JP3322958B2 - Print inspection equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for inspecting printed matter, and more particularly to an apparatus for determining the type of printing defect.

[0002]

2. Description of the Related Art As a method of inspecting a printed matter, when a difference between read image data of a printed matter and a reference value exceeds a set value, it is determined that an error point has occurred, and a printing failure is determined based on the number of error points. A method for determining is known. There is also known a method of detecting a printing defect by determining a difference value between read image data and reference data for each predetermined pixel at a predetermined determination level.

Further, according to the method disclosed in Japanese Patent Laid-Open No. 2-20348, a detection signal is compared with a reference signal to determine a white loss if there is a difference smaller than a predetermined value and a black loss if there is a difference larger than a predetermined value. How to do is shown.

[0004]

A defect detection method based on density comparison, which is widely used as a conventional printed matter inspection method, determines whether printing is good or bad based on physical criteria. There is a problem that it is difficult to obtain inspection results consistent with human judgment such as oil stains and roller scratches.

As described above, various improvements have been made to the inspection method based on the density comparison. However, these methods have the following problems. In other words, in the method of detecting a printing defect by determining a difference value between read image data and reference data at a predetermined determination level for each predetermined pixel,
Although there is a feature that a threshold value for determination can be changed for each pixel, there is no mechanism for determining this value, and it takes time and effort for a human to set the threshold value. That is, the inspection is performed by physical judgment, and a human adjusts the threshold, that is, the physical quantity by trial and error.

A read image signal is compared with a reference signal as disclosed in JP-A-2-20348. If there is a difference smaller than a predetermined value, white loss occurs. In the method of determining, a simple physical and simple criterion is used. Therefore, with such a method, it is not possible to know the detailed state of the printing defect such as the degree, position and meaning of the defect.

[0007] As described above, the conventional printed matter inspection apparatus performs only the defect determination based on the simple physical quantity. The correspondence between the physical quantity and the type of the defect and the correspondence between the defect and its print area are described below. There is no means to store and use information that humans can judge in detail. Further, each method is an invention relating only to a print defect inspection, and there is no invention relating to automation relating to an operation for taking measures against a printing machine or the like based on the result of the print defect inspection.

[0008]

According to the present invention, there is provided a printed matter inspecting apparatus for inspecting a printed matter of a printed matter, comprising: reading means for reading a printed image of the printed matter; and image data outputted from the reading means. A comparing unit that compares the standard image data of the printed matter, a first feature extracting unit that extracts a pixel having an absolute value of a difference equal to or more than a certain value from a comparison result obtained from the comparing unit, A second feature extraction unit for calculating a connection amount in which pixels having a difference equal to or more than a certain value continuously exist from the obtained comparison result, and a difference having a certain value or more from the comparison result obtained from the comparison unit. A third feature extraction means for determining the direction of a connected component in which pixels are continuously present, and a fourth feature extraction means for calculating an area of a pixel having a difference equal to or more than a predetermined value from a comparison result obtained from the comparison means. And means, and a determining means for determining the type of printing defect from the characteristics of the printed material to be output from the first to fourth feature extraction means.

[0009]

The printing defects of the printed matter are classified into the types of defects that can be easily understood by humans. Furthermore, depending on the type of defect,
An appropriate countermeasure is output to the printing device.

[0010]

FIG. 1 is a block diagram showing an example of a print inspection apparatus according to the present invention. As shown in FIG. 1, the print inspection apparatus according to the present invention includes a sensor 101, an image input unit 102, an image storage unit 103, a positioning unit 105, and a reference image storage unit 1.
04, data comparison unit 106, defect candidate extraction unit 107, difference code extraction unit 108, difference absolute value extraction unit 109, defect connected component extraction unit 110, defect direction feature extraction unit 111, defect area extraction unit 112, defect type determination unit 113, a defect determination unit 114, a determination result output unit 115, a countermeasure determination unit 118, a printing press countermeasure output unit 116, a display unit 117, a parameter input / output unit 120, an editing unit 119, and a data storage unit 121.

The inspection object P is, for example, a printed matter mainly composed of a line drawing as shown in FIG. 2, and the printed matter passes through a sensor 101 such as a CCD by a transport mechanism (not shown) to collect image data. . Sensor 101
The image output signal of the image input unit 102 including the A / D converter
A / D-converted, and recorded in the image storage unit 103 as digital image data. In the image storage unit 103, data is recorded on the memory in a form as shown in FIG. 3 or FIG. As shown in the figure, the image signal has a size of 1024 pixels horizontally and 512 pixels vertically, and is recorded as image data in which each pixel has 256 gradations.

