JPS63198469A - Image automatic inspecting instrument - Google Patents

Abstract

PURPOSE:To check a dot or a highly sophisticated image which a person is impossible to discriminate by setting a square opening whose sides are in specific length as a window to detect the optical density of an object to be inspected. CONSTITUTION:For instance, the opening in square shape the length of whose sides is 10mum is prepared. And even one piece of toner particle 211 can be detected if present. The shape of the opening can also be rectangle or it can be in a size slightly larger than a toner particle. For instance, a square the length of whose side is 20mum is acceptable as the opening for the particle 211. The smaller the size of the opening is, the easier the detection of a toner particle come, however, if it is too small, the quantity of light that passes through it comes also too smaller, and the S/N is degraded. Accordingly, the opening is made shaped in a rectangle whose short side is about equal to or slightly shorter than the diameter of a coloring unit (for instance a toner) and the long side is about equal to or ten times as long as said diameter.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic image inspection device for inspecting the quality of images reproduced by, for example, a facsimile machine, a printing machine, or a copying machine.

"Prior Art" In offices, various information devices output graphic information such as text and images. A typical example of this is a copying machine that copies original documents. A copying machine forms an electrostatic latent image on a photosensitive drum, reads image information using an image sensor such as a CCD, and develops the image using a developing device, or uses a recording head such as a thermal head. Images are reproduced on paper.

When designing such information devices or shipping these information devices from factories, reproduced images are inspected. There are two types of such tests:

(i) Inspection of whether the information equipment operates normally according to predetermined procedures and reproduces images.

(ii) Checking whether the quality of the reproduced image is within the market acceptable range or the specifications established when the equipment was designed.

For example, in the case of a copying machine, the inspection items include the position of the image relative to the copied paper, the density of the image relative to the original, and the resolution. The inspector inspects using scales, magnifying lenses, measuring instruments, or visually to check whether each process of the copying machine is working properly and that there are no errors in image reading or toner image transfer position. Determine whether or not.

In the case of copying machines, the latter inspection is also performed by an inspector. That is, the quality of the image is determined by the inspector directly comparing the image copied on the paper with the sample.

In conventional inspections as described above, the inspector is the main person, so
There were the following problems.

(i>If the examiner was different, the measured values or test results changed.

(11) Even for the same tester, the measured values or test results may change due to familiarity with the test or the psychological influence of the test subject tested previously.

(iii) Measured values or test results also changed due to physical or mental fatigue of the examiner.

In order to avoid such drawbacks, an image inspection device that automatically performs inspection has been proposed (Japanese Patent Application Laid-Open No. 1034-1989).
No. 5 and Japanese Patent Application Laid-open No. 10346/1983).

This image inspection apparatus is provided with a table on which an object to be inspected having an image is positioned and placed. The object to be inspected is set on this table, and the detection section is placed opposite to it. Then, the detection data output from this detection section is supplied to a data processing section, and the position, density, and resolution of the image are digitized based on the detection data.

However, this proposed image inspection apparatus uses a line sensor (image sensor) or a combination of a low density meter and a high density meter in the detection section. For this reason, the minimum area that can be measured has to be much larger than the minimum area that humans can sense. Therefore, with this image inspection device, the optical density in uniform density areas and the resolution in extremely high quality image areas are
Although it is possible to obtain somewhat good measurement results, image examinations under other conditions may result in a large discrepancy between the results and those perceived by humans.

The cause of this will be explained next.

FIG. 18 is an enlarged view of a part of the chart for resolution inspection. In this way, in this chart, black lines 201 with different intervals and line widths are drawn in parallel, and the resolution of the copied image is determined by how far the line width can be distinguished from the white background part 202. It is designed to be inspected.

FIG. 19 shows a further enlarged part of this chart, and FIG. 20 shows a sample of images copied by a copying machine in correspondence with this.

Here, FIG. 20A is an example in which relatively large unevenness occurs in the boundary area between the ground color portion 202 and the black line 201, and FIG.
This is an example of toner scattering at these boundary areas. Further, C in the same figure is an example in which a blank area 203 to which no toner is attached is generated inside the black line 201. In addition, various other situations may occur, such as the density of the black line 201 not changing all at once at the boundary but changing stepwise, or shading occurring inside the black line 201. Such delicate image conditions are a relatively large factor in image quality evaluation.

By the way, FIG. 21 shows the signal level of the read image signal when the outline of the black line has unevenness, for example, as shown in FIG. 20A. This image signal 205
is a signal obtained by horizontally scanning the chart shown in FIG.
This corresponds to FIG. 4 of Publication No. 5. A signal portion 205' shown by a broken line in this figure is an image signal obtained by scanning the protruding portion of the black line 201 in FIG. 20A, and has a thicker waveform than other portions shown by a solid line. However, in the diagram, the dashed line is 20? The image signal 205 at the threshold level indicated by
．． If 205' is binarized and the image is inspected, the evaluation of the resolution will be exactly the same for both.

This is because in conventional devices, the resolution is determined based on the location where a signal change occurs due to binarization and the number of times the signal changes at that location.

