JP2000094655A - Method and apparatus for detecting defect of ink-jet nozzle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect troubles of an ink-jet nozzle such as clogging, ink splashing, displacement of ink droplets and color irregularity with a high accuracy. SOLUTION: A test pattern 15 is printed on a test recording medium 3 by a printing head. A group 18 comprising a row of ink particles 16 in the pattern is photographed by a sensor means for judging normality or abnormality of the ink ejection by finding the average area, rim end position, and center interval of the ink particles by image process. Moreover, with an RGB filter mounted on a light receiving means of the sensor means, the image is taken in while switching the filter according to the ink color.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for detecting a defect of an ink jet nozzle for detecting a problem such as clogging of a nozzle of an ink jet printer for printing on a cloth, ink jetting, and displacement of ink particles. About.

[0002]

2. Description of the Related Art Recent ink jet printers for printing on fabrics are capable of expressing delicate and subtle patterns. For this reason, the diameter of the nozzles constituting the print head has become increasingly finer, and as a result, problems such as clogging of the nozzles, flying of ink, or misalignment of ink particles have been occurring.

Heretofore, when such a problem as nozzle clogging has occurred, the operator has only been able to judge it from the pattern being printed or to find it out only by the final inspection. However, discovering from the pattern being printed requires considerable experience and is burdensome. In addition, when the final inspection is performed, there is a high possibility that a large number of defective products are generated.

In order to solve these problems, for example, Japanese Patent No. 2708439 discloses a method in which a print head (nozzle head) is provided with an optical reading means for a test pattern so as to be out of the range of the original recording paper. There is disclosed a technique of detecting a non-ejection of ink by executing a test pattern on a test recording paper provided in the vicinity thereof and comparing the test pattern with a reference pattern stored in advance. Further, Japanese Patent No. 2712168 discloses that all ink droplets ejected at one time at the time of a test are imaged by a camera, the position of the center of gravity is calculated for each ink particle, and the position of the center of gravity is defined as a predetermined reference center of gravity. Disclosed is a technique for calculating a displacement in the X and Y directions by comparing the displacement with the position, obtaining a flying angle of the ink particle from the displacement, and comparing the flying angle with a reference flying angle to evaluate the quality of the ink jet. Have been. Japanese Patent Application Laid-Open No. 6-198886 discloses a technique for determining non-ejection and non-uniform density of ink from a density signal of an image obtained by capturing a test pattern.

[0005]

However, in the comparison of the printing patterns as disclosed in Japanese Patent No. 2708439, it is difficult to accurately determine even the very small positional deviation of each ink particle with respect to the reference pattern. This is very difficult because the thresholds can vary widely. On the other hand, Patent No. 27
In Japanese Patent No. 12168, the position shift is determined by comparing the position of the center of gravity of each ink particle with the reference position of the center of gravity. However, in such an absolute comparison, a detection error due to a physical defect such as a shift of the recording paper occurs. , The detection accuracy is deteriorated. A problem of JP-A-6-198866 is that it is particularly difficult to determine light-colored color unevenness. Further, according to the technique disclosed in this publication, even the positional deviation of the ink particles is not detected, but there is a problem that it is more difficult to detect the positional deviation of the light-colored ink particles.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and can detect defects such as clogging of ink jets, ink jumping, displacement of ink particles, and color unevenness with high accuracy. It is an object to provide a method and an apparatus thereof.

[0007]

According to a first method of the present invention, an image of a test pattern of ink droplets printed on a test recording medium by a print head is received, and the image is processed. In the method for detecting normal or abnormal ink ejection of the test pattern, a plurality of rows of ink particles in the test pattern are imaged as one group, and the shape of the ink particles in the group is grasped. An average area is obtained by obtaining individual areas, and a step of determining whether ink ejection is normal or abnormal by comparing the area of each ink particle with a predetermined ratio of the area to the average area as a determination value, A method for detecting a defect of an ink jet nozzle, wherein the step is performed for all nozzles of the print head in a group unit.

With this configuration, it is possible to detect whether or not there is a non-ejection in the ink droplets present in the group. That is, the shapes of a plurality of ink droplets in one group are grasped, an average area is determined by calculating individual areas, and a predetermined ratio of the area to the average area is compared with the area of each ink particle as a determination value. If there is a set number of ink particles equal to or larger than the determination value, it is determined that the ink is ejected normally. If there is at least one ink particle equal to or smaller than the determination value, the nozzle is determined to be abnormal. That is, clogging of the nozzle can be detected. This inspection is performed on a group basis for all the nozzles of the print head, and is a relative comparison within the group. Therefore, the inspection method is simple and sufficiently high accuracy is obtained. In general, the ejection failure of ink droplets occurs in the direction of decreasing the amount of ink, and occurs only in a small proportion of all nozzles, for example, a small number such as one to two. Therefore, the above-described determination by the relative comparison is sufficient.

