JP3231840B2 - Image 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 image inspection apparatus for comparing an image to be inspected with a reference image and detecting a difference between these images, and in particular, can detect a defect having a small density difference. The present invention relates to an image inspection device.

[0002]

2. Description of the Related Art Defective printed materials include stains on printing paper,
There are various things such as poor printing. Even if the defective portion is minute, it may cause a problem in the quality of the printed matter. Further, when a printed product, such as a rotary printing machine, is continuously running at a high speed, the inspection of each printed material must be performed at a high speed.

Conventionally, printed matter inspection has been performed manually by various methods. For example, depending on the type of printing machine and the type of printed matter, the printed matter after printing is appropriately extracted and visually inspected. Alternatively, for a printed material that continuously runs at a constant speed, a strobe light is emitted in synchronization with the running speed, and the printed material is visually inspected.

In recent years, various techniques for automatically performing printed matter inspection have been disclosed.

[0005] For example, Japanese Patent Publication No. 1-47823 has a memory capable of storing image data read from a picture of a traveling printed matter at a predetermined point in time, and uses the image data read out from the memory as reference data and the remaining data of the traveling printed matter. There is disclosed a technique relating to a printed matter inspection apparatus of a system in which image data read from a pattern is used as inspection data, and the quality of printing is determined based on a comparison between the reference data and the inspection data. This technique uses a first feature extraction comparison determination unit that extracts and compares the features of the reference data and the features of the inspection data for each corresponding pixel unit to determine the quality of printing. Using a second feature extraction comparison determination unit that determines the quality of printing by extracting and comparing the feature and the feature of the inspection data as the sum of pixels in a predetermined range within the same picture, A third feature extraction / comparison / determination means for extracting / comparing the feature and the feature of the inspection data as a sum of pixels arranged on the same straight line along the running direction of the printed matter within the same picture to determine whether the printing is good or bad. Is used to comprehensively judge the judgment results of the first to third feature extraction / comparison judgment means to make a final judgment on the quality of the printed matter. Further, in this technique, a comparison unit that compares a difference between a traveling speed of the traveling printed material at a predetermined time and a traveling speed of the traveling printed material at the time of reading the inspection data with a predetermined value is used, and based on a result of the comparison, There is also provided a means for newly rewriting the inspection data read from the picture of the running printed matter as the reference data. According to the technique disclosed in Japanese Patent Publication No. 1-47823, the accuracy of quality determination when the density change such as a picture of a printed material is wide is improved, and a masking function and the like are provided to change the state of the printed material. , A stable operation can always be obtained.

In Japanese Patent Publication No. 1-20477, sample data detected by decomposing a sample pattern of a printed matter into a pixel matrix is compared one by one with sample data previously extracted from a sample printed matter and recorded in a memory. There is disclosed a technique relating to a picture inspection method for a printed matter for determining the quality of a picture. In this technique, one of the sample data and the sample data is displaced by one pixel in the horizontal direction of the printed matter, compared with the other of the two data, and the data between the two data when the two data become closest to the coincidence is compared. This technique involves correcting the sample data or the sample data in the lateral direction of the printed matter based on the position shift amount, and then comparing the sample data with the sample data. According to the technique disclosed in Japanese Patent Publication No. 1-20477, a pattern inspection can be performed with high accuracy using a relatively inexpensive apparatus even if there is a positional shift in the printed matter transport system.

Japanese Patent Application Laid-Open No. Sho 62-11152 discloses a detecting means for extracting pixel data in a state where the visual fields overlap each other in the direction perpendicular to the running direction of the printed matter and the images overlap each other. Then, the data to be inspected is fetched and compared with the reference data. This comparison is to compare each pixel and then compare the added data. According to the technique disclosed in Japanese Patent Application Laid-Open No. 62-11152, an accurate inspection can be performed without being affected by a positional shift that cannot be avoided in a printed material during traveling.

[0008]

However, conventionally,
There were various defects that could not be detected as defects in the inspected image. For example, there are various types of defects in printed matter, and density differences, sizes, shapes, and the like of defective portions vary widely. For example, a defect having a large density difference can be relatively easily detected. However, in the prior art, it is difficult to detect a defect having a relatively small density difference and a large area of the defective portion. Has been confirmed by As described above, if the image inspection is affected by the density difference, the size, the shape, and the like of the defective portion, there arises a problem that the reliability of the image inspection is reduced.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and has as its object to provide an image inspection apparatus capable of detecting a defect or the like having a small density difference.

[0010]

SUMMARY OF THE INVENTION The present invention relates to an image inspection apparatus for comparing an image to be inspected with a reference image and detecting a difference between these images, comprising an image frequency component filter, By extracting, the image hierarchical means for obtaining a plurality of output images having different characteristics, and comparing each of the plurality of output images having different characteristics with a comparison image obtained from the reference image corresponding to each of the output images. The object has been achieved by providing a comparative inspection means for obtaining an inspection result.

[0011] Further, at least the image to be inspected is a pixel image, and the image layering means sequentially performs density smoothing of a predetermined number of adjacent pixels to reduce the number of pixels and sequentially differ in the characteristics. The above object has been achieved by obtaining an output image and by using a means in which parameters used for the density smoothing are variable according to the image to be inspected.

[0012]

The present invention has been made by focusing on the characteristics of image defects which are difficult to detect with a conventional image inspection apparatus. Defects to be detected by the image inspection apparatus have various density differences, sizes, and shapes. For example, when the image to be inspected is a printed matter, there are stains on the printing paper and various printing defects that also depend on the type of printing machine that performed the printing. These print defects have different density differences, sizes and shapes of defective portions. For example, some printing failures that occur in offset lithographic rotary printing presses and the like have relatively small density differences and relatively large areas. According to a study by the inventors, it has been found that it is difficult to detect such a defect having a relatively small density difference and a relatively large area. Even if the density difference is small as described above, if the size of the defective portion is large, it becomes relatively noticeable visually, although it is difficult to detect the defect by inspection, and the quality of printed matter is high. Etc. will be reduced. The present invention has been made to provide an image inspection apparatus capable of stably detecting such a defect.

