JP2001192963A - Inspection device for sheet having tissue structure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inspection device of a sheet, capable of precisely inspecting the defect having a different tissue structure of which the inspection has been difficult, and capable of extracting the defect in high accuracy even if a disturbance such as the bending of the warp and the waving of the surface are present. SOLUTION: This inspection device for sheet having a tissue structure, having a camera-scanning means for scanning a camera in a direction orthogonal to the traveling direction of the sheet, an image-taking means for taking the image of the surface of a sheet-like material in the scanning step, a statistic-processing means for processing the taken image data to a statistic value, and an abnormal value-judging means for judging the presence or absence of the abnormal value by comparing the statistic value extracted by the statistic-processing means with a previously set standard value, has a statistic information-extracting means for scanning the camera over the whole width of the sheet and extracting the static value based on the image taken at each position, and a standard value-setting means for setting the standard value for judging the good or bad, usable for the inspection based on the statistic value of the whole width of the sheet obtained by the statistic information means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a defect inspection apparatus for a sheet having a texture structure, and more particularly, to a defect inspection apparatus for a woven fabric, particularly a warp defect, generated in a woven fabric during weaving.

[0002]

2. Description of the Related Art Conventionally, as an apparatus or method for inspecting the appearance of a woven fabric, an image of the surface of the woven fabric is taken by a camera, and image density data obtained from the taken image is compared with a threshold value to detect abnormalities in the appearance. An automatic inspection system that detects the abnormalities by irradiating the woven fabric with laser light and causing the reflected or transmitted light to be received by the light-receiving element, comparing the level of the amount of light received with a threshold value, and detecting the abnormality. Are known. However, there is also a problem that defects that can be detected by this inspection method are limited to relatively large defects such as thread dropouts that can be easily determined visually by a person.

As a method for improving the detection accuracy, there is an invention disclosed in Japanese Patent Application Laid-Open No. 4-148852, for example. According to the present invention, light transmitted from a light source irradiating a woven fabric is received by a light receiving element through an optical slit arranged in a yarn direction of an inspection target, and abnormality is detected by comparing a received light waveform with a reference waveform. A method for doing so is disclosed. In this method, the presence or absence of a defect is determined based on the characteristic amount of the light transmitting portion, that is, the extracted woven fabric opening.

Japanese Patent Laid-Open Publication No. Hei 3-249243 discloses a method in which a light receiving sensor is formed as a pair of known combs, and an abnormality is detected by comparing a difference value between outputs of the two and a preset threshold value. Have been. In this method, since the density information obtained by dividing the woven cloth narrow region into two parts is reflected on the pair of comb-shaped light receiving sensors, even if there is vibration or disturbance light, there is an effect that the difference value output between the two cancels out. In order to ensure the same accuracy as the inspection performed visually, it is necessary to extract the characteristic of the yarn component itself in addition to the conventional characteristic inspection of the woven fabric opening. For example, in the case of a defect such as a warp inflow defect as described above, the presence or absence of a defect is determined by visual inspection in accordance with only the vertical relation of the yarn to be inspected at the yarn interlacing point. Therefore, in order to mechanically detect this defect, it is only necessary to extract only the yarn component in which the warp is above or below the weft on the yarn interlacing point coordinates.

The present inventors have disclosed a woven cloth inspection apparatus in which the above-mentioned problems have been solved in Japanese Patent Application No. 7-19811. According to the present invention, the light illuminated onto the woven fabric by
An image is picked up by a CD (Charge Coupled Device) element, and woven fabric information (yarn density, yarn inclination, etc.) is calculated based on the obtained image data, and the yarn pitch obtained from the woven fabric information is calculated. By comparing the correlation value over the entire image data between the region having the width and the other region located at an integer multiple of the yarn pitch width with the set threshold value, the plain weave is mainly performed. Defects are detected with high precision over the entire width of the woven fabric under the same optical conditions, even if the woven fabrics have different densities, without distinguishing between warps and wefts. The present inventors have further proposed an inspection apparatus capable of inspecting defects of woven fabrics having different weaving structures such as satin and twill under the same optical conditions.
272400. The present invention adds the woven tissue cycle information to the woven cloth information described in Japanese Patent Application No. 7-98171, and compares the area having the tissue cycle width with another area at an integer multiple of the width. By comparing the overall correlation value with the set threshold value, defects in woven fabrics having different woven structures and different densities can be detected under the same optical conditions without discrimination between warp and weft yarns. Is detected with high accuracy over the entire width of the image.

