JP2012063327A - Defect inspection device and method thereof, and defect inspection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To stably obtain an inspection result of a black spot defect G or the like without the determination result of the black spot defect G or the like being different depending on an inspector.
An imaging unit for imaging an object to be inspected, a defect extracting unit for extracting a defective part in the object to be inspected based on image data acquired by imaging of the imaging unit, and a defect part extracted by the defect extracting unit It is a defect inspection apparatus provided with the outline acquisition part which acquires the outline part, and the defect evaluation part which evaluates a defect part based on the magnitude | size of the change of the light-and-dark level in the outline part acquired by the outline acquisition part.
[Selection] Figure 3

Description

  Embodiments described herein relate generally to a defect inspection apparatus and method for inspecting a defect portion such as a spot defect on an inspection object such as an output phosphor screen provided in an X-ray image intensifier, and a defect inspection program.

  An X-ray image intensifier (X-ray fluorescence intensifier), when X-rays are incident, changes the X-rays into an optical image by an input phosphor screen, and photoelectrons corresponding to the optical image are generated by the photoelectric effect on the photocathode surface. The photoelectrons are accelerated and converged by a high electric field and incident on the output phosphor screen, and a high-brightness optical image is generated on the output phosphor screen. This high-intensity optical image is picked up by an image pickup device such as a charge coupled device (CCD) and output as an image signal.

  In the manufacturing process of this X-ray image intensifier (X-ray fluorescence intensifier), the appearance of the output phosphor screen is inspected. This output phosphor screen is manufactured by depositing a phosphor on a base surface formed in a disk shape such as glass. At the time of manufacturing, a defect portion or a defect due to a spot (hereinafter referred to as a spot defect) may be generated on the output phosphor screen due to foreign matter or the like, and a chip or the like may be generated around the output phosphor screen.

  The appearance inspection is performed by an inspector's visual inspection of defects and spots due to foreign matter or the like existing on the output phosphor screen, and chipping of the periphery of the output phosphor screen. In this appearance inspection, an output fluorescent screen is scanned and irradiated with an electron beam, and a bright spot appearing on the output fluorescent screen is visually observed to determine, for example, a portion where a bright spot does not appear as a defective portion.

JP 2004-294202 A

  Particularly in the case of a spot defect portion, the determination result may differ depending on the inspector, and it is necessary to obtain a stable inspection result.

  According to the embodiment, an image pickup unit that picks up an image of the inspection object, a defect extraction unit that extracts a defect portion in the inspection object based on image data acquired by image pickup of the image pickup unit, and a defect extraction unit. A defect inspection apparatus comprising: a contour acquisition unit that acquires a contour portion of a defect portion; and a defect evaluation unit that evaluates the defect portion based on the magnitude of a change in light and shade level in the contour portion acquired by the contour acquisition unit. is there.

The block diagram which shows the defect inspection apparatus of one Embodiment. The figure which shows the raster scan on the whole surface of the output fluorescent screen by the scanner of the apparatus. The functional block diagram which shows the image processing apparatus in the apparatus. The model which shows the extraction effect | action of the outline of a black spot defect by the outline acquisition part in the same apparatus. The schematic diagram which shows the area for outline extraction used for acquiring the outline of a black spot defect by the outline acquisition part in the same apparatus, and the extraction effect | action of an outline. The schematic diagram which shows the outline of the black spot defect acquired by the outline acquisition part in the same apparatus. The schematic diagram which shows an example which applied the area for density | concentration extraction to the black spot defect by the defect evaluation part in the same apparatus. The model which shows an example which applied the area for light and shade extraction to the black spot defect by the defect evaluation part in the apparatus. The schematic diagram which shows another area for light and shade extraction of the defect evaluation part in the same apparatus. The figure which shows the image data and black spot defect of the output fluorescent screen displayed on the display in the apparatus. The figure which shows the column of the evaluation result containing the final edge intensity | strength etc. of the black spot defect displayed on the display in the apparatus. The defect inspection flowchart in the same apparatus. The figure which shows from extraction of a black spot defect in the same apparatus to extraction of an outline. The figure for demonstrating the modification of the defect extraction part in the same apparatus.

