JP2001092966A - Method and device for processing image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for processing image with which stripes on the surface of an object to be examined can be detected while using a general board for image processing. SOLUTION: This device is composed of a first means 12 for emphasizing a fine signal on the basis of a moving average for each line, a second means 13 for comparatively evaluating the change amount of a peak position and the change amount of an amplitude inside the prescribed block of data obtained from this first means 12, and a means 4 for discriminating the grade of the object to be examined on the basis of the amplitude and the number of generated stripes inside a prescribed area from the result evaluated by this second means 13 and displaying the result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and an image processing apparatus, and more particularly to an image processing method and an image processing apparatus for detecting streaks on a surface of a sheet-like inspection object.

[0002]

2. Description of the Related Art As a method for detecting a streak or a flaw in a sheet-like inspection object, particularly as a method for detecting a streak having a low contrast, a narrow width, and a long length in a transport direction, after emphasizing a change amount of density information, A method for detecting streaks and flaws by comparing and evaluating the amount of change in data divided into a lattice (Japanese Patent Laid-Open No.
No. 74262).

[0003]

In the case of the above-mentioned prior art, the condition is that the width is narrow and continuous for a certain length or more in the transport direction. In the case where there is a minute density change which is not a defect such as the same length / width in the same direction, there is a problem that it cannot be separated from the streak.

The present invention has been made in view of such a problem, and is intended to solve the problem of a minute density change which is noise generated at the same length / width with the same contrast in the same direction as the streak. It is an object of the present invention to provide an image processing method and an image processing apparatus capable of separating and detecting only streaks.

[0005]

According to the first aspect of the present invention, a sheet-like inspection object is imaged by an image sensor, and the presence or absence of a streak of the inspection object is determined from the image data. In an image processing method for detection, a first process for emphasizing a minute signal based on a moving average for each line, and a change in a peak position and a change in amplitude in a predetermined section of data obtained from the first process It is characterized by comprising a second process of comparing and evaluating the amount.

With such a configuration, by comparing and evaluating the small signal emphasis and the amount of change in the peak position and the amount of change in the amplitude, a streak generated in the same direction as the swell from the swell generated in the film forming process is obtained. Separation and extraction are possible.

(2) The invention described in claim 2 is characterized in that the first processing is to add a plurality of line data after subtracting a moving average of the line data from the original line data. .

With this configuration, it is possible to emphasize and extract a small signal buried in the periphery by subtracting the moving average of the line data from the original line data and adding a plurality of lines.

Here, the moving average is a process of shifting the obtained line data in the line direction and taking an average with the original data, and is generally a smoothing process. However, since the result may greatly change depending on the setting of the shift amount, it is necessary to determine an appropriate value based on the frequency of the obtained original signal.

According to a third aspect of the present invention, in the second processing, a change amount of the peak position and a change amount of the amplitude of the peak position within a predetermined section in the transport direction of the inspection object are equal to or less than a predetermined amount. In this case, it is characterized in that a streak is determined.

With this configuration, it is possible to clearly separate the streak and the peripheral undulation by determining whether the amount of change in the signal peak position and the amount of change in the amplitude in the predetermined section are equal to or smaller than the predetermined amount.

(4) A moving average for each line in an image processing apparatus for imaging a sheet-like object to be inspected with an image sensor and detecting the presence or absence of a streak of the object from the image data. And a second means for comparing and evaluating a change amount of a peak position and a change amount of an amplitude in a predetermined section of data obtained from the first means.
And means for judging the grade of the inspection object based on the amplitude and the number of streaks occurring within a predetermined area from the result evaluated by the second means, and displaying the result.

According to this structure, the means for judging the grade of the inspection object based on the amplitude and the number of occurrences within a predetermined area from the results obtained up to the second processing and providing a result display means is provided. This makes it possible to automatically determine the quality of the entire inspection object.

(5) The invention according to claim 5 is characterized in that the comparative evaluation is classified into grades based on a multiplication result of a numerical value weighted according to the magnitude of the amplitude and the number of streaks generated within a predetermined area. Features.

With this configuration, an ambiguous judgment is made by grading the numerical value weighted according to the magnitude of the amplitude in the evaluation result and the result of multiplication of the number of occurrences within a predetermined area. Without this, it is possible to make a clear numerical evaluation of the quality evaluation result.