According to a second method of the present invention, a test pattern of ink droplets printed on a test recording medium is imaged by a print head, and the image is processed to detect the normal or abnormal ink ejection of the test pattern. In the method, a plurality of rows of ink particles in the test pattern are imaged as one group, the shape of the ink particles in the group is grasped, and the X direction of each ink particle (the direction orthogonal to the ink particle row) is determined. That is, a step of determining the edge position of the print head in the main scanning direction) and comparing the difference between the maximum value and the minimum value of the edge position in the X direction with a determination value to determine the displacement of the ink particles in the X direction. And a step of performing the above steps for all the nozzles of the print head in a group unit.

In the second method, the displacement of the ink particles in the group in the X direction can be detected by relative comparison.
An X-direction edge position is determined for each ink particle, a difference between the maximum value and the minimum value is determined, and the difference is compared with a determination value. As a result of the comparison, if there is a value larger than the determination value, it can be determined that the displacement in the X direction is large (therefore, the disorder of the row of ink particles of the group is large).

According to a third method of the present invention, a test pattern of ink droplets printed on a test recording medium is imaged by a print head, and the image is processed to detect the normal or abnormal ink ejection of the test pattern. In the method, a plurality of rows of ink particles in the test pattern are imaged as one group, the shape of the ink particles in the group is grasped, and the Y direction edge position of each ink particle is determined. The center position of the ink particle is determined, the center distance between two adjacent ink particles in the Y direction (row direction of the ink particles) is determined from the center position, and the difference between the maximum value and the minimum value of the center distance in the Y direction is determined. Determining a displacement of the ink particles in the Y direction by comparing the values with the values, wherein the step is performed for all nozzles of the print head in a group unit. It is a poor method for detecting ink jet nozzles.

In the third method, the displacement of the ink particles in the group in the Y direction can be detected by relative comparison.
The displacement of the ink particles in the Y direction means the width of the center distance between the ink particles. Since the center position of an ink particle is determined as the edge position in the Y direction of the ink particle, that is, half of the sum of the Y coordinate values of the upper and lower ends, the center interval between two adjacent ink particles is obtained, and the maximum and minimum values of those two values are determined. The difference is obtained and the difference is compared with a judgment value. As a result of the comparison, if there is a value greater than or equal to the determination value, it can be determined that the positional deviation in the Y direction is large, that is, the center interval is too wide, or in other words, there is one that is extremely narrow. By detecting the displacement of the ink particles in the X and Y directions by the above second and third methods, it is possible to detect whether or not there is a so-called ink fly defect. In addition, the first to third
By comprehensively judging the results obtained by the above method, the presence or absence of color unevenness can be judged.

In the first to third methods, when an image of ink particles in the group is taken in, the image is taken in at an optimum contrast while switching the RGB filters of the light receiving section according to the ink color of the group. I have. Since the ink jet printer for fabric also uses a very light color, the light is received with an optimum contrast while appropriately switching the RGB filters. For this reason, it is easy to detect ink particles of a very light color.

An ink jet nozzle failure detecting apparatus according to the present invention for performing each of the above methods detects a test recording medium driving unit and a test pattern of ink droplets printed on the test recording medium. A light source; and a light receiving unit that collects reflected light from one group consisting of a row of ink particles in the test pattern and receives the reflected light through a slit unit and a filter unit. Scanning means for scanning the sensor means in a direction orthogonal to the ink particle row of the one group; and image processing means for processing an output signal from the light receiving means.

Further, the filter means and the light receiving means comprise a light receiving element array having RGB filters arranged in three rows.

[0016]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows the outline of the apparatus of the present invention. In the figure, reference numeral 1 denotes a sensor, the details of which will be described later. Reference numeral 2 denotes a driving unit for the test recording medium 3.
A test pattern is printed on the test recording medium 3 by the print head 4. Note that a dedicated recording paper is used for the test recording medium 3 in order to increase the detection accuracy. Needless to say, a fabric actually used may be used, but recording paper with less ink bleeding is more suitable. The recording medium 3 includes a delivery roll 5
Rubber rolls 7 and 8 and a plurality of guide rolls 9 for controlling the feed amount of the recording paper and preventing slippage are provided between the paper roll and the take-up roll 6, and are wound so as not to be loosened. . In addition, the print position of the test pattern 10
And their detection positions 11 are blocks 12 and 1 respectively.
The recording medium 3 is provided so as to be perpendicular to the central axes of the print head 4 and the sensor means 1, respectively. Reference numeral 14 denotes a sensor for detecting the presence or absence of the recording medium 3.