  Hereinafter, an embodiment will be described with reference to the drawings.

  FIG. 1 shows a configuration diagram of a defect inspection apparatus. The inspected object 1 is, for example, an output fluorescent screen (hereinafter referred to as an output fluorescent screen 1) provided in an X-ray image intensifier. The output phosphor screen 1 is manufactured by depositing a phosphor on the surface of a base made of glass or the like, and a spot defect is generated on the output phosphor screen 1 due to foreign matter or the like at the time of manufacture. Sometimes. This apparatus extracts a spot defect that occurs on the output phosphor screen 1, and determines the intensity of the spot defect by quantifying it.

  An electron beam source 2 is provided for inspecting the output phosphor screen 1. The electron beam source 2 emits an electron beam. A deflection yoke 3 is provided on the radiation path of the electron beam emitted from the electron beam source 2. The deflection yoke 3 deflects the electron beam emitted from the electron source 2 and raster scans the entire surface of the output phosphor screen 1 as shown in FIG.

  When the electron beam is raster scanned on the output phosphor screen 1, the phosphor on the output phosphor screen 1 emits fluorescence. If there is a spot defect portion on the output phosphor screen 1, the spot defect portion causes non-uniform fluorescence.

  An imaging unit 4 is provided at a position facing the electron beam source 2 through the output phosphor screen 1. The imaging unit 4 captures an optical image generated on the entire surface of the output phosphor screen 1 when X-ray raster scanning is performed on the entire surface of the output phosphor screen 1, and outputs the image signal. The imaging unit 4 is an imaging element such as a CCD. The imaging unit 4 may be the same imaging device as an imaging device such as a CCD provided in the X-ray image intensifier. Note that the imaging device of the X-ray image intensifier captures a high-brightness optical image generated on the output phosphor screen and outputs the image signal. The image signal output from the imaging unit 4 is sent to an image processing personal computer (hereinafter referred to as an image processing apparatus) 10.

The image processing apparatus 10 performs image processing on the image signal from the imaging unit 4 to inspect a defect on the output phosphor screen 1 and obtains it as the intensity of a spot defect existing on the output phosphor screen 1. The image processing apparatus 10 includes a CPU 11. An image input unit 12, a program memory 13, a data memory 14, an output unit 15, and an operation input unit 16 are connected to the CPU 11.
The image input unit 12 receives the image signal output from the imaging unit 4 and acquires image data including the entire surface of the output phosphor screen 1 from the image signal.

  The program memory 13 stores a defect inspection program in advance. The defect inspection program causes the CPU 11 to extract a spot defect on the output phosphor screen 1 based on image data acquired by imaging by the imaging unit 4, and a spot defect contour extracted by the extraction function. An outline information acquisition function for acquiring outline information including the magnitude of the change in the gray level and an evaluation function for acquiring an evaluation of a spot defect based on the outline information acquired by the outline information acquisition function are realized.

In the data memory 14, for example, image data including the entire surface of the output phosphor screen 1 acquired by the image input unit 12 is stored. The data memory 14 temporarily stores data being processed by the image processing apparatus 10, data being processed, and the like. The data memory 14 also includes threshold values used when extracting a spot defect from the output phosphor screen 1, matrix information for density extraction used when acquiring the magnitude of the change in density level at the outline of the spot defect, and the like. It is remembered.
A display 17 made of liquid crystal or the like is connected to the output unit 15. For example, the output unit 15 displays the evaluation result of the spot defect on the display 17.
For example, a keyboard 18 and a mouse are connected to the operation input unit 16. The operation input unit 16 sends this operation information to the CPU 11 when, for example, a keyboard 18 or a mouse is operated by the inspector.