The image data of the inspection object P recorded as described above is corrected by the positioning unit 105 for the displacement of each sample. This is performed by moving the print image in parallel and changing the origin 0o to 0n, as shown in the conceptual diagram of FIG. The positioning unit 105 is specifically realized by the configuration shown in FIG. The buffer control unit 601 outputs the address signal ADR1 to the image storage unit 103, reads the image data from the image storage unit 103, transmits the image data to the binarization processing unit 603, and writes the image data to the image buffer 602. . For the image data read from the image storage unit 103 as shown in FIG. 4, the binarization processing unit 603 sets a threshold value with a constant value THR (for example, 80H = 128) in order to extract a print portion having a small pixel value. Perform processing. As a result, as shown in FIG. 7, the image data is converted into a binary image having a pixel value of “1” for the print portion and “0” for the paper / background. The unit 604 and the horizontal pixel accumulating unit 605 obtain accumulated data in each direction as indicated by P1 and P2 in FIG. 7, respectively. The start points of the accumulated data P1 and P2 are detected by the start point detection units 606 and 607, respectively, and the start point coordinate data X0 and Y0 are obtained. Next, the image data address conversion unit 608 sets (X0, Y0) to (0, Y0) with respect to the address signal ADR2 output from the buffer control unit 601.
0) and a new address signal ADR
3 is output. Specifically, the conversion represented by the following equation is performed using the address ADR2 of the memory storing the data of the coordinate values (xs, ys) to generate ADR3.

ADR3 = ADR2-Y0 × 1024-X0 (1) According to the address signal ADR3, the image buffer 60
2 is read from the image buffer 609.
Is written to. With respect to the image of FIG.
When the position adjustment is performed by the image data 05, the image data is stored in the image buffer 609 in a data form as shown in FIG.

In the data comparing unit 106, the image data shown in FIG.
Compared with the reference image data written in 04 as shown in FIG. 9, the difference between the pixel values is obtained. The data comparison unit 106 is specifically realized by the configuration shown in FIG. The buffer control unit 1001 outputs the address signal ADR4, and simultaneously reads out the input image data from the positioning unit 105 and the reference image data from the reference image storage unit 104 for each pixel. The difference between the two read image data is calculated by a differentiator 1002 for each pixel, and the result is written to an image buffer 1003. When the difference between the input image of FIG. 8 and the reference image data of FIG. 9 is obtained, two-dimensional data having a difference value as a pixel value as shown in FIG. 11 is obtained in the same form as the image data. Data is stored in the image buffer 1003. Note that the difference image data of FIG. 11 is stored in the form of 8 bits per pixel in which the absolute value is represented by the lower 7 bits and the sign is represented by the 8th bit.

For the difference image data calculated by the data comparison unit 106 and stored in the form shown in FIG. 11, a defect candidate extraction unit 107, a difference code extraction unit 108, a difference absolute value extraction unit 109, a defect connection component extraction A processing unit 110, a defect direction feature extraction unit 111, and a defect area extraction unit 112 extract feature amounts relating to defects.

The defect candidate extraction unit 107 extracts pixels whose value is equal to or more than a fixed value THR2 (for example, 5) or less than or equal to a certain value THR3 (for example, -5) from the difference data with respect to the reference image, and simultaneously reads the read pixels Is divided into small areas of 5 × 5 pixels, and the number of extracted pixels in the area is measured. The defect candidate extraction unit 107 is specifically realized by the configuration shown in FIG. As shown in FIG. 12, the defect candidate extraction unit 107 reads the difference image data stored in the data comparison unit 106 using the address signal ADR5 output from the buffer control unit 1201, and
2, when the data value is equal to or greater than THR2, the comparator 120
In No. 3, "1" is output when the data value is equal to or less than THR3. The OR operation circuit 1204 is provided with the two comparators 12
The OR of the outputs 02 and 1203 is obtained. This logical sum data is written into the image buffer 1207 in a form as shown in FIG. Here, in FIG. 13, the blank is a numerical value.
The logical sum data is simultaneously transmitted to the memory data adder 1205. The address converter 1206 outputs the address ADR5 output from the buffer controller 1201.
Is converted into an address corresponding to every five pixels in the horizontal and vertical directions, and the converted address
05 and 1208 are controlled. Memory data adder 12
The step 05 counts the number of pixels of the logical sum “1” in the 5 × 5 small area, and records the counting result in the image buffer 1208. That is, the data shown in FIG. 13 is divided into small areas of 5 × 5 pixels as shown by the solid lines in FIG. 14, and the number of pixels having the value “1” in the small area is determined by an image having a data structure as shown in FIG. The data is stored in the buffer 1208.