After the printing of the test pattern is completed, the pattern is moved and positioned immediately below the sensor means 1 by winding up the recording medium 3. After performing the origin search, the sensor unit 1 scans in the X direction and captures an image in units of groups of ink droplets in the test pattern. Although the scanning means of the sensor means 1 is not shown, it is a commonly used drive mechanism including a pulse motor and a ball screw. In the figure, Y indicates the moving direction of the recording medium 3 at the detection position 11.

FIG. 2 is a diagram schematically showing an example of a test pattern according to the present invention. The test pattern 15 is formed by printing a large number of ink particles 16 in a predetermined arrangement on the test recording medium 3 by the print head 4. In this example, the print head 4 has 32 heads 17 (a portion surrounded by a dotted line is one head), and each head 17 has 8 × 2 rows = 16 nozzles (not shown). ). Therefore, this print head 4 has a total of 5
It has 12 nozzles. Further, each head 17 prints in a different hue. In the present invention, each head 1
In FIG. 7, an array of eight vertically arranged ink droplets 16 is imaged as one group 18 and the presence or absence of nozzle clogging is inspected. The range L1 of one group 18 is about 9 mm, and the imaging range L2 is about 13 mm.

The sensor means 1 of the present apparatus scans in the X direction shown in the figure. The test recording medium 3 moves in the Y direction. Therefore, the first stage is scanned in the X direction for each group 18 for inspection, then the test recording medium 3 is moved in the Y direction, and then the second stage is similarly scanned in the X direction for inspection. Thereafter, the inspection is similarly performed up to the last fourth stage. In this manner, nozzle clogging and the like can be detected by the following method and the sensor unit 1 for all the nozzles of the print head 4.

First, the sensor means 1 has a configuration as shown in FIG. A light source 19 composed of a U-shaped fluorescent lamp,
It comprises a condenser lens 21 and a slit 22 provided in a box 20, and a light receiving element group 24 with a filter 23 mounted on the box 20, and these are integrally movable in the X direction. . Here, as shown in FIG. 4, the light receiving element group 24 includes light receiving element arrays 24a, 24b, and 24c arranged in three rows, and each array includes a CCD having 2592 photodiodes per line. It is provided with a configuration. Further, each of the arrays includes a red filter 23R, a green filter 23G, and a blue filter 23 as RGB filters.
B is attached. In the figure, 25 is a light receiving element group 24
Mounting plate.

FIG. 5 is a side view showing the detection principle of the present invention.
FIG. 6 is a front view thereof. Light rays emitted from the fluorescent lamp of the light source 19 are respectively reflected by the ink droplets 16 which are ink-applied portions on the recording medium 3 and the white paper portions 3a which are non-applied portions.
, And passes through the slit 22, and further, the reflected light 26 a from the ink particles 16 is cut by the filter 23, or only a very small amount of the light passes through. The reflected light 26b from the blank portion 3a passes through the filter 23,
After reaching the CD 27, the output voltage of the CCD 27 increases in the blank portion 3a. Therefore, as shown in FIG. 7 or FIG. 13, the output voltage of the CCD 27 changes at the boundary point of the shape of the ink droplet 16.

(1) First Method The first method of the present invention is carried out as follows. Grasping the shape of ink droplets First, each ink droplet 16 existing in the group 18
Is grasped by image binarization processing. Although an image processing unit is not shown, a known device is connected to the sensor unit 1 by a cable, and performs processing based on an output signal from the light receiving unit 24. CC as described above
Since the output voltage of D27 changes at the boundary point between the shapes of the ink droplets 16, the threshold value c is set for the voltage of the CCD 27 having the characteristics shown in FIG. As shown in FIG.
Scanning is performed in the X direction at m intervals. In this way, the shape of each ink particle 16 can be grasped.

Calculation of Area and Average Area of Ink Particles After grasping the shape of the ink particles 16, each area Si
(I = 1, 2,... N) and their average area Save. First, the area Si of the ink particle 16 is obtained as follows. The size of the ink particle 16 is about 80 to 100 μm, and its area Si is sampled at intervals of δ as shown in FIG. 9 and the dot widths W0, W1, W2,
... It is obtained by adding Wn. That is, it is calculated by: Si = W0 + W1 + W2 +... + Wn. Since the sampling interval δ is as small as 5 μm, the above equation is integrated. Since the amount of light in the dot portion falls, the dot width is smaller than the threshold value of the CCD.
Indicates the number of output pixels. Next, the average area Save of the ink particles 16 is obtained by dividing the total sum of Si by the number of the ink particles 16 existing in one group 18. That is, Save = ΣSi / N where N is the number of ink particles, where N = 8.