FIG. 3 is a functional block diagram of the image processing apparatus 10. The image processing apparatus 10 includes a defect extraction unit 21, a contour acquisition unit 22, a defect evaluation unit 23, and a notification unit 24.
The defect extraction unit 21 extracts a spot defect on the output phosphor screen 1 from the image data acquired by the imaging of the imaging unit 4. For extracting the spot defect, for example, a threshold value for binarization set in advance for spot defect extraction is used. The threshold for binarization includes a threshold for black spot defect extraction and a threshold for white spot defect extraction. A black spot defect has a low density level, and a white spot defect has a high density level. Therefore, the threshold value for black spot defect extraction is set to a shading level that is lower by a predetermined value than, for example, an intermediate shading level for extraction as a black spot defect. A density lower than the threshold for black spot defect extraction is extracted as a black spot defect. The threshold for white spot defect extraction is set to a shading level that is higher by a predetermined value than, for example, an intermediate shading level for extracting as a white spot defect. A density higher than the threshold for white spot defect extraction is extracted as a white spot defect.

The contour acquisition unit 22 acquires a contour line along the contour of the black spot defect and the white spot defect extracted by the defect extraction unit 21.
FIG. 4 shows, for example, a schematic diagram of the action of extracting the outline of the black spot defect G. This figure shows a black spot defect G on the binary image data obtained by binarizing image data including the entire surface of the output phosphor screen 1 using the threshold for black spot defect extraction by the defect extraction unit 21. This black spot defect G indicates, for example, “1” due to the binarization of the image data, and a normal part without a defect around the black spot defect G indicates “0”.

  The contour acquisition unit 22 scans the binarized image data including the black spot defect G and reads the binarization level “1” or “0”. The binarization level is read by setting the X and Y directions orthogonal to the binarized image data, scanning first in the X direction, and then scanning in the X direction by shifting one pixel in the Y direction. Then, the entire surface of the output phosphor screen 1 is scanned by repeating the scanning in the X direction and the shifting of one pixel in the Y direction. When the binarization level “1” is read during scanning of the entire surface of the output phosphor screen 1, the process proceeds to extraction of the contour line C of one black spot defect G.

  That is, for example, when the binarization level “1” is read by scanning in the X direction, the binarization level “1” or the surrounding pixels around the pixel position S1 from which the binarization level “1” is read are centered. Read “0”. When the binarization level “1” is read again by reading the binarization level “1” or “0” of the surrounding pixels, the pixel position S2 from which the binarization level “1” is read is used again as a center. The binarization level “1” or “0” of the surrounding pixels is read. The contour line C of the black spot defect G can be extracted by repeating the reading of the binarization level “1” or “0” and the binarization level “1” or “0” of the surrounding pixels. It becomes.

Specifically, the contour extraction area EC as shown in FIG. The contour extraction area EC is composed of a plurality of pixel areas each having a size corresponding to the pixel size, for example, 3 × 3 pixel areas e1 to e9.
The contour acquisition unit 22 scans the contour extraction area EC in the X direction with respect to the binarized image data, and sets the binarization level “1” in the central pixel area e5 as shown in FIG. 5B, for example. When read, the pixel position S1 corresponding to the pixel area e5 is determined to be the outline portion of the black spot defect G.
The contour acquisition unit 22 centers on the pixel area e5 (pixel position S1) from which the binarization level “1” is read, and the surrounding pixel areas e2, e1, e4, e7, e8, e9, e6, e3. Read the binarization level “1” or “0” of the pixel. The order of reading the surrounding pixel areas e2, e1,..., E3 with the central pixel area e5 as the center may be left or right around the pixel area e5.

  The contour acquisition unit 22 reads each of the surrounding pixel areas e2, e1,..., E3, for example, when the binary level “1” is read in the pixel area e8, the pixel position S2 corresponding to the pixel area e8 is displayed as a black spot. The outline of the defect G is determined. The contour acquisition unit 22 moves the pixel area e5 at the center of the contour extraction area EC to the pixel position S2, and after this movement, the pixel area e2, e1,. , E3 read the binarization level “1” or “0” of each pixel.