[0006]

SUMMARY OF THE INVENTION The above-mentioned JP-A-4-14-1
In the invention disclosed in Japanese Patent No. 48852, for example, in the case of a defect in which the rear part of the woven fabric is not so different from a non-defective product such as a warp yarn pouring defect, the defect cannot be detected, and the detection accuracy is significantly reduced. There was a problem. Further, in this method, it is assumed that the weaving density is constant and the yarn is parallel to the optical slit and the inspection target direction. However, there are various woven densities of actual woven fabrics, and there is a problem that the optical slit needs to be replaced each time. In addition, actual woven yarns, especially warp yarns, are curved on both sides of the woven fabric, so that the above conditions cannot be maintained, and there is a problem that the detection accuracy is reduced. In the invention disclosed in JP-A-3-249243,
As in the invention disclosed in Japanese Patent Application Laid-Open No. 4-148852, the detection accuracy is reduced when there is a limit to the defects that can be extracted, when the weaving density changes, or when the balance between the comb-shaped light receiving sensor and the yarn to be inspected cannot be maintained. There is a problem that. Also,
In the invention disclosed in Japanese Patent Application No. 7-198171, it has been recognized that there is little effect on defects of woven fabrics having different woven structures such as satin weave and twill weave other than plain weave.

Furthermore, in the invention disclosed in Japanese Patent Application No. 8-272400, the binary image shape of the yarn component to be inspected on the yarn entanglement point extracted by the binarization process has a delicate shape on the surface of the woven fabric. It is not always uniform due to the difference in unevenness distribution, fuzzing, flotation adhesion, etc., and as a result, the whole of the area having the tissue cycle width and other areas at an integer multiple of this width When the statistical values over the range were compared, it was found that the statistical values in the normal range varied, and as a result, the S / N ratio decreased. The present invention has been made to solve the above problems, and the object of the invention described in the claims is that the woven fabric, the surface properties are not affected, and disturbances such as fluff, fluff, and dirt are generated. Even in such a case, an object of the present invention is to provide a woven cloth inspection device capable of detecting only a woven defect with high accuracy and automatically performing inspection at low cost.

[0008]

That is, claim 1 of the present invention.
According to the invention, there is provided a camera scanning means for scanning a camera in a direction perpendicular to a traveling direction of a sheet, an imaging means for taking an image of the surface of the sheet-like object in the scanning process, and processing the taken image data into statistical values. In a sheet inspection apparatus including a statistic processing unit that performs the processing, and an abnormal value determination unit that determines whether there is an abnormal value by comparing a statistic value extracted by the statistical processing unit with a preset reference value. A statistical information extracting means for causing a camera to scan the entire width of the sheet and extracting a statistic based on an image taken at each position, and a statistic of the full width of the sheet obtained by the statistical information means are used for inspection. A reference value setting means for setting a reference value for determining whether or not the product is acceptable. According to this configuration, since the reference value for pass / fail judgment required for the inspection is automatically extracted in advance, even if the inspection is performed on a loom for weaving many brands having different weaving patterns and different weaving densities, a completely automatic operation is performed. Inspection becomes possible.

The invention according to claim 2 is characterized in that the reference value set by the reference value setting means is not necessarily uniform over the entire width of the sheet. According to this configuration, even when the warp is inclined near the edge as in a woven fabric immediately after weaving, or when the image range to be imaged changes due to the slack of the woven fabric, such image information is obtained. Since the reference value is set on the basis, the inspection can be performed with high accuracy without being affected. The invention according to claim 3 is characterized in that the statistical information extracting means includes means for processing a statistical value of a tissue repetition unit in the width direction in the captured image or a statistical value of a sheet structure minimum unit. According to this configuration, the captured image data can be processed into the statistical value of the tissue repeating unit or the statistical value of the sheet structure minimum unit.
Irrespective of the density, there is an advantage that the extracted statistical value does not change regardless of where the image is taken.