The difference sign extraction unit 108 extracts the sign of the pixel value whose absolute value is equal to or more than a certain value, for example, 5 or more, from the difference image data shown in FIG. Specifically, the difference code extraction unit 108 is realized by the configuration shown in FIG. Address signal A output from buffer control section 1201 (FIG. 12)
According to DR5, the difference image data is read from the data comparison unit 106. When the pixel value is equal to or larger than THR2 (for example, 5) in the comparator 1601, the comparator 1602 specifies THR3.
When the pixel value is (for example, −5) or less, “1” is output. The OR of the outputs of the two comparators 1601 and 1602 is calculated by a logical OR calculator 1603,
Bit 0 and bit 1 are stored in the image buffer 1604 together with the output of the comparator 1601. As a result, the extraction result of the difference code extraction unit 108 for the difference image data of the data comparison unit 106 shown in FIG.
FIG. 18 corresponding to the plus and minus signs indicated by 7
"3 (11 in binary)" or "1 (0 in binary)
1) The data is represented by the value of ". FIG.
The image buffer 1604 stores the image data in which the difference code is represented by the value indicated by 8. Note that, in FIG. 18, a blank column indicates a pixel value “0”.

The absolute difference extracting unit 109 obtains the absolute value of each pixel from the differential image data output from the data comparing unit 106 as shown in FIG. The difference absolute value extraction unit 109 is specifically realized by the configuration shown in FIG. In accordance with the address signal ADR5 output from the buffer controller 1201 (FIG. 12), the absolute value calculator 1901 forcibly sets the eighth bit of each data to "0" in the signal read from the data comparator 106. Thus, the image data is converted into the absolute value and stored in the image buffer 1902. For example, the data of FIG.
, And stored in the image buffer 1902. In the defect connected component extraction unit 110, the data comparison unit 1
A defect is extracted from the difference image data as shown in FIG. 11 output from 06, and its connected component is detected. The connected component extraction unit 110 is specifically realized by the configuration shown in FIG. According to the address signal ADR5 output from the buffer control unit 1201 (FIG. 12), the signal read from the data comparison unit 106 is compared with the threshold value THR by the comparator 2101.
(E.g., 5) or more, the comparator 2102 sets the threshold value TH
When it is less than R (for example, -5), it is converted to "1". Thereafter, the logical sum data of the outputs of the two comparators 2101 and 2102 is obtained by the logical sum calculator 2103 and stored in the image buffer 2104. The image buffer 2104 stores image data in a data structure as shown in FIG. After the writing to the image buffer 2104 is completed, the CPU 2105 controls the image buffer 2104. The CPU 2105 performs processing in accordance with the flow shown in FIGS. 22 and 23 written in the program memory 2106 to obtain a connected component of the binary image written in the image buffer 2104. As shown in FIG. 22, the binary image written in the image buffer 2104 is represented as f (i, j) as an input image, and
The label image output to 108 is expressed as l (i, j) (ST1). The term "label" used herein is a number for distinguishing a defective image composed of defective pixels connected to each other, and each pixel constituting one defective pixel connection has the same number (label). ) Is appended.
Also, a label table for obtaining the correspondence between label values is secured in the working memory 2107 in the form shown in FIG. 24, and the initial value of f (i, j) is set to 0. For a binary image f (i, j), x =
The pixel is examined from 1, y = 1, and if the pixel value is “1”, S
Proceed to T5. In ST5, for the label values of the five neighboring pixels shown in FIG. 25, the value T (l (i, j)) in the label table is referred to (1) the number of types of values in the label table n, and (2) the minimum The value L1 of the label table and (3) the value L2 of the next smallest label table (these values are only two at the maximum) are obtained. In the following ST6, ST7 and S are determined by the number n of types of values in the label table.
The process branches to T8 and ST9. If n = 0,
Since it means that a new label has occurred at the position of (x, y), the connected component number λ is increased by 1, λ is substituted into the label table T (λ), and the label image l (x, y ) Is assigned to, and label information is allocated (ST8). And n =
In the case of 1, since (x, y) is a state where the neighboring components are dot-dotted, the minimum label table value L1 is substituted into the label image l (x, y) and integrated (ST7). When n = 2, FIG.
5 indicates that there are two components to be integrated in the vicinity area of the label image l.
For (x, y), the value L1 of the smaller label table is substituted, and all other areas taking L2 are replaced with L1 (S
T9). By applying this processing to the entire image, labeling ends when ST13 ends. Further, the values of T (i) having missing numbers between them are rearranged into serial numbers (ST1).
4) By replacing the label images with the values of the rearranged label table, label images in which the connected components are numbered sequentially from 1 are obtained. For example, when the processing shown in FIGS. 22 and 23 is performed on the binary image data recorded in the image buffer 2104 as shown in FIG.
The label image data as shown in FIG. 7 is obtained. Here, the blank in FIG. 27 represents the numerical value “0”. This processing result is stored in the image buffer 2108.