Thereafter, the contour acquisition unit 22 reads the binarization level “1” or “0” of each pixel area e2, e1,. When the conversion level “1” is read, it is repeatedly determined that the pixel positions S3, S4,..., Sk corresponding to the pixel area are the contour portions of the black spot defect G.
As a result, the contour obtaining unit 22 obtains the contour line Lg of the black spot defect G by continuing each pixel position S1, S2,..., Sk determined as the contour part of the black spot defect G as shown in FIG. To do. The contour acquisition unit 22 stores the position information of the pixel positions S1, S2,..., Sk forming the contour line Lg of the black spot defect G, for example, in the data memory 14 as contour information. In FIG. 6, the pixel positions S1, S2,..., Sk forming the contour line Lg are superimposed on the image data D acquired by the imaging unit 4 imaging. Each pixel value of the image data D has a gray level value. The black spot defect G has a lightness level value that is lower in the center portion, for example.

  When acquiring the contour line along the contour of the white spot defect, the defect extraction unit 21 detects the white spot defect as, for example, “1” by binarizing the image data, and normal without any defects around the white spot defect. This portion may be detected as “0”. Thereby, the outline acquisition part 22 can also acquire the outline of a white spot defect similarly to the acquisition of the outline Lg of the black spot defect G.

  The defect evaluation unit 23 determines the magnitude of the change in all shade levels on the contour line Lg acquired by the contour acquisition unit 22, that is, the shades at the edge portions of the black spot defect G and the white spot defect corresponding to the contour line Lg. The magnitude of the level change is acquired, and the black spot defect G and the white spot defect are evaluated based on the magnitude of the shade level change. In the evaluation of the black spot defect G and the white spot defect, the defect evaluation unit 23 acquires a value obtained by averaging the magnitudes of changes in the light and dark levels as the intensity of the black spot defect G and the white spot defect.

  Specifically, the defect evaluation unit 23 has a density extraction area H as shown in FIG. The light and shade extraction area H is composed of a matrix having a plurality of extraction areas each having a size corresponding to the pixel size, for example, 3 × 3 pixel areas f1 to f9. In the grayscale extraction area H, the central pixel area f5 is a point of interest, and a plurality of pixel areas f2, f8, f4, and f6 that exist in the vertical and horizontal directions of the pixel area f5 and a plurality of pixel areas that exist in both the diagonal directions. The pixel areas f1, f9, f3, and f7 of FIG. Note that the pixel area of each peripheral point is represented as f2, f8,..., F7.

  The defect evaluating unit 23 causes the pixel area f5 of the target point in the grayscale extraction area H to follow the contour line Lg while passing through the pixel area f5 of the target point among the pixel areas f2, f8,. Each of the surrounding points facing each other, for example, a value that maximizes the absolute value of the difference between the light and shade levels of the pixel areas f2 and f8 is sequentially obtained as the intensity on the contour line Lg in the pixel area f5 of the point of interest, Are averaged to obtain the final edge strength Fa of the black spot defect G or the white spot defect.

Specifically, the defect evaluation unit 23 acquires the edge strength F at each point of interest according to the following equation (1).
Edge strength F = maximum value (| f2-f8 |, | f4-f6 |, | f1-f9 |, | f3-f7 |)
... (1)
That is, the edge intensity F is the absolute value | f2−f8 | of each gray level of each pixel area f2, f8 existing in the vertical direction and the absolute value of each gray level of each pixel area f4, f6 present in the horizontal direction. | F4−f6 |, absolute values | f1−f9 | of the gray levels of the pixel areas f1 and f9 existing in the diagonally left direction, and gray levels of the pixel areas f3 and f7 existing in the diagonally right direction. Absolute values | f3−f7 | are respectively obtained, and the maximum value among these absolute values is set as the intensity of the point of interest.