Setting of Judgment Value In order to set an allowable range for the judgment, α% of the average area Save obtained in the above is set as a judgment value SA. Therefore, the determination value SA is: SA = Save × α / 100. α is about 80%.

Judgment With respect to this judgment value SA, the individual ink particles 16 obtained above are determined.
Are compared. As a result, if there is a set number of ink particles equal to or larger than the determination value SA, the ink is OK (normal ink ejection).
(Ink ejection abnormality) is determined. This makes it possible to detect the non-ejection of the ink droplets 16 existing in the group 18 or a state close thereto, that is, the clogging of the nozzles by a relative comparison. In addition, this makes it possible to identify the nozzle that has clogged, and inform the operator during the operation of which nozzle has a failure.

(2) Second Method Next, a second method of the present invention will be described with reference to FIG. One edge 16 is determined from the shape of each ink particle 16 determined above.
The positions of a, 16b,... (for example, the start end position) are known. That is, when the shape of the ink particle 16 is grasped as described above, the position where the dot width of λ μm or more first appears is stored, so that the start end position of the ink particle 16 can be determined. Then, the start end position of each ink particle 16 is searched, and the difference ΔX between the maximum value and the minimum value is searched.
Ask for. Setting of judgment value The judgment value is β μm. Judgment The difference ΔX between the maximum value and the minimum value is compared with a judgment value β, and ΔX
If it is within β, it is determined that OK (no position shift), and if it is more than NG (position shift). This allows group 18
The presence / absence of a displacement in the X direction of the ink particles 16 existing in the inside can be detected by relative comparison.

(3) Third Method A third method of the present invention will be described with reference to FIG. The edge 16 in the Y direction is calculated from the shape of each ink particle 16 determined above.
f, 16g,... That is, when grasping the shape of the ink particle 16 as described above,
By storing the position where the above dot width appears,
The upper and lower end positions of the ink droplet 16 can be determined. Then, the upper and lower end positions of each ink particle 16 are searched, and the center of gravity (center) position can be calculated by calculating half of the sum of the Y coordinate values of the upper and lower ends. Thus, the center distance d between two adjacent ink particles 16 is obtained. The difference ΔY between the maximum value dmax and the minimum value dmin is determined. Setting of judgment value The judgment value is γμm. Judgment The difference ΔY between the maximum value and the minimum value of the center interval is compared with a judgment value γ. If ΔY is equal to or more than γ, it is judged OK (no displacement), and if less than NG, it is judged NG (position displacement). This makes it possible to detect the displacement of the ink particles 16 existing in the group 18 in the Y direction, that is, the width of the center interval, by relative comparison.

(4) Effect of RGB Filter Since the ink colors of the print heads 4 are different from each other, FIG.
As shown in FIG. 2, the digital signal SG from the CPU unit 28
The image data is read by switching the RGB filters based on No.4. That is, the red filter 23R is applied to the blue ink, the green filter 23G is applied to the yellow ink, and the blue filter 23B is applied to the red ink to obtain a stable output. I'm trying to get. The color of the ink is stored in advance as data. In particular, in the case of a light-colored ink, in which light-dark contrast is hardly produced, so-called overprinting is performed to increase the absolute density and increase the sensitivity. The result of reading under the optimum contrast by switching the filter in this manner is converted into an analog signal SG5 by the sensor means 1, and then converted to the CPU section 2 via a cable.
8 and determines whether the nozzle is clogged, has a poor flight, or is misaligned in accordance with the above-described process. As a result of the determination, if a defect such as clogging of the nozzle is recognized, the abnormality is notified to the inkjet printer 29 via the digital signal SG1. Further, since the nozzle where the trouble has occurred can be specified, the operator can know the abnormality of the nozzle.

FIG. 13A is a diagram showing the effect of the filter for yellow ink, and FIG. 13B is a diagram showing the effect of the filter for blue ink. As described above, the difference between the output voltages when the filter is provided is more remarkable than when the filter is not provided, so that the detection accuracy can be improved.

[0030]

As described above, according to the present invention,
Relative comparison of the area based on the average area in groups consisting of a plurality of rows of ink particles in the test pattern,
Relative comparison of the X direction displacement based on the difference between the maximum value and the minimum value of the edge position of the ink particles, and relative comparison of the Y direction displacement based on the difference between the maximum value and the minimum value of the center distance of the ink particles,
Since the determination is made as to whether the ink ejection is normal or abnormal, it is possible to detect defects such as nozzle clogging, ink jump, displacement, and color unevenness with a simple method with high accuracy.
Further, since the contrast of light and dark is optimized and detected while switching the RGB filters according to the color of the ink, it is possible to effectively cope with light ink colors.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of an inkjet nozzle failure detection device according to the present invention.