FIG. 1 is a block diagram showing the gist of the present invention.

As shown in FIG. 1, in the image inspection apparatus of the present invention, an image hierarchization unit 20 and a comparison inspection unit 2
2 is a characteristic part.

The image layering means 20 has, for example, a low-pass filter, a high-pass filter, a band-pass filter, or a filter having various frequency characteristics. Further, the image layering means 20 uses such an image frequency component filter to extract the frequency components of the image to be inspected.
Obtain output images with different characteristics. For example, in FIG.
In, a total of n output images having different frequency components and different characteristics, such as a first output image, a second output image, and an n-th output image, are obtained.

The present invention does not limit whether the image to be inspected or the reference image is a pixel image, nor does it limit the specific configuration of the image layering means 20. . For example, the image to be inspected is a pixel image, and the image layering means 20 sequentially performs density smoothing of a predetermined number of adjacent pixels, reduces the number of pixels, and sequentially outputs output images having different characteristics. In addition, the parameter used for the density smoothing may be variable depending on the image to be inspected. In this case, more effective determination can be made in accordance with the characteristics of the image to be inspected. For example, the parameters may be varied according to whether or not the image to be inspected includes a thin pattern or the like.

The comparison / inspection means 22 converts the first output image to the n-th output image obtained from the image hierarchical means 20 into
The image inspection is performed by comparing the first output image to the n-th output image corresponding to the first output image to the n-th output image. This image inspection compares the first output image with the first comparison image 24, compares the second output image with the second comparison image 24, and thus, one-to-one for each output image. And the n-th output image is compared with the n-th comparison image 24.

The first comparative image 24 to the n-th comparative image 2
Reference numeral 4 denotes an image obtained from the reference image, that is, an image serving as a reference corresponding to the image to be inspected input to the image layering means 20. Such a first comparative image 24 to an n-th comparative image 24 may be generated in advance by using the image layering means 20 as in an embodiment described later. That is, a reference image may be input to the image layering means 20, and a total of n first comparison images 24 to n-th comparison images 24 may be obtained.

In the present invention, the image hierarchization means 20
It does not limit the specific configuration of the image frequency component filter, and does not specifically limit the image frequency component filter. For example, as in an embodiment described later with reference to FIG. 10 and the like, the image layering means 20 may be a cascade-connected image layering circuit 120 having the same circuit configuration.

The present invention does not specifically limit the comparison and inspection means 22. For example, each of the first output image to the n-th output image and the first comparison image
The inspection determination method after the one-to-one comparison with each of the n-th comparison images 24 is not specifically limited. For example, in such a one-to-one comparison, if there is at least one defect, the inspection result of the comparison / inspection means 22 may be determined to be defective, or a predetermined number or more of one-to-one comparison results becomes defective. Only when this occurs, the inspection result of the inspection unit 22 may be determined to be defective. Alternatively, how to use a threshold value for performing such an inspection is not limited.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 2 is a block diagram showing a configuration of a printing machine and a printed matter centralized inspection apparatus to which the present invention is applied.

FIG. 2 shows a printed matter centralized inspection apparatus main body 80 for inspecting a plurality of printing presses including a printing press (not shown) at one place. In the printing press 10 shown in FIG. 1, the band-shaped printed matter 1 sent out from the final printing unit 12 is wound into a roll by a feed roll 14 and a winding part 16. The printing machine 10
Thereafter, a printing state signal indicating whether or not the printing press is in operation and the like, and a pulse signal according to the running of the printed matter 1 are output by a rotary encoder 18 attached to the printing unit 12. When the circumference of the plate cylinder of the printing unit included in the printing machine 10 such as the printing unit 12 is longer than one pattern to be printed (hereinafter, also referred to as one surface),
A plurality of pictures are printed by plate cylinder-rotation. In order to detect such a surface position, the rotary encoder 18 includes:
A pulse signal is output for each predetermined rotation.

In the printing press 10 shown in FIG. 2, a detecting unit 30 for taking an image of the surface state of the printed material during traveling and outputting a predetermined image signal is provided between the printing unit 12 and the feed roll 14. Are located. The repeater 40 receives the print state signal and the pulse signal from the rotary encoder 18,
0 is connected. Note that, like the three repeaters 40 shown in the figure, the other repeaters not shown are also
It is provided corresponding to another printing machine not shown. These repeaters 40 are respectively connected to the centralized inspection device main body 80.
It is connected to the judging unit 100 inside.

FIG. 3 is a side view of the detecting section of the printed matter centralized inspection apparatus as viewed from the printed matter running direction.

In FIG. 3, the printed matter 1 is traveling from the near side to the other side. The detecting section 30 mainly includes a light source 32b housed in a lamp house 32a and an optical fiber 33 for guiding light from the light source 32b to a light guide 34, and a slit provided in the light guide 34 The printed matter 1 is illuminated. A radiating fan (not shown) is built in the lamp house 32a, and the heat generated by the light source 32b is radiated. As the light source 32b, for example, a discharge light source such as a halogen lamp or a xenon lamp is used. The detection unit 30 has a plurality of line sensor cameras 36 arranged orthogonally to the running direction of the printed matter 1. The one-dimensional line-shaped portion to be photographed on the printed matter 1 photographed by each of the line sensor cameras 36 is also orthogonal to the running direction of the printed matter 1. The field of view of each of the plurality of line sensor cameras 36 overlaps the field of view of another adjacent line sensor camera 36. The electric circuit 38 obtains one one-dimensional line-shaped image signal covering substantially the entire width of the printed matter 1 from the one-dimensional line-shaped image signal from the plurality of line sensor cameras 36, Processing such as transmitting this to the repeater 40 is performed.