FIG. 22 shows an image signal for the image state shown in FIG. 20, and FIG. 23 shows an example of an image signal for the image state shown in FIG. 20C. In the example shown in FIG. 22, the level of the image signal 205 becomes high in a portion 208 where the scattered toner is scanned. However, since the scattered portion is a relatively small area, the signal level does not increase sufficiently in this portion and is ignored in the binarization process.

FIG. 23 shows the opposite case, where the signal level at the arrow 2090 is lowered due to the blank area 203 existing at the black line 201. However, even in this case,
Since the signal change in the minute portion is not sufficient, this change is ignored in the binarization process. As described above, the conventional apparatus has a problem in that images are discriminated at a level different from the quality of images perceived by the human eye.

Therefore, automatic image inspection apparatuses have been proposed that can perform an inspection that is almost equivalent to what a person who inspects an image perceives visually. This is described in Japanese Patent Application No. 142976/1988 filed on June 20, 1986.

In this patent application No. 61-142976, for example, as shown in FIG. 24, the optical density of an object to be inspected is inspected using an aperture of 10 .mu.m.times.500 .mu.m. As a result, images can be inspected for some inspection items of the object to be inspected at a level approximately equivalent to human sensation.

``Problems to be Solved by the Invention'' By the way, at the design or research stage of copying machines, etc., it is necessary to perform image evaluation with a precision higher than that of human vision.

However, conventionally proposed automatic image inspection apparatuses have limitations in such highly accurate image evaluation, and have been unable to perform sufficient inspections. Furthermore, with this previously proposed device, it was possible to perform relatively good measurements such as inspection of optical density over a relatively wide range and reproducibility of certain straight lines, but it was not possible to measure "points". If it is something that humans can barely discern, it may be ignored in the measurement, making it impossible to measure the optical density.

This will be explained using FIGS. 24 to 28.

As already explained with reference to FIG. 24, it is assumed that image inspection is performed using an aperture of 10 μm×500 IJm. In this case, as shown in FIG. 25, colored units 21 having a diameter slightly larger than 10 μm,
Even if only one 1 is present in the aperture, the level of the read signal will be low, as shown in FIG. Here, the coloring unit 211 refers to the smallest unit constituting the image to be inspected, and for example, in an electronic copying machine, one toner particle corresponds to this. In devices employing other recording methods, the minimum unit of coloring is the coloring unit 211 here. In this specification, for ease of understanding, unless otherwise specified, the coloring units 211 will be described as toner particles 211 as an example.

Toner particles 2 are placed well within the 10 μm x 500 μm opening.
Even if 11 were captured, the resulting signal level would be extremely low. Therefore, as shown in FIG. 26, when this signal is binarized at the threshold level TH,
The presence of toner particles 211 cannot be detected. That is, the presence of toner particles 211 will be ignored.

On the other hand, as shown in FIG. 27, when the toner particles 211 are linearly connected over a certain length, the area occupied in the opening increases. As a result, in the case shown in FIG. 27, the signal level reaches M,t below the threshold level TH, and its presence is detected as shown in FIG. As described above, although conventional devices have reached the human inspection level, they are only effective for lines or objects with a certain amount of area, and there is a problem in distinguishing "points."

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an automatic image inspection apparatus that can mainly inspect "points" effectively.

"Means for solving the problem" In the present invention, the short side is approximately equal to or shorter than the diameter of the coloring unit (for example, toner particles), and the length is approximately equal to or within 10 times the diameter of the coloring unit (for example, toner particles). A rectangular opening formed by the long sides of is set as a window for detecting the optical density of the object to be inspected.

For example, as shown in FIG. 29, a square opening with one side of 10 μm is prepared. In this way, the presence of even one toner particle 211 can be detected as shown in FIG. 30.

The shape of the opening may be rectangular or may be slightly larger than the size of the colored toner particles. For example, as an opening for the toner particles 211,
As shown in FIG. 31, it is also possible to have a square shape with one side of 20 μm.

This is because, as shown in FIG. 32, if a toner particle 211 just enters into this opening, its presence can be detected. However, if the size of the opening is increased in this way, the toner particles 211 may move away from the center of the opening, making it impossible to detect their presence.

In such a case, the pitch of window movement may be finely set to less than the length of one side in the main scanning direction or the sub-scanning direction. FIG. 33 shows an example of this, in which toner particles 211 are captured by setting the window movement pitch to half of one side.

Note that the smaller the aperture size, the easier it is to detect toner particles or colored units, but if the aperture is too small, the amount of light that passes through it will decrease, and the level of noise in the image signal will increase. The S/N ratio deteriorates. Also, the time to scan the object to be inspected becomes relatively long.
It's not a good idea.

Therefore, in the present invention, the size of the opening is determined from a short side that is approximately equal to or shorter than the diameter of the coloring unit (for example, a toner particle), and a long side that is approximately equal to this diameter or within 10 times the diameter. It is assumed to be a rectangle.

According to the present invention, it is possible to obtain image data that is higher than the visual resolution, and by processing this, it becomes possible to perform "point" inspection and advanced image inspection that cannot be discriminated by humans.

"Example" The present invention will be described in detail below with reference to Examples.

Overview of the Apparatus FIG. 1 shows the appearance of an automatic image inspection apparatus according to an embodiment of the present invention. This automatic image inspection apparatus is composed of an inspection section 1, a computer section 2, and a printer section 3.