(6) The invention according to claim 6 is characterized in that the test object is a triacetyl cellulose film. With this configuration, by using a triacetylcellulose film as an object to be inspected, streaks generated on the surface of the triacetylcellulose film can be accurately detected.

[0017]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the present invention. In the figure, reference numeral 1 denotes a linear sensor as an image sensor, for example, a CCD or the like is used. In addition, a photoelectric conversion element such as a photomultiplier tube (PMT) is used as the imaging element. 10 is CC
An image processing board that receives the output of D1 and detects and outputs streaks of the sheet-like inspection object. In the present invention, a general-purpose image processing board is used.

The image processing board 10 includes a primary image memory 1
1, first processing unit 12, second processing unit 13, and determination processing unit 1
4. The primary image memory 11 is a CCD
1, the first processing unit 12 emphasizes a minute signal based on a moving average for each line, and the second processing unit 13 stores data obtained from the first processing unit 12. The comparison processing unit 14 compares and evaluates the amount of change in the peak position and the amount of change in the amplitude in the predetermined section. This is to determine the grade of the inspection object.

A secondary image memory 2 for storing image information;
Reference numeral 3 denotes a CPU which is connected to the secondary image memory 2 and controls various image processing. Reference numeral 4 denotes a result display unit that displays a result of the determination process performed by the image processing board 10. As the result display unit 4, for example, a monitor or a printer is used.
The operation of the device configured as described above will be described below.

FIG. 2 is an explanatory view of the transport direction of the present invention.
Reference numeral 21 denotes a sheet as an object to be inspected, arrow A denotes a conveying direction of the sheet 21, and arrow B denotes a line from which an image is captured by a linear sensor (for example, a CCD camera) 20.

FIG. 3 shows that the sheet 21 is conveyed by an optical system of 0.5 m.
This represents each line data of a section of a predetermined width when measured every m. In the figure, 0.5 mm pitch, 20
The figure shows an example of scanning up to measurement. In the figure,
There is a streak in the part where "streak" is written. The horizontal direction in the figure is the width position (main scanning direction), and the vertical axis is the measured value.

The image data detected by the CCD 1 is converted into digital data and stored in the primary image memory 11 once. The image data stored in the primary image memory 11 is stored in the secondary image memory 2 by the CPU 3. The measurement data shown in FIG. 3 shows the raw data measured in this manner. The CPU 3 drives the first processing unit 12 to perform the first processing.

The first processing unit 12 performs processing for emphasizing a minute signal based on a moving average for each line. First, a moving average process is performed based on the original data shown in FIG. The moving average process is a process of calculating an average value of the signal of each line shown in FIG. 3 and a signal shifted by, for example, 30 points from the original signal. Thereafter, a process of subtracting the moving average result from the original data is performed. As a result, large undulations as shown in FIG. 3 disappear, and substantially flat image data is obtained for each line as shown in FIG.

The first processing section 12 performs an addition process on the image data shown in FIG. 4 every four lines. As a result, an image subjected to the enhancement processing as shown in FIG. 5 is obtained. By this processing, an image in which the streak component hidden until then is emphasized is obtained. This image data is stored in the secondary image memory 2.

Next, the CPU 3 activates the second processing unit 13. The second processing unit 13 compares and evaluates the amount of change in the peak position and the amount of change in the amplitude of the data obtained from the first processing in a predetermined section in the transport direction of the inspection object. Second processing unit 1
3 compares and evaluates the change amount of the peak position and the change amount of the amplitude for each predetermined section. As a result, by determining whether or not the amount of change in the signal peak position and the amount of change in the amplitude in the predetermined section are equal to or less than the predetermined amount, it is possible to clearly separate the streak and the surrounding undulation.

According to this embodiment, by comparing and evaluating the small signal emphasis and the amount of change in the peak position and the amount of change in the amplitude, a small undulation generated in the film forming process is generated in the same direction as the undulation. It is possible to separate and extract streaks.

According to the embodiment, after the moving average of the line data is subtracted from the original line data,
By adding a plurality of line data, it becomes possible to emphasize and extract a small signal buried in the periphery.

Next, the CPU 3 drives the determination processing section 14. As a result, the result processed by the second processing unit 13 enters the determination processing unit 14. The determination processing unit 14 includes the second processing unit 1
From the result obtained in step 3, the grade of the inspection object is determined based on the amplitude and the number of streaks generated within a predetermined area, and the result is displayed on the result display unit 4. According to this, it is possible to automatically perform quality control of the entire inspection object.