FIG. 4 is an electric circuit diagram of the electric circuit of the detecting section of the printed matter centralized inspection apparatus.

As shown in FIG. 4, the detection unit electric circuit 38 for outputting image signals obtained from the plurality of line sensor cameras 36 to the repeater 40 mainly includes an A / D converter. A unit 38a, a frame memory 38b,
Image compression circuit 38c, optical communication unit 38d, driver 3
8e.

The A / D converter 38a converts an image analog signal from each of the plurality of line sensor cameras 36 into an image digital signal by an A / D converter 38g. The A / D converter 38a generates a continuous one-dimensional linear digital image over substantially the entire width of the printed matter 1 by integrating the images of the adjacent line sensor cameras 36. The generated digital image is sequentially written into the frame memory 38b in synchronization with the running of the printed matter 1. This makes it possible to obtain a two-dimensional frame image scanned by the plurality of line sensor cameras 36 in synchronization with the running of the printed matter 1.

The 1 written in the frame memory 38b
The digital image for one screen (one frame) is data-compressed by the image compression circuit 38c, and the optical communication unit 38d
Is output to. The data compression performed by the image compression circuit 38c is a JPEG (joint photographic coding exp.) Which has recently been used in still image coding.
ert group) coding. The data compression performed by the image compression circuit 38c is for improving the transmission efficiency such as shortening the transmission time of the image digital signal.
Other data compression methods may be used. For example, it is possible to use a single known entropy coding, Huffman coding, run-length coding, or the like. The image digital signal from the image compression circuit 38c is optically modulated in the optical communication section 38d and transmitted to the repeater 40.

On the other hand, a timing signal transmitted from the repeater 40 and synchronized with the traveling of the printed matter 1, that is, a signal of a timing for photographing the printed matter 1 in a one-dimensional line shape while scanning by the traveling is The data is output to each of the plurality of line sensor cameras 36 by the driver 38e. Also, from the repeater 40, a timing signal for each surface corresponding to a picture to be printed is also sent,
The signal is used to detect the start of one frame when the A / D converter 38a writes a digital image to the frame memory 38b.

FIG. 5 is a block diagram showing a hardware configuration of a repeater of the printed matter centralized inspection apparatus.

As shown in FIG. 5, the repeater 40 installed close to each printing machine is mainly
CPU (central processing unit) 40a and ROM
(Read only menory) 40b and RAM (random acces)
s menory) 40c, parallel input / output circuit 40d, counter 40e, O / E (optical to electrical) converter 40f, serial converter 40g, and serial converter 4
0h and E / O (electrical to optical) converter 40
i and a bus 40j.

The CPU 40a operates according to a program previously written in the ROM 40b.
The RAM 40c, the parallel input / output circuit 40d, the serial converter 40g, and the serial converter 40h are accessed via 0j to perform predetermined processing. For example, the CPU 40a generates a timing signal to the detection unit 30 mainly in accordance with a signal from the printing press 10 according to a program pre-written in the ROM 40b, and generates an image digital signal from the detection unit 30. And the like to relay to the determination unit 100.

The RAM 40c is used not only for temporarily storing data when the CPU 40a executes a program, but also for buffering the image digital signal when relaying the image digital signal from the detection unit 30 to the determination unit 100. Also used as a memory. The parallel input / output circuit 4
0d is an input / output for each bit or a predetermined number of bits with respect to the printing press 10 and the detection unit 30.
The processing is performed in the bit width as a processing unit in the U40a and the bus 40j. The counter 40e determines an absolute traveling position of the printed matter 1 from an encoder signal transmitted from the rotary encoder 18 and relatively corresponding to the traveling position of the printed matter 1. The O / E converter 40f demodulates the light-modulated image digital signal sent from the detection unit using an optical fiber, and
Output to 0g. The serial converter 40g is provided with the O
The serial image digital signal output from the / E converter 40f is converted into parallel data having a predetermined bit width and accessible by the CPU 40a. The serial converter 4
0h converts an image digital signal of a predetermined bit width written by the CPU 40a into a serial signal, and outputs this to the E / O converter 40i. The E / O converter 40i
Performs optical modulation of the serial image digital signal sequentially transmitted from the serial converter 40h and transmits the modulated digital image signal to the determination unit 100 via an optical fiber.

FIG. 6 is a block diagram showing a functional configuration of the repeater.

As shown in FIG. 6, the repeater 4
The main functions of 0 include an image timing control unit 40m, an image signal relay unit 40p, and a memory 40n.

The image timing control unit 40m is a timing signal used by the detection unit 30, for example, the camera start signal and the read enable signal, according to a signal from the rotary encoder 18 of the printing press 10, for example, the A-phase signal. Generate At this time, the image timing control unit 40m also uses imposition information previously input by the operator from a keyboard 62 described later. The imposition information is information relating to the printing contents of the plate cylinder of the printing press 10 and, for example, information on how many pages are to be printed.

The image signal relay section 40p receives the image digital signal sent from the detection section electric circuit 38 via an optical fiber, and stores it in the memory 40n. Further, the image signal relay section 40p transmits the digital image stored in the memory 40n to the determination section 100 as an image digital signal in response to a request from the determination section 100 of the centralized inspection apparatus main body 80. Output via. The memory 40n can store a plurality of frames of digital images. The judgment unit 100 is connected to a plurality of repeaters 40 (40a to 40c).
Thus, the image signal relay unit 40p and the memory 40
By the relay processing of the relay device 40 using n, it is possible to sequentially perform the printed matter inspection of each printing machine corresponding to each relay device 40.

FIG. 7 is a block diagram showing a hardware configuration of the printed matter centralized inspection apparatus main body.

As shown in FIG. 7, the printed matter centralized inspection apparatus main body 80 mainly includes a CPU (central pr).
ocessing unit) 50, a main storage device 52, a hard disk device 54, a floppy disk device 58, an input / output device 60, an O / E converter 60a, and a keyboard 62.
CRT (cathode ray tube) controller 64a, C
It is composed of an RT 64b, a network control device 66, a printer device 68, and a system bus 70.