An embodiment of a sheet inspection apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram of an inspection apparatus according to an embodiment of the present invention. The inspection device includes a camera lens 3, a CCD camera 4, an A / D converter 5, a frame memory 7, various image processing and statistic extraction circuits 8, an image bus 13, a CPU bus 14, a CPU (Cent
ral Processing Unit) 9, ROM (Read Only Memor)
y) 10, a RAM (Random Access Memory) 11 is included.
Light emitted from the light source 1 is reflected on the surface of the woven fabric 2, condensed by the camera lens 3, and is imaged on a CCD element in the camera 4. The light source 1 is preferably a light source that emits a highly directional light flux in order to enhance the effect of enhancing a yarn to be inspected, which will be described later. In the present embodiment, an example (FIG. 2) in which a plurality of light emitting diodes are combined into a module to secure an imaging field of view and illuminance is shown. In the present embodiment, the area type C
An example using a CD camera is shown. If the line type CC
When a D camera is used, a one-dimensional / two-dimensional converter 6 is provided between the A / D converter 5 and the frame memory 7 to convert one-dimensional image data into two-dimensional image data, as shown in FIG. Just add.

The wavelength of the light reflected from the woven cloth is
There is no particular problem if is a wavelength having sensitivity. There is no particular problem in the intensity of light applied to the woven fabric as long as sufficient charge can be accumulated within the update cycle of the CCD. By irradiating the woven fabric under these illumination conditions, the camera 4 captures an image in which only the yarn components to be inspected, which are arranged above the interlace points of the warp and the weft constituting the woven fabric 2, are brightly emphasized. it can. An example showing this effect is shown in FIG.

The grayscale image data picked up by the CCD camera 4 is converted into 8-bit digital image data by an A / D conversion circuit 5 and then stored in a frame memory 7. As described above, when the CCD camera 4 is of a line type, 1
The dimensional / two-dimensional converter 6 converts the one-dimensional image data into two-dimensional image data and stores it in the frame memory 7. The tissue repetition period of the warp to be inspected is obtained from the original image. In the embodiment, in order to calculate the tissue cycle, the captured image data is added to the density in the direction of the warp to be inspected, converted into one-dimensional data, and a peak value of a spectrum obtained by performing FFT processing on the density data is obtained. The repetition period of the warp was calculated. As another method, for example, a method of extracting the characteristics of the density waveform in the direction orthogonal to the inspection target yarn to obtain the periodicity can be considered, but the method is not particularly limited.

A rectangular area is set in the captured image based on the obtained warp tissue repetition cycle size, and the statistical value of this area is extracted. This statistic value extraction method is a statistic value extraction method necessary for calculating a preset pass / fail judgment reference value. The same is true for the method of extracting statistical values extracted during inspection. This indicates that inspection can be performed even if the state of the woven fabric is not necessarily uniform over the entire width. The statistics of the rectangular area include the maximum value, average value, minimum value, variance value, covariance value, variation coefficient, distortion degree, sharpness, correlation coefficient, standard deviation value, and mode density value of the density, Defects are extracted by comparing these statistics with a difference value, image pattern correlation, or statistics when comparing them in the vicinity area in the image and the reference data. FIGS. 5A and 5B are diagrams illustrating an example of defect extraction. (A) shows a normal tissue, and (b) shows an example of a flow-in defect. Although it is difficult to extract such a defect by comparing the conventional woven fabric opening, as shown in FIG. 5, the inspection target yarn component (above described illumination) Comparing the thread components that are emphasized and brightly imaged), the positional pattern consistency in the left and right rectangular regions in (b) is completely different. That is, the position patterns of the warp components (represented by black squares in FIG. 5) on the wefts at the intersections are completely different. Therefore, it can be seen that a defect in the woven tissue can be easily extracted by calculating the pattern consistency between a pair of rectangular regions. Note that the CPU 9 performs the extraction of the defect based on the above statistics. The image including the extracted defect is output by the display of the image display 12.