Of these, the inspection section 1 is a section that continuously inspects copy paper 4 as an object to be inspected. The inspection section 1 includes a supply tray 5 and a discharge tray 6.

When inspecting a copying machine, a desired chart is set in the copying machine, and copy sheets 4 obtained thereby are stacked on a supply tray 5 as shown.

The copy sheets 4 are fed one by one into a cylindrical chart holder 8 by a feed roller ruff. The surface of the chart holder 8 is covered with an insulating film, and the copy paper 4 is electrostatically attracted to this surface by a charging operation by an electrostatic charge supply device (not shown). In this state, an image inspection of the copy paper 4 as the object to be inspected is performed.

The copy paper 4 that has been inspected is peeled off from the chart holder 8 by a peeling mechanism that will be described later. The peeled copy sheets 4 are sequentially discharged onto the discharge tray 6.

An operation display panel 9 is arranged in this inspection section 1.
Disposed here are a power switch 11, a movement key 12 used when manually specifying the pattern of the object to be inspected, and a display 13 for displaying concentration data as a measurement result.

The computer section 2 can be configured by a commercially available computer, and performs identification of inspection items, processing of data such as concentration data, and various displays. This part includes a keyboard 15 as an input means and a CRT 16 as a display means.
, a disk drive device 17 for driving a floppy disk, etc., and a CPU (central processing unit) for data processing is installed inside.

The printer section 3 is a section that outputs test results and the like, and in this embodiment a dot printer is used.

FIG. 2 shows an outline of the inspection section of this automatic image inspection apparatus. A shaft 21 for rotating the feed roller 7 of the inspection section 1 receives driving force from a feed roller drive motor 23 via a chain 22. The supply tray 5 is adapted to be given a force to move upward by the excitation of a solenoid (not shown), and at the time of this excitation, the top layer surface of the copy paper 4 as the object to be inspected comes into contact with the feed roller 21. When the feed roller 21 rotates by a predetermined amount in this state, only one sheet of copy paper 4 in the uppermost layer is fed out. Prior to this feeding, the surface of the chart holder 8 is uniformly charged by a charging mechanism (not shown). As a result, the fed-out copy paper 4 is electrostatically attracted to the chart holding section 8. The rotation of the cylindrical chart holder 8 in the circumferential direction (Y-axis direction) is performed by the driving force of a chart holder drive motor 26 connected to a speed reducer 25 .

In this embodiment, the outer diameter of the chart holder 8 is 162 mm in diameter.
77 mm, the step angle of the chart holder drive motor 26 is 1.8 degrees, and the reduction ratio of the reducer 25 is 1/256.
And so. As a result, the chart holder drive motor 26 is driven one step, and the surface of the chart holder 8 is moved by 10 μm in the Y-axis direction. Control of the rotational position of the chart holder 8, that is, control of the position in the Y-axis direction, is performed using a point detected by a photo sensor 28 in a notch 27 provided at the end of the cylinder as a reference point.

A ball screw 32 rotated by an X-axis stepping motor 31 is arranged above the chart holder 8 so that its axis is parallel to the rotation axis of the cylindrical chart holder 8. The Y-axis direction movement hole 34 of the optical head mounting block 33 is threadedly engaged with the ball screw 32. Therefore, when the X-axis stepping motor 31 rotates, the ball moves! It is guided by two guide bars 35 and 36 arranged parallel to I=-32 to move in the X-axis direction.

In this embodiment, the pitch of the ball screw 32 is 5 mm. The step angle of the X-axis stepping motor 31 is set to 0.
With the configuration set at 72 degrees, the optical head mounting block 33 moves by 10 μm in the X-axis direction in one step of driving. Two limit switches 37 and 38 are arranged in the X-axis direction to limit the movement range of the optical head mounting block 33.

A density detection section 41, which will be described next, is attached to the optical head attachment block 33. A magnifying eyepiece lens 42 is also attached to the density detection unit 41, and the position of the image captured by the objective lens 43 can be checked during focus adjustment and especially during manual operation.

In the automatic image inspection apparatus of this embodiment, the density detection section 41 is moved at a pitch of 10 μm in both the X-axis direction and the Y-axis direction, but it may be set at a finer pitch than this. In this case, for example, the pitch of the ball screw 32 in the X-axis direction and the reduction ratio in the Y-axis direction may be made finer.

FIG. 3 shows the optical structure of the optical head.

The concentration detection section 41 includes a tungsten lamp 51 for illumination. The light emitted from the tungsten lamp 51 is focused by the illumination lens 52 and
The measurement site 53 is illuminated. The reflected light from the measurement site 53 is collected by an objective lens 43 and split into two directions by a prism 54 equipped with a semi-transparent mirror (beam splitter).

One of the branched lights is reflected by a mirror 55, passes through a measurement field of view adjustment mechanism 56, has its wavelength component corrected by a color correction filter 57, and enters a photomultiplier tube 58. Here, the measurement field of view adjustment mechanism 56 has an aperture plate 59 and a field lens 61 disposed in the optical path.

The aperture plate 59 is a plate provided with a rectangular opening as shown in FIG. In this 50 μm x 50 μm opening area, an image of the measurement site on the copy paper is magnified five times and formed.