The CPU 50 executes program modules and the like read from the hard disk device 54 into the main storage device 52, such as those relating to printed matter inspection processing. The hard disk device 54 stores program modules, data, and the like according to the present embodiment.
The data is read out to the main storage device 52 as needed. The floppy disk device 58 is used for transferring various program modules and data to and from other computer systems. The input / output device 60 is used for connection with another digital processing device. For example, it is connected to the plurality of repeaters 40 (40a to 40c) via the plurality of O / E converters 60a, and receives an image digital signal related to a printed matter inspection of each printing press.

The keyboard 62 is used together with the CRT controller 64a and the CRT 64b when an operator operates the main body 80 of the centralized inspection apparatus.
The keyboard 62 is also used for setting various data. The CRT control device 64a and the CRT
The reference numeral 64b is used for transmitting a printed matter inspection result in the centralized inspection apparatus main body 80 and an abnormality detection based on the result to the operator.

The network control device 66 is used to connect the centralized inspection device body 80 to another computer system online. For example, data and printed matter inspection results that can be collected by the centralized inspection apparatus main body 80, further, self-diagnosis results of the centralized inspection apparatus main body 80, and devices and devices connected to the centralized inspection apparatus main body 80, for example, The repeater 40 and the detection unit 30
It is possible to centrally monitor the abnormality diagnosis result of the printer and the printing press 10 with another computer system.
The printer device 68 prints a printed matter inspection result by the centralized inspection device main body 80 and information related thereto.

The system bus 70 is connected to the CP
U50 is the main storage device 52, the hard disk device 54, the floppy disk device 58, the input / output device 60, the keyboard 62, the CRT control device 64a, the network control device 66, Used when accessing the printer device 68. The system bus 70 includes an address bus used for selecting a device connected to the system bus 70 and the like.
Has a data bus used for data transfer when accessing.

FIG. 8 is a block diagram showing the functional configuration of the printed matter centralized inspection apparatus main body.

As shown in FIG. 8, the printed matter centralized inspection apparatus main body 80 mainly includes a determination section 100, a defective image storage section 82, a failure display section 84, a data totalization section 86, a network connection section. 88. or,
Keyboard 62, CRT controller 64a, CRT 64b
And a printer 68 and the like. These determination units 1
The processes and operations performed in the 00 and the defective image storage unit 82 are as follows.
This is performed for a plurality of printing presses 10 connected to the centralized inspection apparatus main body 80. Further, data such as a printed matter inspection result is shared between the printing presses 10.

The judging section 100 demodulates the image digital signal optically modulated and transmitted from the repeater 40 by an optical fiber, and restores the data compression performed by the image compressing circuit 38c of the detecting section electric circuit 38. Then, the printed matter is inspected according to the digital image of the result.

The defective image storage unit 82 stores the judgment unit 1
If a defect in the printed matter is detected in 00, the digital image of the corresponding frame is data-compressed and stored in the hard disk device 54 together with information on the location where the defect occurred. The defect occurrence location information is information indicating a frame number at which a defect of a printed matter is detected, and a location of a defective portion in the frame.

The defective display section 84 is provided with the keyboard 6.
2, the image stored in the defective image storage section 82 and the information on the location of the occurrence of the defect are automatically displayed by the operation of the operator using the CRT control device 64a or the CRT 64b, or automatically when a print defect occurs. At this time, the defective display section 84 restores the stored digital image that has been stored and compressed.

As described above, in the present embodiment, when a print defect occurs, the image of the frame and the defect occurrence location information relating to the image are stored, so that the operator can grasp the state of the print defect at any time. And can be observed. When a printed matter is found visually by a rotary printing press, label insertion or marking has conventionally been performed, but in order to observe the defective part or defective part, the wound printed matter is unwound. There was a problem that had to be. However, according to this embodiment,
At any time, it is possible to grasp and inspect the defect of the printed matter.

The data totaling section 86 is provided with the judgment section 10
The number of print defects detected at 0 is counted and the result is stored. Further, the totaling result can be displayed on the CRT 64b or printed on the printer device 68 by the operation of the operator from the keyboard 62 or the like, and output.

The network connection unit 88 includes an abnormality such as a failure of the main body of the centralized inspection device, and a device or device connected to the main body of the centralized inspection device 80, for example, the detection unit 30, the repeater 40, or the printing device. An abnormality such as a failure of the machine 10 is detected and transmitted to another computer system online.

FIG. 9 is a block diagram showing the functional configuration of the determination unit of the printed matter centralized inspection apparatus main body.

As shown in FIG. 9, the judgment section 1
00 is mainly composed of the light receiving unit 102 and the image restoring unit 104
, A temporary image memory 106, a comparison determination unit 110, and a recording unit 118. These configurations are provided for each of the detectors 30 of the printing press 10. Alternatively, one set of these configurations is provided, and the printed matter defect determination for each detector 30 is sequentially performed.

The optical receiving section 102 demodulates an image digital signal optically modulated and transmitted from the repeater 40 via an optical fiber. The image restoration unit 104 restores the digital image data compressed by the image compression circuit 38c in the detection unit electric circuit 38.

The temporary image memory 106 temporarily stores an image for one screen (one frame) until the processing performed by the comparison determination unit 110 or the recording unit 118 is completed. The comparison / determination unit 110 compares the image to be inspected stored in the temporary image memory 106 with a reference image previously stored in the comparison / determination unit 110, and determines a difference between these images. The determination result is stored in the recording unit 11
8 is output. The recording unit 118 reads the digital image of the temporary image memory 106 corresponding to the print defect when there is an input of the print defect from the comparison determination unit 110, and reads the digital image together with various information such as a defect occurrence time and a defect occurrence location. Is temporarily stored as print defect information. At this time, the recording unit 118 also performs data compression of the digital image. Further, it outputs the various information temporarily stored in response to a request from the defective image storage unit 82 or the data totaling unit 86.