In the case of in-line inspection of the woven fabric during weaving, for example, the warp yarns in the central region of the woven fabric are arranged in parallel to the flow direction, but the warp yarns are inclined at 3 to 10 degrees in the woven fabric edge region. In addition, for example, when the tension of the woven fabric is poor, the woven fabric may run in a undulating state in the width direction.
It is obvious that if such a woven fabric is inspected, if the presence / absence of a defect is determined using a fixed reference value for quality determination, the accuracy is greatly reduced. For example, if an image of a non-defective part is recorded in advance and the inspection is performed by a so-called pattern matching method in which the quality is compared with an image captured during the inspection, the accuracy is greatly reduced. As a measure for avoiding this problem, statistical values of the entire width of the woven fabric are extracted in advance by the above-described method, and a reference value for the entire width of the woven fabric is automatically generated based on this information. This example is shown in FIG. It can be seen that the reference value is set higher in the edge area (C) of the woven fabric than in the central area because the warp is inclined. The reason why the level is slightly higher in the area (B) is that the woven fabric has undulation due to unevenness in the tension of the sheet. In other words, it can be seen from FIG. 8 that the woven fabric information is successfully reflected in the reference value. In the embodiment, after the statistical values of the full width of the extracted statistical values are smoothed, a value to which an offset of 10% is added is used as a reference value at each coordinate. However,
The offset value, the smoothing constant, and the like may be set according to the brand and the management standard, and are not particularly limited. Table 1 shows an example in which the pattern matching method and this embodiment are compared. The inflow defect of the warp was intentionally generated at each position with respect to the center of the woven fabric (0 mm), and the repetition detection rate was confirmed.
Note that a normal image of the central part of the woven fabric was stored as the reference image of the pattern matching method. It can be seen that the detection rate is greatly reduced in the pattern matching method, whereas 100% is detected in the present embodiment. FIG. 7 shows a camera scan pattern for setting a pass / fail judgment threshold value and a camera scan pattern at the time of inspection. The scanning area for reference value setting is from the position A to the position B, and the inspection area is the following. However, the starting position of each step may be the pattern of FIG. 7A or the pattern of FIG. 7B, and is not particularly limited. When setting the pass / fail judgment threshold value, the woven fabric does not necessarily need to be running, but since the image captured during running the woven fabric is slightly different from the still woven fabric image, the inspection state is further improved. It is preferable to extract the reference value while running the woven fabric in order to extract the reference value for pass / fail judgment in a state close to.

[0015]

[Table 1]

FIG. 6 is a flowchart showing the processing procedure of the woven cloth inspection device according to the present embodiment. First, the camera is moved to a reference position (point A in FIG. 7) (S1),
The scanning of the CCD camera is started. After the grayscale image data captured by the camera 4 in the scanning state is converted by the A / D converter 5 into 8-bit digital image data,
The data is taken into the frame memory 7 (S2). Next, CP
U9 extracts the warp tissue cycle based on the grayscale image data by the above-described method, and sets a rectangular area in the captured image based on the cycle data (S3). Image processing circuit 8
Extracts a statistic from the gray image data in the set rectangular area (S4).

In step S5, it is determined whether or not statistical values have been extracted for all of the captured image data. If the extraction has not been completed (S5, No),
The rectangular area set in the image is shifted in the direction orthogonal to the warp and set (S6). Then, the processing of steps 4 and 6 is repeated. In step 7, the maximum value of the obtained statistical values is obtained and temporarily stored (S7). If the statistical value extraction of the full width is not completed (S8, No), the processing from S2 to S8 is repeated until it is completed. When the processing of the full width is completed (S8, Yes), the reference value for the pass / fail judgment of the full width is set by the above-described method based on the stored statistical value data of the full width (S9). Thereafter, the inspection is continuously performed in the camera scanning state. The statistical value extraction (S10) method for making a quality judgment from the captured image data is the same as the statistical value extraction method from S2 to S6. In step 11, the extracted statistic is compared with the aforementioned reference value, and if the statistic value indicates a value larger than the reference value (S11, Yes), it is determined that a defect has occurred (S12). If a defect occurs, the inspection is interrupted and the process waits until a defect repair completion input is received. When the input of the completion of the defect repair is received (S13, Yes), the inspection is continued and the inspection is restarted (S14, Yes). If the statistical value is smaller than the reference value, the processing of S10 and S11 is repeated. The program for controlling the entire system is RO
It is stored in M10. When inspecting the full width of the woven fabric,
The sensors may be traversed in the woven fabric width direction or a plurality of sensors may be arranged at equal intervals in the woven fabric width direction.