The defect direction feature extraction unit 111 extracts the directionality of the defect from the difference image data as shown in FIG. 11 output from the data comparison unit 106. The defect direction feature extraction unit 111 is specifically realized by the configuration shown in FIG. According to the address signal ADR5 output from the buffer control unit 1201 (FIG. 12), the difference image data read from the data comparison unit 106 is equal to or larger than the threshold value THR2 (for example, 5) in the comparator 2701, and is compared in the comparator 2702. When THR3 (for example, −5) or less, it is converted to “1”. Then, two comparators 2701
And 2702 output the OR data to the OR operator 27
03 and stored in the image buffer 2704. The image buffer 2704 is stored in a data structure as shown in FIG. After the writing to the image buffer 2704 is completed, the CPU 2705
Control. The CPU 2705 is a program memory 27
The processing is performed according to the processing flow shown in FIGS. 29 and 30 written in the image buffer 06, and the feature of the directional component of the binary image written in the image buffer 2704 is obtained. FIG.
As shown in FIG. 9, f (i, j) is the binary image data initially written in the image buffer 2704, and d (i, j) is image data having the directional feature to be output to the image buffer 2708 as a pixel value. Define and start the process (ST2)
1). Then, for each pixel of the image, FIG.
A product-sum operation value with the weights of the four types of neighboring pixels shown in FIGS. 33 and 34 is obtained (ST25 to ST30). The four parameters of FIG. 31, FIG. 32, FIG. 33, and FIG. 34 are for detecting a vertical direction, a horizontal direction, a diagonally rising leftward, and a diagonally falling leftward component, respectively. The product sums of these four types of parameters and pixels are obtained as m1, m2, m3, and m4, respectively, and it is determined in ST31, ST33, ST34, and ST35 whether or not each of them takes a predetermined threshold THR4 (for example, 3) or more. Then, 1 (vertical), 2 (horizontal), 3
(Upward diagonal) and 4 (downward diagonal) are defined as d (x,
y) (ST39). By performing this process on the entire image, the binary image stored in the image buffer 270 as shown in FIG. 13 is converted into image data as shown in FIG. Is done. Here, in FIG. 36, a blank represents a numerical value “0”.

The defect area extraction unit 112 is a data comparison unit 1
The area is obtained for each connected component with respect to the difference image data as shown in FIG. The defect area extraction unit 112 is specifically realized by the configuration shown in FIG. In accordance with address signal ADR5 output from buffer control section 1201 (FIG. 12), data comparison section 106
Is read by the comparator 3601 as a threshold value TH.
When the value is equal to or greater than R2 (eg, 5), the comparator 3602 converts the value to “1” when the value is equal to or smaller than the threshold value THR3 (eg, −5). Thereafter, the logical sum data of the outputs of the two comparators 3601 and 3602 is obtained by the logical sum calculator 3603 and stored in the image buffer 3604. Image buffer 3604
Stores image data in a data structure as shown in FIG. After the writing to the image buffer 3604 is completed, the CPU 3605 controls the image buffer 3604. The CPU 3605 includes a program memory 3606
38, 39, 40, and 41 written in the image buffer 3604, the connected components of the binary image written in the image buffer 3604 are obtained, and the area of each connected component is obtained. As shown in FIG. 22 and FIG.
From T44 to ST63, the same labeling process as that performed by the defect connected component extraction unit 111 is performed. On the other hand, the number of pixels of each label is measured up to the maximum label value LM obtained in ST57 (ST64 to ST72). At the same time, the image data is converted into image data as shown in FIG. 42 in which an area value is assigned to each connected component (ST73 to ST79). As a result, the binary image data recorded in the image buffer 3604 is converted into image data in which the area of the connected component as the pixel value as shown in FIG. 42 and corresponding data of the label area shown in FIG. The data is written to the buffer 3608 and the data buffer 3609. Here, in FIG. 42, a blank represents a numerical value “0”.