FIG. 10 is a block diagram showing the configuration of the comparison and judgment section of the judgment section.

As shown in FIG. 10, the comparing and judging section 11 of the judging section 100 described with reference to FIG.
0 has a total of four image layering circuits 120, a total of five comparison inspection circuits 130, and an inspection result determination circuit 170.

A digital image stored in the temporary image memory 106 is input to one of the image hierarchical circuits 120. Further, the output of the one image layering circuit 120 is input to one of the other image layering circuits 120, and the output of the input image layering circuit 120 is output to the next image layering circuit 120. Is input to one of them. The other image layering circuit 120 is connected in the same manner. As described above, these four image layering circuits 12 are used in total.
Numerals 0 are daisy-chain connected so that each digital image input portion and each digital image output portion are sequentially connected.

The total of four image hierarchical circuits 12
0, i.e., these image layering circuits 1
The output of each of the 20 extracted images is input to the comparison and inspection circuit 130. The digital image output from the fourth image layering circuit 120 is input to the fifth comparison inspection circuit 130.

The provisional judgment results output from each of the five comparison inspection circuits 130 are all input to the inspection result determination circuit 170.

The image layering circuit 120 has an image frequency component filter, that is, a low-pass circuit 122 and a high-pass circuit 126, which will be described later. Obtain different output images. That is, the image layering circuit 120 obtains an output of the digital image having different characteristics and an output of the extracted image. Therefore, by connecting the image layering circuit 120 as shown in FIG. 10, a total of five output images having different characteristics can be obtained. In this embodiment, a total of 4
The image layering means 120 of the present invention is constituted by the image layering circuits 120.

The image hierarchization means of the present invention is not limited to the daisy chain connection of the plurality of image diffraction circuits 120 as described above. If 1
By designing the two image layering circuits 120,
It is possible to design various image hierarchical means having different numbers of output images.

The comparison / inspection circuit 130 compares the extracted image output from the image layering circuit 120 with a reference image stored in the comparison / inspection circuit 130 corresponding to the extracted image and determines a predetermined value. Obtain the inspection results. The inspection result determination circuit 170 obtains an overall determination result based on the temporary determination results input from each of the five comparison inspection circuits 130 in total. The inspection result determination circuit 170 determines that the overall determination result is defective if at least one of the input provisional determination results is defective. Specifically, the five-input OR circuit is used. Have been.

FIG. 11 is a block diagram showing the configuration of a first example of the image hierarchical circuit of the determination section.

As shown in FIG. 11, the image layering circuit 12 used in the determination section 100
0, that is, one of the image layering circuits 120 in FIG. 10 mainly includes a low-pass circuit 122 and an image thinning circuit 1
24. The low-pass circuit 122 is a low-pass filter having a predetermined cutoff frequency FC. The low-pass circuit 122 converts the input digital image
An image spatial frequency component lower than the cutoff frequency FC is extracted and output. The image thinning circuit 124 reduces the number of pixels of the digital image (pixel image) output from the low-pass circuit 122. Note that the low-pass circuit 122 and the image thinning circuit 124 may be configured as a composite.

As described above, the first example of the image layering circuit outputs the input digital image as it is as the extracted image, and passes the digital image through the low-pass circuit 122 and the image thinning circuit 124 to output a predetermined image. Output a digital image of In this way, the first example of the image hierarchical circuit 120 has two different characteristics based on the input digital image.
Get two output images.

FIG. 12 shows the configuration of the second example of the image layering circuit 120 used in the comparison / determination section 110, that is, one of the image layering circuits 120 shown in FIG. Image layering circuit 1 shown in FIG.
Reference numeral 20 mainly includes a low-pass circuit 122, an image thinning circuit 124, and a high-pass circuit 126. The low-pass circuit 122 and the image thinning circuit 124
Is the same as that of the first example of the image layering circuit 120 having the same reference numeral.

The high-pass circuit 126 is a high-pass filter having a predetermined cutoff frequency FC, and extracts and outputs an image spatial frequency component of the input digital image that is higher than the cutoff frequency FC. The high-pass circuit 12
The output of 6 is the output of the extracted image of the image hierarchical circuit 120.

As described above, the second example of the image layering circuit of the present embodiment is such that, in addition to the low-pass circuit 122, the high-pass circuit 126 having the same cutoff frequency FC as the low-pass circuit 122. And outputs two output images having different characteristics in total, that is, an extracted image and a digital image.

The low-pass circuit 122 and the high-pass circuit 126 used in the first and second examples of the image layering circuit include a digital signal processor (DSP) and a normal CPU (central CPU). processing unit
) Or a predetermined arithmetic and logic operation circuit. Further, the image thinning circuit 124 can be similarly configured.

FIG. 13 is a diagram showing the relationship between the image hierarchy and the total number of pixels in this embodiment.

As shown in FIG. 13, in the present embodiment, the 0th-layer digital image V ₀ having a total number of pixels of 256 × 256 pixels is obtained by extracting the spatial frequency components of the low-frequency image and thinning out the total number of pixels. And sequentially, the first hierarchical digital image V ₁
4th go a hierarchical digital image V ₄ ~. These 0
The thinning out of the total number of pixels from the hierarchical digital image V ₀ to the fourth hierarchical digital image V ₄ means that the total number of pixels is sequentially reduced to ４, and four adjacent pixels are sequentially integrated. Things. Therefore, the total number of pixels of the digital image of each layer is as follows.