And on the chart, that is, in this example, copy paper 4
Square area with all four sides of the top 10 μm (Figure 4)
The light beam reflected from the photomultiplier tube 58 passes through this opening and enters the photomultiplier tube 58 described above. The direction of the opening of the aperture plate 59 can be set to an arbitrary angle by an aperture plate rotation step motor 62.

The other light branched by the prism 54 has its traveling direction changed by the roof-shaped prism 64, and forms an erect image on the observation screen 65.

The image of the measurement site 53 thus formed can be enlarged and observed using the magnifying eyepiece 42.

Circuit Configuration of the Device (Principle Configuration of the Device) Before specifically explaining the device, the fundamental configuration of the circuit will be explained.

The following FIG. 5 shows an outline of the circuit configuration of the automatic image inspection apparatus. This device is equipped with external signal input means 72 for instructing desired inspection items. The measurement control means 73 is configured to set the position and type of the pattern to be inspected and the inspection processing procedure according to the inspection item indicated by the external signal manual means 72. The pattern information storage means 74 stores the pattern to be inspected within the object to be inspected, and the processing procedure storage means 75 stores the inspection processing procedure for the pattern to be inspected.

The measurement means 76 scans the surface of the object to be inspected under the control of the measurement control means 73 and detects the image density. The arithmetic processing means 78 performs arithmetic processing on the data obtained from the measuring means 76 according to a processing procedure instructed by the measurement control means 73. The test results obtained thereby are outputted by the output means 83. The output means 83 is typically the printer section 3 shown in FIG. 1, but the test results can also be displayed on the CRT screen of the computer section 2.

The operation of this automatic image inspection device will be explained in more detail. In the automatic image inspection apparatus, the type of the object to be inspected and the inspection items are coded by the external signal manual means 72 during inspection. A chart code 84 for identifying the chart copied to the object to be inspected and an inspection item code 85 representing the inspection item are sent to the measurement control means 73. The measurement control means 73 sends the chart code 84 to the pattern information storage means 74. Pattern information storage means 7
4 is a pattern code 86 representing a pattern to be inspected included in the chart represented by the chart code 84, and a representative point position 8 representing a representative position of this pattern to be inspected.
Outputs 7. Of these, the pattern code 86 is sent to the processing procedure storage means 75 together with the inspection item code 85,
The image density detection format 88 corresponding to the inspection item and pattern to be inspected and the arithmetic processing code 89 representing the arithmetic processing procedure are read into the measurement control means 73.

At this stage, ■ the type of pattern to be inspected required for inspection, ■
Positional information on where the pattern exists on the copy paper, information on (1) an image density detection method for that pattern, and (2) information about a method for calculating and processing the results corresponding to the inspection items are stored in a code in the measurement control means 73. It will be set in a formatted state.

Among these pieces of information, the representative point position 87 representing the coordinates of the position where the pattern exists and the image density detection format 8
8 is sent to measuring means 76.

The measuring means 76 moves to the representative point position 87 instructed by the measurement controlling means 73, and reads the image density detection format 8.
8, the image density to be measured is detected. The detection result is outputted to the arithmetic processing means 78 as a concentration data string 91. At the end of the concentration data string 91, an end signal 9
2 is added to instruct the start of arithmetic processing.

Upon receiving the end signal 92, the arithmetic processing means 78 executes the arithmetic processing code 89 previously supplied from the measurement control means 73.
Based on this, the corresponding calculation processing routine is selected. This arithmetic processing routine is then loaded into the internal arithmetic processing routine memory area. In the arithmetic processing means 78, the density data string 91 described above is stored in a density data string memory area. The arithmetic processing means 78 processes this concentration data string 91 using the routine loaded into the arithmetic processing routine memory area, and outputs the results according to the test items as the test results 9.
3 and is supplied to the output means 83.

The output means 83 will output this content.

The image density detection and density data arithmetic processing operations described above are sequentially performed for all patterns to be inspected that are preset in the measurement control means 73.

The arithmetic processing means 78 not only performs arithmetic processing on each pattern, but also performs statistical processing of the arithmetic processing results corresponding to all set patterns to be inspected. In this way, the desired test results for the object to be tested will be obtained.

(Configuration of External Signal Input Means) Next, the configuration of the external signal input means will be explained using FIG. 6.

The external signal input means 72 includes encoding means 101 . Chart salt 102 and test items 103 input by the operator are encoded by this encoding means 101. When the code type discrimination means 104 receives the coded information, it separates it into a chart code and a test item code. Then, it is output as a chart code 84 and a test item code 85 via the code control unit 105.

(Structure of Pattern Information Storage Means) FIG. 7 shows the structure of the pattern information storage means. The pattern information storage means 74 supplies the chart code 84 to the pattern information storage position retrieval means 107. The pattern information storage position search means 107 searches for the position of the pattern to be inspected, and searches the pattern information storage unit 1.
08 as a pointer 109.

FIG. 8 shows the contents of the pattern information storage section. The pattern information storage unit 108 stores as data the pattern codes of all the objects to be inspected in the corresponding chart using the chart code as a key, and the representative point positions of the respective patterns represented by these pattern codes. . In this figure, for example, for chart code “xxx”, there are three pattern codes a
1a and 1b are available. This means that two types of patterns specified by the pattern code aSb are displayed on the chart specified by this chart code "xxx", and the coordinates of the three patterns are the representative point positions. It is as shown in.

