold value 2. The threshold value 2 is to see how close it is to the characteristic amount of securities, and is set smaller than the threshold value 1. Comparator 34 is 1, if the number is greater than 0,
Otherwise, 0 is output. The counter 35 is a comparator 34.
Measure the number of 1 for each securities size. sync_msk
The signal is a signal synchronized with the size of the mask (security size). This pass is for checking whether a certain amount of features are distributed in a wider range than the securities size. Therefore, counting is performed in synchronization with the size of the mask (size of securities). The comparator 36 is
When the output of the counter 35 is larger than the threshold value 4, 1 (not securities) is output, and when it is smaller, 0 is output. Then, in the final stage, the results of each pass are integrated. The truth table is shown in FIG. That is, if the upper path is judged as 1 (security securities), the final judgment is unconditionally 1 (for securities), and if the middle path is 1 and the lower path is 0, the final judgment is 1. Becomes Other final judgments are 0 (not securities). This judgment is made by the OR circuit 37 and the NOT circuit 3.
8 and AND circuit 39.

FIG. 5 is a block diagram showing an embodiment of the third judging section 6. In the third determination unit 6, the selector 52 selects the logical product of the first and second determination units formed by the AND circuit 51 or the result of the first determination unit as the final determination result. The selection is made by a serviceman. FIG. 6 is a diagram showing the relationship among pixels, local areas, and global areas. The local area is made up of a plurality of adjacent pixels, and is large enough to contain a pattern that characterizes securities. Strictly speaking, considering, for example, a situation in which securities are copied by a color copier, it is not known at which position and at what angle the securities are placed. The cut-out mask 21 shown in (4) cuts out the image, and the feature amount conversion unit 22 calculates the feature amount of this local area. The global area is composed of a plurality of the local areas.

FIG. 7 is a diagram showing an example of the feature quantity output from the first judging section 5. In this example, a feature amount (matching degree) of 0 to 10 is calculated for each local region, and it is assumed that the larger the numerical value, the higher the probability of securities. For example, if the threshold value given to the comparator 23 in FIG. 2 is 8, the shaded portion in the figure exceeds the threshold value, and it is determined that the securities are securities. The area shown in FIG. 7 is an area scanned by the scanner, and corresponds to, for example, a sheet size of A4. FIG. 8 shows a processing unit of the lower path shown in FIG. For example, if the threshold value 2 in FIG. 3 is 6, the feature amount in the local region is 7 or more (circled numbers in the figure).
Is measured by the comparative measuring device 33. The comparator 34 outputs 1 if the number is larger than 0, and outputs 0 otherwise.
The counter 35 measures the number of times of 1. However, it is measured in synchronization with the securities size. FIG. 9 is a diagram for explaining the input to the comparator 34. The result measured by the comparative measuring device 33 has a value as shown in FIG. This corresponds to the number of circled numbers in the securities size of 3 × 3 from the upper left of FIG. Whereas FIG. 8 shows the feature amount in units of local area,
It should be noted that FIG. 9 is a numerical value in securities size unit. The comparator 34 outputs 1 if it is larger than 0, and outputs 0 otherwise. FIG. 10 shows the input value to the counter synchronized with the security size. The counter 35 measures the number of 1s (in this example, if "..." is set to 0, the output of the counter 35 finally becomes 5), that is, the counter 35 exceeds the threshold value 2 in the scan area (somewhat valuable). It counts how many securities-sized areas (which have the characteristics of securities). The comparator 36 outputs 1 (not securities) when the value becomes larger than the threshold value 4, and outputs 0 otherwise. The threshold value 4 is a threshold value for determining how wide the distribution area is for securities. For example, the threshold 4 may be set to 3 if the securities are not securities if they are distributed in four or more regions.

[0010]

As described above, according to the present invention, if a feature very similar to securities is found in a scanned image, it is judged as securities, but if a certain number of suspicious features are found, the securities will not be recognized. Since it is judged that it is not securities, images that closely resemble securities can be surely judged, and on the other hand, images that are not very similar are found, and it is not harmful even if it is judged that it is not securities, copying operation and print output are unnecessary. Deterring can be avoided. Further, since the feature amount calculated by the first determining unit is used by the second determining unit, the amount of calculation can be reduced and the time required for determination can be shortened.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a color image output device using a color image determination device of the present invention.

FIG. 2 is a configuration diagram illustrating an example of a first determination unit.

FIG. 3 is a configuration diagram illustrating an example of a second determination unit.

FIG. 4 is a diagram showing a truth table of a second determination unit.

FIG. 5 is a configuration diagram illustrating an example of a third determination unit.

FIG. 6 is a diagram showing a relationship among a pixel, a local area, and a global area.

FIG. 7 is a diagram showing an example of feature quantities.

FIG. 8 is a diagram showing a processing unit of the lower path shown in FIG.

FIG. 9 is a diagram illustrating an input to a comparator 34.

FIG. 10 is a diagram showing an input value to a counter synchronized with a security size.

[Explanation of symbols]

5 First determination unit, 7 Storage unit, 8 Second determination unit, 9
User I / F

Continued front page    F-term (reference) 2H134 NA05 NA14 NA29 NA30 NA32                       NA43                 3E041 AA01 BA11 BA12 BC06 CA01                       CA09 CA10 CB03 DB01                 5C077 LL14 MP08 PP32 PP65 PQ20                       TT06                 5L096 AA02 BA03 BA18 FA23 FA35                       GA51 JA11

Claims (7)

[Claims] 1. A color image determining apparatus for determining whether or not an image is forbidden to copy, and a first determining unit for determining whether or not the image is securities based on data in a local predetermined area, Second determining means for determining whether or not the securities are large-scale areas larger than the local predetermined area, and determining the final judgment from the judgment results of the first and second judging means. A color image determination device characterized by. 2. The first determining means calculates and determines a feature amount by using a local feature included in securities, and the second determining means localizes a feature amount close to securities. The color image determination apparatus according to claim 1, wherein the determination is performed by using a degree of spread of the distribution of the target area. 3. The color image determination apparatus according to claim 1, wherein the second determination unit makes a determination using the feature amount calculated by the first determination unit. 4. The first determining means calculates a feature quantity by using a local feature included in the security and compares the calculated feature quantity with a predetermined threshold value to determine that the security is securities. The color image determination device according to claim 1, wherein the color image determination device is a color image determination device. 5. The second determination means determines that the securities are not securities when the number of securities size regions having a feature amount larger than a predetermined threshold is larger than a predetermined number. 4. The color image determination device according to any one of 1 to 3. 6. The threshold to be compared with the feature quantity is the first
6. The color image determining apparatus according to claim 1, wherein the value used for the determining means of is larger than the value used for the second determining means. 7. The color image determination device according to claim 1, wherein each determination condition in the first and second determination means can be set / cancelled by a serviceman. .