The total number of pixels of the 0th hierarchical digital image V ₀ : 2
56 × 256 = 65536 pixels Total number of pixels of the first hierarchical digital image V ₁ : 128 × 128
= 16384 pixels total number of pixels of the second hierarchical digital image V _2: 64 × 64 = 4
096 pixels total number of pixels of the third hierarchy digital image V _3: 32 × 32 = 1
024 pixels total number of pixels of the fourth hierarchy digital image V _4: 16 × 16 = 2
56 pixels

FIG. 14 is a diagram showing the relationship between the image hierarchy and the image spatial frequency distribution in this embodiment.

As shown in FIG. 14, FIG.
Each cut-off frequency FC ₁ ~FC ₄ of a total of four image layering circuit 120 shown in is a low frequency enough to go downward in FIG. 10. In the present embodiment, particularly when the first example of the image hierarchical circuit 120 is used,
The output of the extracted image from the image layering circuit 120 is, in order, the 0th layer digital image V ₀ , the 1st layer digital image V ₁ ,
Second hierarchical digital image V ₂ , third hierarchical digital image V ₃
And a fourth hierarchical digital image V _4. On the other hand, when the second example of the image layering circuit 120 is used in this embodiment, the outputs of the extracted images output from the total of four image layering circuits 120 are sequentially higher frequency bands. The component image O ₀ , the high frequency component image O ₁ , the high frequency component image O ₂ , the high frequency component image O _3, and the fourth hierarchical digital image V ₄ are obtained.

FIGS. 15 to 24 show the 0th hierarchical digital image V ₀ to the 4th hierarchical digital image V ₄ , the high frequency component image O _{0 to} O ₃ , and the 4th hierarchical digital image, respectively. of the second example of the image V _4, it is an image diagram illustrating an example of each image.

In other words, FIG.
Image V₀FIG. 4 is an image diagram showing an example of the above. FIG.
The high frequency component image O₀FIG.
You. FIG. 17 shows the first hierarchical digital image V₁One
It is an image figure showing an example. The 18 is the high frequency component.
Minute image O₁FIG. 4 is an image diagram showing an example of the above. FIG.
The second hierarchical digital image V_TwoFIG.
You. 20 is the high frequency component image O_TwoAn example of
FIG. FIG. 21 shows the third-layer digital
Image V_ThreeFIG. 4 is an image diagram showing an example of the above. 22 is the front
High frequency component image O_ThreeFIG. 4 is an image diagram showing an example of the above.
FIG. 23 illustrates the fourth hierarchical digital image V_FourFirst example of
FIG. FIG. 24 shows the fourth-layer digital
Tal image V _FourIt is an image figure showing the 2nd example of.

Hereinafter, the operation of the comparison / determination unit 110 using the second example of the image hierarchical circuit 120 will be described using mathematical expressions.

In FIG. 12, the n-th hierarchical digital image V _n (x, y) input to the image hierarchical circuit 120 is shown.
), A low-pass filter realized by the low-pass circuit 122 (that is, an image spatial frequency filter)
The processing using K _L (i, j) can be expressed as the following equation when the thinning processing is performed simultaneously.

V _{n + 1} (x, y) = K _L (i, j) * V _n (2x, 2y) (1)

When the thinning process is separately performed in the process using the low-pass filter K _L (i, j), first, the process represented by the following equation is performed (filter process).

V _n ′ (x, y) = K _L (i, j) * V _n (x, y) (2)

Thereafter, a process represented by the following equation is performed as a thinning process (thinning process).

V _{n + 1} (x, y) = V _n ′ (2x, 2y) (3)

Here, “*” represents a convolution operation. Incidentally, as the low-pass filter K _L , the following equation is usually used.

1/9 1/9 1/9 1/9 1/9 1/9 1/9 1/9 1/9 (4)

In the above equation (2), an example of a low-pass filter K _L of 3 × 3 pixels is shown, but a case of 5 × 5 pixels or 7 × 7 pixels may be used. Needless to say, as the number of pixels of the filter increases, the cutoff characteristics of the spatial filter can be improved.

Subsequently, a high-pass filter (image spatial frequency filter) K _H (i, j) realized in the high-pass circuit 126 is used. When component image O _n, processing performed by the high-pass circuit 126 can be expressed by the following equation.

[0091] _{O n (x, y) =} K H (i, j) * V n (x, y) ... (5)

The high-pass filter K _H (x, y
) Can be used, for example, as shown in the following equation for 3 × 3 pixels.

-1-1-1-1 -18-1-1-1-1 ... (6)

The high-pass filter K _H (i, j) shown in the equation (4) is a filter known as Laplacian in eight directions.

FIG. 25 is a block diagram showing a configuration of the comparison and inspection circuit of the comparison and determination section.

FIG. 25 shows one configuration of a total of five comparison / inspection circuits 130 of the comparison / judgment unit 110 shown in FIG. This figure 2
As shown in FIG. 5, the comparison and inspection circuit 130 used in the present embodiment mainly includes a switch 132, a reference image memory 134, an image to be inspected memory 136, and a threshold determination unit 1.
38, a threshold memory 140, and a comparison / inspection means 142.

The switching unit 132 switches the extracted image output from the image layering circuit 120 to the reference image memory 134 or the inspected image memory 136. When the contacts a and c are turned on, the switch 132 stores the extracted image in the inspection image memory 136.
Is written to. On the other hand, when the contacts b and c are turned on, the extracted image is written into the reference image memory 134.

In the image inspection apparatus of the present embodiment, when starting a predetermined image inspection, first, the switch 132 is switched to the contact point b, so that a good digital image is first written into the reference image memory 134. During the subsequent image inspection, the switching unit is switched to the contact point a, and sequentially transmitted from the detection unit 30 and the repeater 40, and the image restoration unit 104 and the temporary image memory 106, if necessary. The extracted images input through some of the image layering circuits 120 in response are written into the image memory 136 to be inspected.

The threshold value determining section 138 determines a predetermined threshold value T (x
, Y). The obtained threshold value T (x, y) is
It is stored in the threshold memory 140. The threshold memory 14
0 has a storage capacity capable of storing a plurality of thresholds T (x, y) of the same number as the total number of pixels of the extracted image input to the comparison inspection circuit 130.