The difference image data output from the data comparison unit 106 is subjected to six types of feature extraction processing, and the result is transferred to the defect processing determination unit 113.

The defect type discriminating section 113 estimates and outputs the type of defect from the characteristic amounts and their degrees using the plurality of characteristic amounts extracted in this way as input data, and is realized by the configuration shown in FIG. Data from the six extraction units, which is input data, is transmitted via transceivers 4101 to 4108. This is because the CPU 4109 has the transceiver 4
This is realized by reading necessary data from the image buffers or data buffers of the six extraction units via 101 to 4108. The CPU 4109 stores the program memory 41
The processing is performed according to the procedure shown in FIGS. As shown in FIG. 45 and FIG. 46, the value of the two-dimensional data as shown in FIG. 14 or FIG. If it is not 0, five pieces of information are examined for the small area to obtain information about the type of defect.

As shown in FIG. 45, in ST103, the first
As the third detection amount, the feature amount data (FIG. 18) stored in the difference code extraction unit 108 via the transceiver 4103 is read out, and the plus or minus of the larger number of pixels in the small area of 5 × 5 pixels is read. Select either sign. A feature amount / defect type correspondence table 4112 in which correspondence data between the feature amount of a defect and its value is described (FIGS. 47 and 48).
The defect type candidate D1 is obtained by collating with the feature amount (ST104).

As a second detection amount, in ST 105, the absolute difference value extracting unit 1
The feature amount data (FIG. 20) stored in the area 09 is read out, and the largest absolute value is detected in a 5 × 5 small area. A feature amount / defect type correspondence table 4112 in which correspondence data between the feature amount of a defect and its value is described (FIGS. 47 and 48).
And the feature amount are compared to obtain a defect type candidate D2 (ST106).

As a third detection amount, in ST 107, feature amount data (FIG. 27) stored in defect connected component extraction section 110 is read out via transceiver 4105, and differs in a 5 × 5 small area. Detect the number of connected components. A feature amount / defect type correspondence table 4112 in which correspondence data between the feature amount of a defect and its value is described (FIGS.
8) is compared with the feature amount to obtain a defect type candidate D3 (ST108).

As the fourth detection amount, in ST109, the data (FIG. 35) stored in the defect direction feature extraction unit 111 is read out via the transceiver 4106, and the direction feature which is the largest in the small area is set to the vertical and vertical directions. Horizontal, diagonally up left,
Select one of the diagonally lower left. The feature amount / defect type correspondence table 4112 (FIGS. 47 and 48) in which correspondence data between the feature amount of the defect and its value is described is collated with the feature amount to obtain a defect type candidate D4. (ST110).

As a fifth detection amount, in ST 111, the defect area extraction unit 11
2 (FIG. 42) is read, and the largest area value is detected. The feature quantity / defect type correspondence table 4112 (FIGS. 47 and 48) in which correspondence data between the feature quantity of the defect and its value are described is collated with the feature quantity to obtain a defect type candidate D5 (ST112). . The logical sum of the obtained five defect type candidates D1, D2, D3, D4, and D5 is set as the defect type of the small area (S
T113). Here, the characteristic amount shown in FIG. 47 and FIG.
The defect type correspondence table 4112 describes the name and the degree of the defect. The name of the defect is simply ORed.
Those with no information on the degree are not used as OR. For example, two logical ORs of blur (degree: dark) and blur (degree: unspecified) are blur (degree: dark).
If the pixel value of the small area data recorded in the defect candidate extraction unit 107 is 0, the detection process is bypassed and “no defect” is set (ST102). When this process is performed on all the small area data, two-dimensional array data having a print defect as shown in FIG. 49 can be obtained as a result. An embodiment of another defect type determination unit 113 will be described. This is realized by the configuration shown in FIG. 44 as in the first example. The feature amount / defect type correspondence table 4112 specifies a feature amount range corresponding to the defect type as shown in FIG.
The type of defect is output for each obtained feature amount.