Here, the pattern represented by the pattern code a is, for example, the electrophotographic society test chart "No. 1-
This is a resolution measurement pattern (not shown) for "R1975".This pattern is placed in the upper left and lower right parts of this test chart.In addition, pattern code b
The pattern represented by is a pattern for density measurement in this electrophotographic society test chart.

In this test chart, various density samples are displayed in a row at the bottom, forming a pattern for density measurement.

The pattern information output means 110 includes the pattern information storage section 1
08 is read out from the position indicated by the pointer 109 output by the pattern information storage position search means 107. The read content is the pointer 109
All pattern codes and their representative point positions for one chart code indicated by . Pattern code 86 and representative point position 870 for it
The combinations are sequentially read out under the control of the measurement control means 73 shown in FIG. 5 and sent into the measurement control means 73.

(Configuration of processing procedure storage means) Next, the contents of the processing procedure storage means 75 are shown in FIG.

The processing procedure storage means 75 is supplied with an inspection item code 85 and a pattern code 86. Among these, the inspection item code 85 is detected by the inspection item code detection means 11.
2, and the pattern code 86 is detected by the pattern detection means 113. The detection result of the test item code detection means 112 is outputted as a first pointer 114 to the processing code storage means 115, and the detection result of the pattern detection means 113 is outputted as the second pointer 115 to the processing code storage means 116.

FIG. 10 shows the contents of the processing code storage means. The processing code storage means 116 stores (i) an arithmetic processing code, (ii) a pattern code, and (iii) an image density detection code for each inspection item code. The first pointer 114 output from the above-mentioned test item code detection means 112 specifies the test item code for specifying the test item. Then, the second pointer 115 output from the pattern detection means 113 selects the arithmetic processing code in that inspection item code. In the example shown in FIG. 10, the pattern code “a
It can be seen that for the pattern specified by ``, image density detection specified by image density detection code A'' and calculation processing specified by calculation processing code A'' are performed.The code specified by the two pointers 114 and 115 is performed. The contents are temporarily stored in a storage area within the processing code storage means 116.

Returning to FIG. 9, the explanation will be continued. Inspection procedure search means 11
7 reads out the image density detection code 118 stored in the processing code storage means 116.

In the example of FIG. 10 described above, the image density detection code 118
is “i”. Based on this, the third pointer 119 as address information is stored in the inspection procedure storage means 121.
Output for.

FIG. 11 shows the contents of the inspection procedure storage means. The inspection procedure storage means 121 stores image density detection formats for each image density detection code. There are multiple sets of image density detection formats, each of which is stored in block units. The contents of these block units are, for example, (i) measurement start position, (ii) direction,
(iii) interval, (1v) total number of points, and (V) slit direction.

Here, (i) the measurement start position is shown as a relative representative point. As described above, the representative point is indicated by the coordinates that serve as a reference for each pattern, whereas the counter representative point is the difference between the coordinate value of the starting position and the coordinate value of the representative point when scanning the pattern.

(ii) The direction is the direction of scanning, and there are two types of directions: the X-axis direction and the Y-axis direction. (iii) Interval is the sampling interval for concentration detection, and (iv)
The total score is the total number of sampled data. (
V) The slit direction refers to the direction of the opening of the aperture plate 59 shown in FIG. In this example, the opening is set to
It is set either parallel to the axis or rotated 90 degrees from it. Further, this opening can be set at an arbitrary position during measurement. Especially if the opening is rectangular,
By setting this angle, it is possible to effectively measure diagonal lines and the like.

The third pointer 119 specifies the image density detection code. In the example shown in Figure 11, the image density detection code
is selected. The first code output means 122 reads out the image density detection format 88 selected by the third pointer 119, and the measurement control means 73 shown in FIG.
is supplied to the measuring means 76 under the control of. On the other hand, the second code output means 123 is processed by the processing code storage means 116.
The arithmetic processing code 89 is read out from the arithmetic processing means 78 and similarly supplied to the arithmetic processing means 78 under the control of the measurement control means 73.

(Configuration of measurement means) The following FIG. 12 shows the contents of the measuring means.

The measuring means 76 can be roughly divided into three parts: (i) an image density detection section, (ii) a detection aperture control section, and (iii) a moving section. The measuring means 76 moves over the object to be inspected (in this embodiment, the copy paper 4 on which the chart has been copied) based on the image density detection format 88 supplied from the measurement control means 73, and generates an image in a predetermined format. The concentration will be detected. That is, the image density detection format 88 (see FIG. 11) supplied from the measurement control means 73 is supplied to the data sampling control section 131, and based on the format 88 decoded here, the image density detection section and the detection aperture control section ,
The moving part will be controlled accordingly.

By the way, the data sampling control section 131 controls the light receiving means 1 based on the data 133 obtained from the drive control section 132.
The current location of 33 is known. Therefore, the data sampling control section 131 determines the amount by which the light receiving means 133 should be moved by comparing it with the measurement start position obtained from the image density detection format 88. Data 134 regarding the determined movement amount etc. is sent to the drive control section 132.