FIG. 2 is a diagram illustrating an example of a test pattern.

FIG. 3 is a schematic diagram of a sensor unit.

FIG. 4 is a schematic diagram of a light receiving unit.

FIG. 5 is a side view of the sensor means showing the detection principle of the present invention.

FIG. 6 is a front view of FIG. 5;

FIG. 7 is a diagram illustrating a method of detecting the shape of an ink particle.

FIG. 8 is a diagram illustrating a sampling method when detecting the shape of an ink particle.

FIG. 9 is a diagram illustrating a method for calculating the area of an ink particle.

FIG. 10 is a diagram illustrating a method for determining a displacement of an ink particle in the X direction.

FIG. 11 is a diagram illustrating a method for determining a displacement of ink particles in the Y direction.

FIG. 12 is a block diagram of a control device of the present apparatus.

FIG. 13 is a diagram illustrating an effect of an RGB filter.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Sensor means 2 Drive means 3 Test recording medium 4 Print head 5 Sending roll 6 Take-up roll 7 Rubber roll 8 Rubber roll 9 Guide roll 10 Printing position 11 Detection position 12 Block 13 Block 14 Paper out detection sensor 15 Test pattern 16 Ink grain 17 Head Reference Signs List 18 group 19 light source 20 box 21 condenser lens 22 slit 23 filter 24 light receiving element group 25 mounting plate 26 reflected light 27 CCD 28 CPU unit 29 inkjet printer

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Shoji Takezawa 1-10-1 Kiya, Fukui City, Fukui Prefecture F-term (reference) 2C056 EB27 EB40 HA58 KD06 2C057 DD09 DD10 2C061 AQ05 KK13 KK18 KK26 KK28

Claims (6)

[Claims] 1. A method for detecting a normal or abnormal ink ejection of the test pattern by imaging a test pattern of ink droplets printed on a test recording medium by a print head and processing the image. A plurality of rows of ink particles in are imaged as one group, the shape of the ink particles in the group is grasped, an average area is obtained by obtaining individual areas, and a predetermined ratio of the area to the average area is obtained. And determining whether the ink ejection is normal or abnormal by comparing the area of each ink particle with a determination value as a determination value, wherein the step is performed for all nozzles of the print head in a group unit. Defect detection method. 2. A method for imaging a test pattern of ink droplets printed on a test recording medium by a print head and processing the image to detect whether or not ink ejection is normal or abnormal for the test pattern. Are imaged as one group, the shape of the ink particles in the group is grasped, the X-direction edge position of each ink particle is obtained, and the maximum value of the X-direction edge position is obtained. An ink jet nozzle, comprising: determining a displacement of an ink particle in the X direction by comparing a difference between the printhead and the minimum value with a determination value; and performing the process in a group unit for all nozzles of the print head. Defect detection method. 3. A method for detecting a normal or abnormal ink ejection of the test pattern by imaging a test pattern of ink droplets printed on a test recording medium by a print head and processing the image. Are imaged as one group, the shape of the ink particles in the group is grasped, and the center position of each ink particle is obtained by obtaining the edge position in the Y direction of each ink particle. From the center position, a Y-direction center interval between two adjacent ink droplets is obtained, and a difference between a maximum value and a minimum value of the Y-direction center interval is compared with a determination value to thereby determine a displacement of the ink particles in the Y direction. A method of detecting a defect of an ink jet nozzle, comprising: a step of determining; and performing the step in a group unit for all nozzles of the print head. . 4. When capturing an image of ink droplets in the group, the R of the light receiving unit is determined according to the ink color of the group.
The method for detecting a defect of an inkjet nozzle according to any one of claims 1 to 3, wherein capturing is performed with an optimum contrast while switching a GB filter. 5. A test recording medium driving unit, and a sensor unit for detecting a test pattern of ink droplets printed on the test recording medium, wherein the sensor unit includes a light source and ink in the test pattern. Light receiving means for condensing reflected light from one group consisting of a row of particles and receiving the reflected light through a slit means and a filter means, and further comprising: a direction orthogonal to the ink particle row of the one group. 1. A defect detection device for an ink jet nozzle, comprising: a scanning unit that scans an image; and an image processing unit that processes an output signal from the light receiving unit. 6. The apparatus according to claim 5, wherein said filter means and said light receiving means comprise a light receiving element array having RGB filters arranged in three rows.