Subsequently, the defect judging unit 114 judges the presence or absence of a two-dimensional array data defect as shown in FIG. 49 in which data indicating the type of defect is described for each 5 × 5 small area. The defect determination unit 114 is realized by the configuration shown in FIG. The output data of the defect type determination unit 113 is
Is input via the CPU 4703. CPU 4703
FIG. 51 stored in the program memory 4704
The processing is performed according to the procedure shown in FIG. The coordinate / region correspondence database 4702 stores the range of the x-coordinate and the y-coordinate for each of the partial print regions set in advance as shown in FIG. The relationship between the name and the area number is described as shown in FIG. As shown in FIG. 54, the print defect database 4706 includes, as reference information for determining a print defect, the defect information determined by the defect type discriminating unit for each print area and the small area of the defect candidate included in the print area. It is described in terms of numbers. As shown in FIG. 54, the allowable lower limit value of each defect type is different between the various regions in the upper left corner and the central region. As shown in FIG. 51, for each small area output from the defect type discriminating unit, two-dimensional data (FIG. 49) describing the type of defect is printed for each characteristic print as shown in FIG. Processing is performed on each of the divided areas. ST
At 122, the type of defect and the number of the small areas are obtained for the small areas included in each print area, as shown by the broken lines in FIG. 55, using the information of the coordinate / area correspondence database 4702 (FIG. 53). Subsequently, in ST123,
Print defect database 47 based on the result of ST122
It is determined whether the defect is a defect to be detected using the reference information 06 (FIG. 54). This processing is performed for each print area shown in FIG. 50, and the determination result is output to the output circuit 4708 and the data buffer 4709.

Finally, the countermeasure determining unit 118 determines whether the print defect determined by the defect determining unit 114 is correct.
13 to determine countermeasure information for the printing press. The countermeasure determination unit 118 is realized by the configuration shown in FIG. The transceiver 5301 obtains defect information from the defect type determination unit 113, and the transceiver 5302 obtains defect information from the defect determination unit 114. The CPU 5303 determines the corresponding information of the printing press according to the processing procedure of FIG. 57 written in the program memory 5304. Here, the printing press countermeasure information database 5306 describes defect types and countermeasure information for printing presses for each area as shown in FIG. As shown in FIG.
The presence or absence of a defect is determined for each of the small areas using the defect determination data (FIG. 54) output from the printer (ST132), and output to the printing press using the information of the printing press countermeasure information database 5306 (FIG. 58). Information is selected (ST133).
This processing is performed for each print area shown in FIG. 52, and the determination result is output from the output circuit 5307.

The judgment result output from the defect judgment unit 114 is output from the judgment result output unit 115 as a system selection signal and simultaneously to the display unit 117. The printing press countermeasure information also output from the countermeasure determination unit 118 is output from the printing press countermeasure output unit 116 to the display unit 117 to alert the operator. Output.

The defect type determining section 113 and the defect determining section 11
4. The parameters of the countermeasure determination unit 118 are controlled via the parameter input / output unit 120, and these parameters are adjusted while reading data from the data storage unit 121 or displaying an image as shown in FIG. can do.

In the above-described invention, the defect type is determined for each small area shown in FIG. 14, but the defect type can be determined for different units. FIG. 60 shows a second embodiment of the present invention in which a defect is determined for each connected component of a pixel. According to FIG. 60, according to the present invention, the sensor 101 and the image input unit 10
2. Image storage unit 103, positioning unit 105, reference image storage unit 104, data comparison unit 106, defect candidate extraction unit 1
07, difference code extraction unit 108, difference absolute value extraction unit 10
9. Defect direction feature extraction unit 111, defect area extraction unit 11
2. Defect type determination unit 113, defect determination unit 114, determination result output unit 115, countermeasure determination unit 118, printing press countermeasure output unit 116, display unit 117, parameter input / output unit 120, editing unit 119, data storage unit 121 Consists of The defect candidate extraction unit 107 has the same configuration as the defect connected component extraction unit 110 in FIG. 1 that extracts connected components. afterwards,
For each connected component, the difference code extraction unit 108 determines the difference code, the difference absolute value extraction unit 109 determines the difference absolute value, the defect direction feature extraction unit 111 determines the direction of the defect, and the defect area extraction unit 112 determines the area. The defect type is extracted by the defect type determination unit 113 for each connected component, and the defect determination unit 114 determines each connected component based on the position. Each of these processing units has the same configuration as that of the first embodiment. Here, since the degree of the defect is determined for each connected component to make a comprehensive determination, the feature amount / defect type correspondence table 42, 48, or 59 in the defect type determination unit 113 is slightly different. FIG. 48 shows a print defect database in the defect determination unit 114.
The printing press countermeasure determination database FIG. 58 in the countermeasure determination unit 118 is slightly different from that of the first embodiment.
Editing unit 119 in any correspondence table and database
To correct the data each time.