When the threshold value determining section 138 and the threshold value memory 140 shown in FIG. 26 are used, the comparison and inspection means 142 generates the inspection result G (x, y
) Is determined for each pixel of the extracted image.

G (x, y) = NG: | S (x, y) -I (x, y) |> T (x, y) GOOD: | S (x, y) -I (x, y) | ≦ T (x, y) (7)

Here, (x, y) is the position of each pixel of the extracted image, the reference image stored in the reference image memory 134, or the image to be inspected stored in the image to be inspected memory 136. And S (x
, Y) are each pixel of the reference image, and I (x, y)
Is each pixel of the image to be inspected. "NG" indicates that the inspection result is defective, and "GOOD" indicates that the inspection result is good.

As described above, in this embodiment, the inspection of the image to be inspected is performed for each pixel, and the permissible threshold at this time is provided for each pixel. If there is a displacement between the reference image and the image to be inspected, the comparison inspection means 142
It is impossible to perform an accurate inspection. In particular, in a portion of the reference image or the image of the image to be inspected where the density change is remarkable, even if the displacement is less than one pixel, the adverse effect is extremely large. However, in the present embodiment, the threshold value of the portion where the density change is drastic can be automatically increased to a large value, and it is possible to reduce erroneous inspection.

FIG. 26 is a block diagram showing the configuration of a first example of the threshold value determining section of the comparison test circuit.

FIG. 26 shows the configuration of the threshold value determination unit 138 of FIG.

The threshold value determining section 138 mainly comprises a maximum value calculation circuit 150, a minimum value calculation circuit 152, a difference circuit 154, and an arithmetic circuit 156.
In the threshold value determination unit 138, the maximum displacement of the reference image and the image to be inspected in the horizontal and vertical directions,
That is, it is provided in order to allow the maximum positional deviation amount MX in the horizontal direction (X-axis direction) and the maximum positional deviation amount MY in the vertical direction (Y-axis direction), noise that can be expected in the input system up to the comparison inspection means, and the like. The allowable value NMIN and the threshold T
(X, y) is required.

T (x, y) = Max [NMIN, Max {S (x + i, y + j)} − Min {S (x + i, y + j)}] (8)

Here, Max (A, B...) Is A, B
... is a function that returns the maximum value of Min. (A, B
..) Is a function that returns the minimum value of A, B,. Also, i is a variable that changes from -MX to MX, and j is a variable that changes from -MY to MY. or,
In the above equation (6), “Max ｛S (x + i, y + j
)｝ ”Is performed by the maximum value calculation circuit 150, and finds the maximum density value in the range of i from −MX to MX and j in the range of −MY to MY. “Min {S (x + i, y + j)}” is an operation performed in the minimum value calculation circuit 152, and i is −MX
To MX, and j determines the minimum density value in the range of -MY to MY. Therefore, the above (6)
The expression is based on the difference between the maximum density value and the minimum density value when a displacement within the allowable range occurs, taking into account noise and the like expected in the input system, and the allowable threshold value T (x
, Y).

In the above equation (6), when the maximum displacement is smaller than one pixel both in the horizontal and vertical directions, the following equation can be used.

T (x, y) = Max (NMIN, [Max {S (x + i, y + j)} − Min {S (x + i, y + j)}] × r%) (9)

Here, i and j are in the range of ± 1, and r
Is a numerical value from 0% to 100%.

FIG. 27 is a block diagram showing the configuration of a second example of the threshold value determining section of the comparison test circuit.

The threshold value determining section 138 shown in FIG.
Is mainly composed of a maximum value calculation circuit 160, a minimum value calculation circuit 162, and a calculation unit 164. In the threshold value determination unit 138, the upper limit allowable threshold value T _H (x, y)
And a lower limit threshold T _L (x, y) are obtained, and these limit thresholds are respectively written to the upper threshold memory 140a or the lower threshold memory 140b in the threshold memory 140.

The threshold value determining section 138 performs an operation as shown in the following equation.

[0115] _{T H (x, y) =} Min [MXV, Max {S (x + i, y + j)} + NMIN] ... (10)

T _L (x, y) = Max [0, Min {S (x + i, y + j)} − NMIN] (11)

Here, the MXV is determined by the threshold value determination unit 138.
And the maximum value that can be expressed by the threshold memory 140 or the like. When the maximum displacement is less than one pixel, the equations (8) and (9) can be modified in the same manner as the equation (7).

When the threshold value determination section 138 is of the second example described later, the comparison inspection means 142 obtains the inspection result G (x, y) by the following equation.

G (x, y) = OVER: I (x, y)> _TH (x, y) UNDER: I (x, y) < _TL (x, y) GOOD: otherwise (12)

As described above, the threshold value determining section 138
In the case of using the example, it is possible to more finely determine the inspection failure in the comparison inspection means 142. That is, it is possible to discriminate whether the defect is equal to or larger than the allowable range or equal to or smaller than the allowable range. When the second example of the threshold value determination unit 138 is used,
The storage capacity of the threshold memory 140 is 2 compared to the first example.
However, since the reference image memory 134 is not used at the time of inspection, the reference image memory 134 may be used. Therefore, when the second example is used, it is possible to distinguish whether the defect is due to lack of color or addition of color,
It is possible to reduce the problem of light source unevenness and to speed up the inspection process by not performing the absolute value process.

[0121]

As described above, according to the present invention, an excellent effect of being able to provide an image inspection apparatus capable of inspecting even a defect having a small density difference can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the gist of the first invention of the present application.

FIG. 2 is a block diagram illustrating a configuration of a printing machine and a printed matter centralized inspection apparatus to which the present invention is applied.

FIG. 3 is a side view of a detection unit of the printed matter centralized inspection device as viewed from a printed matter running direction.