The drive control unit 132 determines the amount of movement in the X-axis direction and the amount of movement in the Y-axis direction based on the data 134, and outputs an X-axis direction drive signal 135 and a Y-axis direction drive signal 136 with the corresponding number of pulses. The X-axis direction drive signal 135 is supplied to the X-axis stepping motor 31, and the Y-axis direction drive signal 136 is supplied to the chart holder drive motor 26 (both shown in FIG. 2) which also serves as a stepping motor.

As already explained, the concentration detection section 41 (FIG. 2) is moved in the X-axis direction by the X-axis stepping motor 31.

Further, the drum-shaped chart holder 8 is rotated in the Y-axis direction by the drive of the chart holder drive motor 26, and the light receiving means 133 consisting of the photomultiplier tube 58 and the like is moved to a desired measurement position.

The data sampling control unit 131 then decodes the image density detection direction, sampling interval, total number of sampling points, and slit direction. First, the aperture direction is compared with the current aperture direction, and an angle signal 138 representing the comparison result with the designated angle is output. Angle signal 138
is supplied to an angle signal generator 139. The angle signal generator 139 supplies a control signal 141 to the aperture plate rotation step motor 62 (see FIG. 3) to rotate the aperture plate 61 by a desired angle.

When the setting of the light receiving means 133 is completed as described above,
The data sampling control section 131 sets the total number of points obtained from the image density detection format 88 in a counter (not shown) within the control section. Then, the image density detection direction and the sampling interval are outputted to the drive control section 132 as data 134 and set.

The drive control section 132 moves the concentration detection section 41 or the chart holding section 8 by a predetermined amount in accordance with the instructed detection direction.

By the way, the detection output 14 output from the light receiving means 133
3 is amplified by an amplifier 144 in the image density detection section, and its output 145 is logarithmically converted by a logarithmic converter 146. Conversion output 147 is provided to A/D converter 148. A/D
The converter 148 is supplied with an A/D signal 151 for specifying the time at which A/D conversion is performed by the A/D signal generator 149. A/D signal signal generation section 1
49-A/A supplied from the data sampling control section 131
The A/D signal 151 is generated by the D start signal 152, and the output of a counter (not shown) in the data sampling control section 131 is used as the A/D start signal 152.

That is, this counter is preset with a count value corresponding to the measurement start position, and the light receiving means 1
As 33 begins to move, the count increases. When the count value of the counter reaches the preset value, the A/D
A start signal 152 will be output. When the A/D conversion is completed, the A/D signal generation section 149 completion signal 153
Output. When the data sampling control section 131 receives the end signal 153, it manages the above-mentioned counter and instructs the drive control section 132 to move the light receiving means 133, and if necessary, starts the next A/D at a predetermined timing. A D start signal 152 will be output. In this way,
It becomes possible to manage the sampling interval of concentration data.

On the other hand, when A/D conversion is instructed by the A/D signal 151, the A/D converter 148 converts the conversion output 147 from analog to digital. The density data 154 after conversion is
The images are sequentially stored in the image density buffer 155. The stored concentration data 154 is supplied to the arithmetic processing means 78 as a concentration data string 91, and is subjected to arithmetic processing.

Now, when the sampling of the concentration data progresses and the built-in counter counts a certain value as the final value, the data sampling control section 131 outputs an end signal 156 to the measurement control means 73.

When the measurement control means 73 receives this end signal 156,
Data regarding the next block is supplied to the data sampling control section 131 as an image density detection format 88. In this way, concentration data is collected for each part to be measured.

(Configuration of arithmetic processing means) FIG. 13 shows the configuration of the arithmetic processing means. The calculation processing means 78 includes a calculation control section 161.

The calculation control unit 161 is supplied with the calculation processing code 89 from the measurement control means 73 .

The arithmetic processing code 89 is a code representing the arithmetic processing procedure. The arithmetic control unit 161 decodes the arithmetic processing code 89 and supplies the result as an arithmetic processing code 162 to the arithmetic processing address search means 163.

The arithmetic processing address search means 163 uses this arithmetic processing code 162 to search the arithmetic processing storage section 164.

The arithmetic processing storage unit 164 stores arithmetic processing data groups necessary for various tests. When the arithmetic processing address search means 163 moves the pointer 165 to the start address of the corresponding memory area by searching, the arithmetic processing address search means 163
A termination signal 166 is sent to 61. After receiving the end signal 166, the arithmetic control unit 161 generates a start signal 167,
Supply to loader starter 168.

When loader starter 168 receives activation signal 167, it generates load signal 169.171.

(i) Pointer 1 in the arithmetic processing storage unit 164
A series of arithmetic processing contents 172 indicated by 65 and (11) the density data string 91 stored in the image density buffer 155 (see FIG. 12) of the measuring means are transferred to the working area 17.
Load into 4. After loading, loader starter 16
8 supplies a start signal 175 to the working area 174, activates it, and transfers control to the working area 174 itself.

Along with this, the working area 174 contains the concentration data string 91.
The desired arithmetic processing is performed on the data. The result is stored in the calculation result output buffer 177 as the calculation result 176. The output means 83 shown in FIG. 5 inputs this stored content as an inspection result 93 and visualizes it.