The gist of the present invention does not change even if other related feature values are used among the feature values. For example, together with the area indicating the spread of the defect, the vertical length and the horizontal length of the defect are obtained by, for example, projection of a binary image and used for determination, or a feature in which the same defect is scattered exists within a certain distance, for example. It is also possible to obtain the number and use it for the determination.

It should be noted that the present invention is not limited to the above-described embodiment, and it is needless to say that various modifications can be made without departing from the spirit of the invention.

[0035]

As described above, the print inspection apparatus according to the present invention performs registration with respect to an input image to be inspected, detects a plurality of features from data obtained by comparing the standard image with the input image, and detects the characteristic. It is possible to extract the type and the degree of the printing defect from the feature amount, determine the printing defect to be dealt with from the type and the degree of the extracted printing defect, and output the printing press countermeasure information from the information.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a print inspection apparatus according to one embodiment of the present invention.

FIG. 2 is a diagram showing an example of a printed material to be inspected according to an embodiment of the present invention.

FIG. 3 is a diagram showing a form of image data stored in an image storage unit according to one embodiment of the present invention.

FIG. 4 is another diagram showing a form of image data stored in an image storage unit according to one embodiment of the present invention.

FIG. 5 is a conceptual diagram showing an origin at which positioning is performed by a positioning unit according to one embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration of a positioning unit according to one embodiment of the present invention.

FIG. 7 is a diagram showing an example of data accumulated by a positioning unit according to one embodiment of the present invention.

FIG. 8 is a diagram illustrating an example of an image that has been aligned by the alignment unit according to an embodiment of the present invention.

FIG. 9 is a diagram showing an example of reference image data stored in a reference image storage unit according to one embodiment of the present invention.

FIG. 10 is a diagram showing a configuration of a data comparison unit according to one embodiment of the present invention.

FIG. 11 is a diagram illustrating an example of a processing result of a data comparing unit according to an embodiment of the present invention.

FIG. 12 is a diagram showing a configuration of a defect candidate extraction unit according to one embodiment of the present invention.

FIG. 13 is a diagram showing an example of pixels extracted by a defect candidate extraction unit according to one embodiment of the present invention.

FIG. 14 is a diagram illustrating an example of a small area determined by a defect candidate extraction unit and the number of extracted pixels according to an embodiment of the present invention;

FIG. 15 is a diagram showing an example of a processing result of a defect candidate extraction unit according to one embodiment of the present invention.

FIG. 16 is a diagram showing a configuration of a difference code extracting unit according to one embodiment of the present invention.

FIG. 17 is a diagram showing an example of code data extracted by a difference code extraction unit according to one embodiment of the present invention.

FIG. 18 is a diagram showing an example of a processing result of a difference code extracting unit according to one embodiment of the present invention.

FIG. 19 is a diagram showing an example of a difference absolute value extracting unit according to one embodiment of the present invention.

FIG. 20 is a diagram showing an example of a processing result of a difference absolute value extraction unit according to one embodiment of the present invention.

FIG. 21 is a diagram showing a configuration of a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 22 is a view showing a processing procedure performed by a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 23 is a diagram showing a processing procedure performed by a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 24 is a diagram showing an example of a label table performed by a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 25 is a diagram showing a neighborhood area of an input image of interest in a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 26 is a diagram showing a region near a label image of interest in a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 27 is a diagram showing an example of a processing result of a defect connected component extraction unit according to one embodiment of the present invention.

FIG. 28 is a diagram showing a configuration of a defect direction feature extraction unit according to one embodiment of the present invention.

FIG. 29 is a diagram showing a processing procedure performed by a defect direction feature extraction unit according to one embodiment of the present invention.

FIG. 30 is a diagram showing a processing procedure performed by a defect direction feature extraction unit according to one embodiment of the present invention.

FIG. 31 is a diagram showing parameters for detecting a vertical defect according to an embodiment of the present invention.

FIG. 32 is a view showing parameters for detecting a lateral defect according to one embodiment of the present invention.

FIG. 33 is a diagram showing parameters for detecting a defect in a diagonally left-upward direction according to one embodiment of the present invention.