Next, the operation of the determination processing section 14 will be described in detail. The amplitude is divided into the following three stages according to the magnitude of the amplitude. S (relatively small streaks) M (medium streaks) L (large streaks) Next, weighting is performed on the amplitude. S → 1 point M → 2 points L → 5 points The grade judgment is divided into five ranks A to E. A = Total score is 0 to 2 B = Total score is 3 to 5 C = Total score is 6 to 10 D = Total score is 11 to 15 E = Total score is 16 or more The total score is calculated from the number of streaks generated within a predetermined area. Ask for and make a decision. (A) S → 3 (pieces), M → 1 (pieces), L → 1 (pieces) (evaluation) The total score is 1 × 3 + 2 × 1 + 5 × 1 = 10 C (b) S → 2 (pieces), M → 0 (pieces), L → 0 (pieces) Total score is 1 × 2 + 2 × 0 + 5 × 0 = 2 A (c) S → 0 (pieces), M → 1 (pieces), L → 3 (pieces) From 1 × 0 + 2 × 1 + 5 × 3 = 17 E or more, (b) has the highest evaluation, (c) has the worst evaluation, and (a) has an intermediate evaluation.

As described above, according to the present embodiment, the determination processing unit 14 classifies a grade from the multiplication result of the numerical value weighted according to the magnitude of the amplitude in the evaluation result and the number of occurrences of streaks within a predetermined area. By doing so, it is possible to make a clear numerical evaluation of the quality evaluation result without making an ambiguous determination.

When a triacetylcellulose film is used as an object to be inspected, it is possible to accurately detect streaks generated on the surface of the triacetylcellulose film.

FIGS. 6 and 7 are flowcharts showing an image processing algorithm according to the present invention. First, the primary image memory 11 and the secondary image memory 2 for temporarily storing image data are cleared (S1). Next, image data of a sheet-like inspection object is taken in line by line (S2). Next, the captured image data is converted into digital image data using an A / D converter (S3). Next, the CPU 3 performs filtering processing on the A / D-converted image data to remove noise included in the image data (S
4). Image data from which noise has been removed (density data)
Is stored in the secondary image memory 2 (S5).

When the capture of the image data for one screen is completed, the CPU 3 activates the first processing unit 12. The first processing unit 12 reads the image data from the secondary image memory 2 and first calculates a moving average of one line (S6). next,
Subtract the moving average from the original data. As a result, the undulation component of the original image data is removed, and the image data becomes substantially flat data. Next, the first processing unit 12
Checks whether the processing for M lines is completed (S8). If not, the process returns to step S6. When the process is completed, an addition process is performed for each M lines (S9).

When the processing for M lines is completed, C
PU 3 starts the second processing unit 13. Second processing unit 13
Calculates the peak position fluctuation amount and the amplitude fluctuation amount within a predetermined section in the transport direction of the inspection object between n lines of the addition result (S10).

Next, the second processing unit 13 sets the amplitude fluctuation amount <
It is checked whether “a” and the peak position fluctuation amount <b (S11). If so, the second processing unit 1
3 stores the occurrence position and its amplitude (S12).

Next, it is checked whether the processing has been completed for all data (S13). If the processing has not been completed, the process returns to step S10, and the peak position fluctuation amount and the amplitude fluctuation amount are calculated. If the processing has been completed for all data, the CPU 3 activates the determination processing unit 14 and performs a grade determination (S14). Details of the grade determination are as described above. When the grade determination is completed, the CPU 3 displays the determination result on the result display unit 4 (S15).

[0037]

As described above, according to the present invention, the following effects can be obtained. (1) According to the first aspect of the present invention, a first process for emphasizing a minute signal based on a moving average for each line, and a peak position in a predetermined section of data obtained from the first process. The second process for comparing and evaluating the change amount and the change amount of the amplitude makes it possible to compare and evaluate the small signal emphasis and the change amount and the amplitude change amount of the peak position, thereby producing a minute undulation generated in the film forming process. Can be separated and extracted from the streaks that occur in the same direction as the swell.