1] 11 weeks] Error 1 history In this automatic image inspection apparatus, first, the copy paper 4 is held in the chart holding section 8, and after its positioning, density measurement for each pattern is performed.

Therefore, next we will explain positioning with respect to the copy paper 4,
Next, the density measurement work for each pattern will be explained.

(Positioning for each measurement site) In order to measure an image from the copy paper on which the chart has been copied, the objective lens 43 of the optical head must correctly capture the target measurement site. By the way, if the density detection unit 41 side is unilaterally set to a predetermined coordinate position, the desired position on the copy paper 4 is ±5
An error of about mm occurs. This is due to the following reasons.

(i) Misalignment that occurs when a chart is copied using a copying machine.

In addition to alignment errors (registration misalignment) between the copy paper 4 and the photosensitive drum of the copying machine and magnification setting errors, there are also Contains errors due to skew (rotation).

(ii) Misalignment when setting the chart in the chart holder 8.

This is a deviation that occurs when the copy paper 4 sent out from the supply tray 5 is set in the chart holding unit 8, and is caused by an error in the setting of the supply tray 5 or a misalignment in position alignment when the copy paper 4 is sent out. do.

In this embodiment, the target position can be reached with an accuracy of ±9.5 mm. For this purpose, a chart 191 used in an automatic image inspection device, as shown in FIG.
For example, there are position detection patterns 192 to 1 at those three locations.
94 was placed. These position detection patterns 192~
The coordinates of 194 are grasped by the chart type on the automatic image inspection device side. These patterns on the chart 1
The coordinate values from 92 to 194 are called actual coordinate values, and these are (L, Yt, Xa, Yz, Xa,
Y3).

The device calculates these real coordinate values (X+, Yt, Xa
. These position detection patterns 192 to 194 on the copy paper 4
The coordinate value of (X+ + y+, Xa * '72
, Xa + y3). Surface coordinate system (X, Y),
By assembling the relational expression (x, y), the coordinates (xo,
The actual position (Xo, Yo) corresponding to the coordinate value (Xo, Yo) is calculated, and this coordinate value (Xo, Yo) is used to cause the density detection unit 41 to reach the target image area.

The position detection pattern does not need to be placed in three places on the copy paper; for example, by placing it in two places,
It is possible to understand the rotation and positional shift of the Further, by arranging more points, it is possible to grasp the elongation of each part of the copy paper 4, and it is possible to further improve the accuracy of reaching the measurement site.

(Pattern Scanning) FIG. 15 shows a certain pattern for density measurement. In this pattern 221, a point 223 is a representative point, and a point 224 is a detection start point in the X-axis direction. The detection starting point 224 is expressed as a relative coordinate value with respect to the representative point 223.

When scanning this pattern 221 also in the Y-axis direction,
A detection starting point different from point 224 may be provided for this purpose.

As described above, in the automatic image inspection apparatus of this embodiment, a rectangular area of 10 μm×10 μm on a copy sheet is set as a reading range at one time. The reading mode is determined by the image density detection format shown in FIG. That is, in this example, the point 224 is the measurement start position, and the measurement direction is the X-axis direction. The measurement interval, ie, the sampling width, is determined depending on the purpose of measurement and the like. In the automatic image inspection apparatus of this embodiment, the aperture is a rectangular (square) area of 10 μm x IQ μm, so the density detection area for one time in the X-axis direction is 10 μm. Therefore, when scanning the copy paper 4 all over in the X-axis direction, the measurement interval is 10μ.
m.

FIG. 16 shows the scanning situation in this case. By setting the sampling width smaller than the visual resolution in this way, it is possible to extract data representing minute image conditions as shown in Figure 20, and by processing this data, it is possible to understand the human sense of density and resolution. almost identical to
Or it becomes possible to make a determination that goes beyond this. The measurement does not necessarily require sampling at the same width as the short side or the non-long side of the rectangular area where the optical density is to be detected, and can be freely set in the image density detection format.

Therefore, depending on the inspection item, the image may be scanned roughly.
Alternatively, it is also possible to scan finely. For example, an example of a case where an image is roughly scanned at a pitch longer than the short side is when the density of a relatively wide area is to be determined. An example of fine scanning is scanning at a pitch of 5 μm as shown in FIG. This is useful when analyzing It is also important to perform such detailed inspections on images that are enlarged and projected using an overhead projector.

The same applies when scanning in the Y-axis direction as in the X-axis direction. As a result, in the case of this embodiment, it is possible to sample an image with a width of 10 μm.

As already explained, in this embodiment, the concentration detection section 41 is moved in the X-axis direction or the Y-axis direction in units of 10 μm by stepping the X-axis stepping motor 31 or the chart holding line drive motor 26 one pulse at a time. be able to.

As described above, since the automatic image inspection apparatus of this embodiment measures the object to be inspected using the rectangular area as the minimum measurement range, the object to be inspected can be efficiently scanned without any gaps. Of course, the rectangular area is not limited to a square, and may be a rectangle. In this case, it is important that the short side of this rectangle is approximately the same as or slightly shorter than the toner particles after being fixed on the recording paper.

Also, the basic area to be measured does not have to be a strict rectangle;
For example, it may be a rectangle that is close to a circle or an oval. However, in these cases, in order to scan the entire image, some of the images are read overlappingly, so it is necessary to process these overlapping parts.