In the above description, the processing relating to the in-line inspection of the woven fabric during weaving has been described. However, the present invention can also be applied to automatic inspection of a woven fabric that has been woven. Further, the present invention can be applied to a sheet having regular characteristics other than the woven fabric. FIG. 9 is a diagram illustrating a case where the inspection apparatus according to the present embodiment is applied to a loom. The light source 1 is irradiated from above onto the woven fabric being woven, and an image is taken by the CCD camera 4. C
The CD camera 4 is mounted so as to be movable along a movement axis 40.

[0019]

According to the present invention, it has become possible to reliably detect a defect having a different woven texture, which has been difficult to detect.
Further, even when there is disturbance such as warp of the warp or undulation of the surface, the defect can be extracted with high accuracy.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram of an inspection apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining an arrangement of a lighting module in the embodiment of the present invention.

FIG. 3 is a diagram for explaining an effect of illumination in the embodiment of the present invention.

FIG. 4 is an example for explaining a rectangular area for calculating a statistic in the embodiment of the present invention.

FIG. 5 is an example for explaining a defect extraction method in the embodiment of the present invention.

FIG. 6 is a flowchart illustrating an operation according to the exemplary embodiment of the present invention.

FIG. 7 is a diagram illustrating a scanning pattern of a camera according to the embodiment of the present invention.

FIG. 8 is an example in which a pass / fail judgment reference value in the embodiment of the present invention is automatically extracted.

FIG. 9 is a perspective view showing an appearance according to the embodiment of the present invention.

[Explanation of symbols]

1: LED, 2: Woven cloth, 3: Camera lens, 4: Camera, 5: A / D converter, 6: 1D / 2D converter,
7: Image memory, 8: Image processing circuit, 9: CPU, 1
0: ROM, 11: RAM, 12: Image display, 13:
Image bus, 14: CPU bus, 31: Camera case, 3
2: transparent member, 33: LED module, 40: single axis moving stage

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G051 AA40 AB02 BA01 BA20 CA03 CA04 CB01 CD04 DA01 DA06 EA08 EA12 EA14 EB01 EB02 EC02 EC03 ED11 FA10 3B154 AB20 BA53 BC42 BC48 CA16 CA23 CA27 DA13 DA30

Claims (3)

[Claims] 1. A camera scanning means for scanning a camera in a direction perpendicular to a running direction of a sheet, an imaging means for taking an image of the surface of the sheet-like object in the scanning process, and processing the taken image data into statistical values. In a sheet inspection apparatus including a statistic processing unit that performs the processing, and an abnormal value determination unit that determines whether there is an abnormal value by comparing a statistic value extracted by the statistical processing unit with a preset reference value. A statistical information extracting means for causing a camera to scan the entire width of the sheet and extracting a statistic based on an image taken at each position, and a statistic of the full width of the sheet obtained by the statistical information means are used for inspection. An inspection apparatus for a sheet having an organizational structure, comprising: a reference value setting unit for setting a reference value for determining whether the sheet is good or bad. The reference value set by the reference value setting means is:
The sheet inspection apparatus having a tissue structure according to claim 1, wherein the sheet width is not necessarily uniform. 3. The statistical information extracting means according to claim 1, further comprising means for processing a statistical value of a tissue repeating unit in a width direction in the captured image or a statistical value of a sheet structure minimum unit. Inspection device for sheet with tissue structure.