FIG. 4 is a block diagram of a detection unit electric circuit of the detection unit.

FIG. 5 is a block diagram illustrating a hardware configuration of a repeater of the printed matter centralized inspection device.

FIG. 6 is a block diagram showing a functional configuration of the repeater;

FIG. 7 is a block diagram showing a hardware configuration of a main body of the printed matter centralized inspection apparatus.

FIG. 8 is a block diagram showing a functional configuration of the printed matter centralized inspection apparatus main body.

FIG. 9 is a block diagram illustrating a functional configuration of a determination unit of the printed matter centralized inspection apparatus main body.

FIG. 10 is a block diagram showing a configuration of a comparison / determination unit of the determination unit;

FIG. 11 is a block diagram showing a configuration of a first example of an image hierarchical circuit of the comparison / determination unit;

FIG. 12 is a block diagram showing a configuration of a second example of the image hierarchical circuit of the comparison / determination unit;

FIG. 13 is a diagram showing the relationship between image hierarchies and the total number of pixels in the embodiment.

FIG. 14 is a diagram showing a relationship between image hierarchization and spatial frequency distribution in the embodiment.

FIG. 15 is an image diagram showing an example of a 0th-layer digital image targeted by the embodiment.

FIG. 16 is an image diagram showing an example of a high-frequency component image of the 0th hierarchy targeted by the embodiment.

FIG. 17 is an image diagram showing an example of a first hierarchical digital image targeted by the embodiment.

FIG. 18 is an image diagram showing an example of a high-frequency component image of the first hierarchy targeted by the embodiment.

FIG. 19 is an image diagram showing an example of an image to be inspected in the second hierarchy targeted by the embodiment;

FIG. 20 is an image diagram showing an example of a high-frequency component image of the second hierarchy targeted by the embodiment.

FIG. 21 is an image diagram showing an example of a third hierarchical digital image targeted by the embodiment;

FIG. 22 is an image diagram showing an example of a high-frequency component image of the third hierarchy targeted by the embodiment.

FIG. 23 is an image diagram showing a first example of a fourth hierarchical digital image targeted by the embodiment;

FIG. 24 is an image diagram showing a second example of the fourth hierarchical digital image targeted by the embodiment.

FIG. 25 is a block diagram showing a configuration of a comparison test circuit of the comparison determination unit.

FIG. 26 is a block diagram showing a configuration of a first example of a threshold value determination unit of the comparison inspection circuit;

FIG. 27 is a block diagram showing a configuration of a second example of the threshold value determination unit of the comparison test circuit;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Printed matter 10 ... Printer 12 ... Printing unit 14 ... Feed roll 16 ... Wound upper part 18 ... Rotary encoder 20 ... Image hierarchical means 22 ... Comparative inspection means 24 ... Comparative image 30 ... Detector 32a ... Lamp house 32b ... Light source 33 ... optical fiber 34 ... light guide 34a ... slit 34b ... casing 34c ... glass rod 34d ... paint 36 ... line sensor camera 38 ... detector electric circuit 38a ... A / D converter 38b ... frame memory 38c ... image compression circuit 38d ... light Communication unit 38e Driver 38g A / D converter 40 Repeater 40a CPU 40b ROM 40c RAM 40d Parallel input / output circuit 40e Counter 40f O / E converter 40g, 40h Serial converter 40i E / O converter 40j ... bus 40m ... image timing control unit 40n ... Memory 40p image signal relay unit 50 CPU 52 main storage device 54 hard disk device 58 floppy disk device 60 input / output device 60a O / E converter 62 keyboard 64a CRT control device 64b CRT 66 network Control device 68 Printer device 80 Printed matter centralized inspection device main body 82 Defective image storage unit 84 Defective display unit 86 Data aggregation unit 88 Network connection unit 100 Judgment unit 102 Optical receiving unit 104 Image restoration unit 106 Temporary image memory 110 ... Comparative determination unit 118 ... Recording unit 120 ... Image hierarchical circuit 122 ... Low-pass circuit 124 ... Image thinning circuit 126 ... High-pass circuit 130 ... Comparative inspection circuit 132 ... Switching unit 134 ... Reference image memory 136: Image memory to be inspected 138: Threshold determination unit 140: Threshold memory 14 ... comparison inspection unit 150 ... maximum value calculation circuit 152 ... minimum value calculating circuit 154 ... difference circuit 156 ... arithmetic circuit 160 ... maximum value calculation circuit 162 ... minimum value calculating circuit 164 ... arithmetic unit

Continuation of front page (72) Inventor Hiroshi Sato 1-1-1 Ichigaya-Kaga-cho, Shinjuku-ku, Tokyo Inside Dai Nippon Printing Co., Ltd. (56) References JP-A-60-205681 (JP, A) JP-A-1 JP-A-293485 (JP, A) JP-A-63-173465 (JP, A) JP-A-62-89175 (JP, A) JP-A-63-293686 (JP, A) (58) Fields investigated (Int. . ^7, DB name) G06T 1/00 280 - 340 G06T 5/20 G06T 7/00 G07D 7/00 patent file (PATOLIS) JICST file (JOIS)

Claims (2)

(57) [Claims] An image inspection apparatus for comparing an image to be inspected with a reference image and detecting a difference between the images has an image frequency component filter and extracts a plurality of characteristics by extracting frequency components of the image to be inspected. Image hierarchical means for obtaining different output images, and comparison inspection means for obtaining an inspection result by comparing each of the plurality of output images having different characteristics with a comparison image obtained from the reference image corresponding to each of the output images. An image inspection apparatus comprising: 2. The method according to claim 1, wherein at least the image to be inspected is a pixel image, and the image hierarchical means sequentially performs density smoothing of a predetermined number of adjacent pixels to reduce the number of pixels. An image inspection apparatus, wherein output images having different characteristics are sequentially obtained, and a parameter used for the density smoothing is variable according to the image to be inspected.