(2) According to the second aspect of the invention, the first processing is to subtract a moving average of the line data from the original line data and then add a plurality of line data. By subtracting the moving average of the line data from the original line data and adding a plurality of lines, it becomes possible to emphasize and extract a minute signal buried in the periphery.

(3) According to the third aspect of the present invention, in the second process, the change amount of the peak position and the change amount of the amplitude within a predetermined section in the transport direction of the inspection object are equal to or less than a predetermined amount. By determining whether the amount of change in the signal peak position and the amount of change in the amplitude in a predetermined section are equal to or less than a predetermined amount by determining the streak, the streak and the peripheral undulation can be clearly separated.

(4) According to the fourth aspect of the present invention, the first means for emphasizing the minute signal based on the moving average of each line, and the data obtained from the first means in a predetermined section are provided. A second method for comparing and evaluating the amount of change in peak position and the amount of change in amplitude
And means for judging the grade of the inspection object based on the amplitude and the number of streaks occurring within a predetermined area from the result evaluated by the second means, and displaying the result. From the results obtained up to this point, determine the grade of the inspection object based on its amplitude and the number of occurrences within a predetermined area, and provide means for displaying the result, so that the quality of the entire inspection object can be automatically determined. Becomes possible.

(5) According to the fifth aspect of the present invention, the comparative evaluation is performed by grading the numerical value weighted according to the magnitude of the amplitude and the result of multiplication of the number of streaks occurring within a predetermined area. By doing so, the numerical value weighted according to the magnitude of the amplitude in the evaluation result and the result of multiplication of the number of occurrences within a predetermined area, so as to be graded, without making a vague determination, the quality evaluation result It is possible to clearly determine the numerical value.

(6) According to the invention described in claim 6, the test object is a triacetylcellulose film, and the test object is a triacetylcellulose film. The streaks that occur in the area can be accurately detected.

As described above, according to the present invention, it is possible to provide an image processing method and an image processing apparatus capable of detecting a streak on the surface of a test object using a general-purpose image processing board.

[Brief description of the drawings]

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

FIG. 2 is an explanatory view of a transport direction of the present invention.

FIG. 3 is a diagram showing raw measurement data.

FIG. 4 is a diagram showing image data after moving average processing.

FIG. 5 is a diagram showing image data after addition processing.

FIG. 6 is a flowchart illustrating an image processing algorithm.

FIG. 7 is a flowchart illustrating an image processing algorithm.

[Explanation of symbols]

 Reference Signs List 1 CCD 2 Secondary image memory 3 CPU 4 Result display unit 10 Image processing board 11 Primary image memory 12 First processing unit 13 Second processing unit 14 Judgment processing unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G051 AA41 AB02 CA03 DA06 EA08 EA11 EA25 EB01 EC01 EC03 5B057 AA01 BA02 CA02 CA08 CA12 CA16 CC02 DA03 DB02 DB05 DB09 DC02

Claims (6)

[Claims] 1. An image processing method for imaging a sheet-like object to be inspected by an image sensor and detecting presence or absence of a streak of the object from the image data, wherein a minute signal is emphasized based on a moving average for each line. 1. An image processing method comprising: a first process; and a second process of comparing and evaluating a change amount of a peak position and a change amount of an amplitude in a predetermined section of data obtained from the first process. . 2. The image processing method according to claim 1, wherein the first processing is to add a plurality of line data after subtracting a moving average of the line data from the original line data. 3. The method according to claim 2, wherein the change in the peak position and the change in the amplitude of the peak position within a predetermined section in the transport direction of the inspection object are equal to or less than a predetermined amount. The image processing method according to claim 1, wherein 4. An image processing apparatus for imaging a sheet-like object to be inspected with an image sensor and detecting the presence or absence of a streak of the object from the image data, wherein a small signal is emphasized based on a moving average for each line. A first means, a second means for comparing and evaluating the amount of change in the peak position and the amount of change in the amplitude in a predetermined section of the data obtained from the first means, and a result evaluated by the second means Means for judging the grade of the inspection object based on the amplitude and the number of streaks generated within a predetermined area and displaying the result. 5. The image according to claim 4, wherein the comparative evaluation is performed based on a result obtained by multiplying a numerical value weighted according to the magnitude of the amplitude and the number of occurrences of streaks within a predetermined area. Processing equipment. 6. The image processing apparatus according to claim 4, wherein the inspection object is a triacetyl cellulose film.