Although a photomultiplier tube is used as the light receiving means in the embodiment, the same optical density measurement operation can be performed using a photomultiplier using a semiconductor or a one-dimensional sensor such as a CCD. Further, in the embodiment, the optical density was detected using reflected light, but it may be detected using transmitted light, for example, when inspecting the development state of photographic film.

Further, in the embodiment, the colored unit as the smallest unit constituting the image to be inspected has been explained using toner particles as an example. For other non-impact type devices or impact type devices, the size of the opening, the rotation angle, etc. may be considered in the corresponding color unit.

"Effects of the Invention" As explained above, according to the present invention, the short side is approximately equal to or shorter than the diameter of each colored unit as the smallest unit constituting the image to be inspected on the object to be inspected. , and long sides having a length approximately equal to or within 10 times the diameter was set as a window for detecting the optical density of the object to be inspected. As a result, extremely accurate image measurement data can be obtained, image inspection accuracy is improved, and work that relies on humans can be almost completely automated.

[Brief explanation of the drawing]

1 to 16 are for explaining one embodiment of the present invention, of which FIG. 1 is a perspective view of an automatic image inspection device;
Fig. 2 is a schematic configuration diagram showing the main parts of the inspection section, Fig. 3 is an explanatory layout diagram showing the arrangement of optical components of the optical head, and Fig. 4 shows the size of the area that is the measurement unit on copy paper. 5 is a block diagram showing the outline of the circuit configuration of the automatic image inspection device, FIG. 6 is a block diagram showing the configuration of external signal input means, and FIG. 7 is a block diagram showing the configuration of pattern information storage means. , FIG. 8 is an explanatory diagram showing the structure of the pattern information storage section, FIG. 9 is a block diagram showing the structure of the processing procedure storage means, FIG. 10 is a block diagram showing the structure of the processing code storage means, and FIG. 11 is a block diagram showing the structure of the processing code storage means. FIG. 12 is a block diagram showing the structure of the test procedure storage means, FIG. 12 is a block diagram showing the structure of the measuring means, FIG. 13 is a block diagram showing the structure of the arithmetic processing means, and FIG. 14 is a block diagram showing the arrangement of position detection patterns. FIG. 15 is a plan view of the chart shown, FIG. 15 is a plan view showing an example of a pattern for density measurement, FIG. 16 is an explanatory diagram showing an example of scanning in the X-axis direction, and FIG. 17 is shown in FIG. FIG. 18 is an explanatory diagram showing an example of finer scanning than the conventional scanning; FIG. 18 is a plan view showing an enlarged part of a chart for resolution inspection;
19 is a plan view further enlarging a portion of the chart shown in FIG. 18, FIG. 20~C is a partially enlarged plan view showing various states of the sample of the copy image, and FIG. A
22 is a waveform diagram showing the signal level of the image signal obtained by reading the image portion shown in FIG. A waveform diagram showing the signal level of the image signal obtained by reading the image portion shown in Figure C; Figure 24 is an explanatory diagram showing an example of the size of the aperture used in the conventionally proposed device;
The figure is an explanatory diagram showing the case where one toner particle is captured by the aperture shown in this proposal, and Fig. 26 is a signal level showing how the image signal is binarized in the case shown in Fig. 25. The characteristic diagram, Figure 27 is an explanatory diagram showing the case where a part of the straight line is captured by the opening shown in Figure 24, and Figure 28rllJ is this second
A signal level characteristic diagram showing how the image signal is binarized in the case shown in FIG. 7, FIG. 29 is an explanatory diagram showing the size of the aperture in the embodiment of the present invention, and FIG. FIG. 31 is a plan view showing an example of an opening with a slightly larger size;
FIG. 32 is an explanatory diagram showing a state in which toner particles are read at a pitch of 20 μm, and FIG. 33 is an explanatory diagram showing a state in which toner particles are captured at a pitch of half of one side. 1...Inspection section, 2...Computer section, 3...Printer section, 4...Copy paper, 26...Chart holding section drive motor, 31
...X-axis stepping motor, 41...
・Concentration detection unit, 58...Photomultiplier tube, 59...Aperture plate, 62...Aperture plate rotation step motor, 21
1...Toner particles.

Claims (1)

[Scope of Claims] 1. A holding means for holding an object to be inspected, and an optical density measuring means for measuring the optical density at a specified position of the object to be inspected; A short side approximately equal to or shorter than the diameter of each colored unit as the smallest unit constituting the image to be inspected in the inspection object, and a length approximately equal to or within 10 times the diameter. An automatic image inspection device characterized in that a rectangular opening formed by long sides is set as a window for detecting the optical density of an object to be inspected. 2. A patent characterized in that the optical density measuring means is a means for detecting the optical density of the object to be inspected at each position while moving the window at a pitch equal to or shorter than the short side of the opening. An automatic image inspection device according to claim 1. 3. The automatic image inspection device according to claim 2, wherein the opening has a square shape with each side of 10 μm. 4. The automatic image inspection device according to claim 2, wherein the opening has a short side of 10 μm, a long side that is longer than the short side, and a length that is within 10 times the short side. .