FIG. 34 is a view showing parameters for detecting a defect in a diagonally left-downward direction according to one embodiment of the present invention.

FIG. 35 is a diagram showing an example of direction data detected by a defect direction feature extraction unit according to one embodiment of the present invention.

FIG. 36 is a view showing an example of a processing result of a defect direction feature extraction unit according to one embodiment of the present invention;

FIG. 37 is a diagram showing a configuration of a defect area extraction unit according to one embodiment of the present invention.

FIG. 38 is a diagram showing a processing procedure performed by a defect area extraction unit according to one embodiment of the present invention.

FIG. 39 is a view showing a processing procedure performed by a defect area extraction unit according to one embodiment of the present invention.

FIG. 40 is a diagram showing a processing procedure performed by a defect area extraction unit according to one embodiment of the present invention.

FIG. 41 is a view showing a processing procedure performed by a defect area extraction unit according to one embodiment of the present invention.

FIG. 42 is a diagram showing an example of a processing result of a defect area extraction unit according to one embodiment of the present invention.

FIG. 43 is a view showing another example of the processing result of the defect area extraction unit according to one embodiment of the present invention;

FIG. 44 is a diagram showing a configuration of a defect type determination unit according to an embodiment of the present invention.

FIG. 45 is a diagram showing a processing procedure of a defect type determination unit according to one embodiment of the present invention.

FIG. 46 is a diagram showing a processing procedure of a defect type determination unit according to one embodiment of the present invention.

FIG. 47 is a view showing an example of a feature amount / defect type correspondence table according to an embodiment of the present invention;

FIG. 48 is a view showing an example of a feature amount / defect type correspondence table according to an embodiment of the present invention;

FIG. 49 is a diagram showing an example of the processing result of the defect type determination unit according to one embodiment of the present invention;

FIG. 50 is a diagram showing a configuration of a defect determination unit according to one embodiment of the present invention.

FIG. 52 is a view showing a processing procedure of a defect judgment unit according to one embodiment of the present invention.

FIG. 52 is a diagram showing an example of a print area used in the defect determination unit according to one embodiment of the present invention.

FIG. 53 is a diagram showing an example of a coordinate / region correspondence database of the defect determination unit according to one embodiment of the present invention.

FIG. 54 is a diagram showing an example of a printing press correspondence information database of a defect determination unit according to an embodiment of the present invention.

FIG. 55 is a diagram showing the correspondence between the print area and the small area of the defect determination unit according to one embodiment of the present invention.

FIG. 56 is a diagram showing a configuration of a measure judging unit according to one embodiment of the present invention.

FIG. 57 is a view showing a processing procedure of a measure judging unit according to one embodiment of the present invention;

FIG. 58 is a view showing an example of a print defect database of a measure judging unit according to one embodiment of the present invention;

FIG. 59 is a view showing an example of a feature amount / defect type correspondence table according to an embodiment of the present invention;

FIG. 60 is a block diagram showing a configuration of a print inspection apparatus according to another embodiment of the present embodiment.

[Explanation of symbols]

P: printed matter, 101: sensor, 102: image input unit, 1
03 ... image storage unit, 104 ... reference image storage unit, 105 ...
Positioning unit, 106 ... data comparing unit, 107 to 112
... Feature extracting section, 113... Defect type determining section, 114... Defect determining section, 115... Determination result output section, 116... Printing machine countermeasure output section, 117.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G01N 21/84-21/958 G06T 1/00-9/40

Claims (2)

(57) [Claims] 1. A printed matter inspecting apparatus for inspecting a printed matter of a printed matter, a reading means for reading a printed image of the printed matter, and a comparing means for comparing image data output from the reading means with standard image data of the printed matter. And a first feature extraction unit that extracts a pixel having an absolute value of a difference equal to or greater than a certain value from the comparison result obtained from the comparison unit. Second to calculate the amount of connection where the pixels have
From the comparison result obtained from the comparison means, a third feature extraction means for determining the direction of a connected component in which pixels having a difference equal to or more than a certain value are continuously present, and A fourth feature extraction unit for calculating an area of a pixel having a difference equal to or greater than a predetermined value from the comparison result; and determining a type of a print defect from the features of the printed matter output from the first to fourth feature extraction units. A printed matter inspection device, comprising: 2. The printed matter inspection apparatus according to claim 1, further comprising a countermeasure information generating unit configured to generate countermeasure information for a printing machine for coping with the print defect determined by the determination unit.