  The defect evaluation unit 23 sequentially uses each pixel position S1, S2,..., Sk on the contour line Lg as a focus point, and acquires each edge strength F at each focus point. The defect evaluation unit 23 averages each edge intensity F at each pixel position S1, S2,..., Sk on the contour line Lg, and uses the averaged value as the final edge intensity of the black spot defect G or the white spot defect. Acquired as Fa.

In order to obtain the edge strength F, the absolute value | f2-f8 || f4-f6 | ... | f3-f7 | of the shade levels of the pixel areas f2 and f8 existing in the vertical and horizontal directions and the diagonal directions is obtained. Although obtained, these absolute values | f2−f8 || f4−f6 |... | F3−f7 | indicate the magnitudes of changes in the gray level at the edge of the black spot defect G or the white spot defect.
The light and shade extraction area H uses, for example, 3 × 3 pixel areas f1 to f9, and therefore indicates the magnitude of the light and shade level change in the three pixels. The maximum value of the absolute values in the up / down / left / right directions and the left / right diagonal directions is set as the edge strength F of the target point because the change in the gray level in the direction close to the vertical direction with respect to the contour line Lg is the largest. . Therefore, a large change in the shading level indicates that the intensity of the black spot defect G or the white spot defect, that is, the spot intensity is high.

FIG. 8 shows an example in which the density extraction area H is applied to a black spot defect G, for example. The change Gb of the light and shade level on one line Ga of the black spot defect G is shown. For example, when the absolute value | f4−f6 | of each gray level of the pixel areas f4 and f6 existing in the left-right direction is obtained, the absolute value | f4−f6 | This represents the steepness of each gray level of f6.
Accordingly, when the change Gb of the darkness level Gb on one line Ga of the black spot defect G is compared with the change Gc of the lightness level, the absolute value | f4−f6 | indicating the magnitude of the change Gb of the lightness level is greater. It becomes larger than the absolute value | f4−f6 | indicating the magnitude of the level change Gc. As a result, the edge strength F of the black spot defect G having the density level change Gb is higher than the edge strength F of the black spot defect G having the density level change Gc.

  The density extraction area H uses, for example, 3 × 3 pixel areas f1 to f9. However, the present invention is not limited to this. For example, as shown in FIG. 9, the density extraction area is formed of a matrix having a 5 × 5 pixel area. Ha may also be used. Also in this shading extraction area Ha, the central pixel area f5 is taken as a point of interest, and a plurality of pixel areas f2, f8, f4, f6 and left and right diagonals existing at each of the vertical and horizontal ends of the pixel area f5. A plurality of pixel areas f1, f9, f3, and f7 existing at each end in the direction are set as peripheral points. Since the light and shade extraction area Ha has a 5 × 5 pixel area, it depends on the number of pixels from which a change in light and shade level is extracted as compared to the light and shade extraction area H having 3 × 3 pixel areas f1 to f9. The width becomes longer. The density extraction area Ha is suitable for acquiring the edge strength of a black spot defect G or white spot defect having a large area, for example. The edge strength F at each point of interest is acquired according to the above equation (1). Also, the pixel area between the pixel area f1 and the pixel area f2 is f12, the pixel area between the pixel area f8 and the pixel area f9 is f89, and the absolute value of the difference between the light and shade levels of the pixel area f12 and the pixel area f89. | F12−f89 | may be obtained. Similarly, the edge strength F may be obtained by obtaining each absolute value of the difference in light and shade level between the pixel areas between the pixel areas f1 to f9 and taking these absolute values into consideration.

  The notification unit 24 displays on the display 17 the evaluation result of the black spot defect G and the white spot defect acquired by the defect evaluation unit 23, that is, the final edge strength Fa. The notification unit 24 displays the image data of the output phosphor screen 1 acquired by imaging of the imaging unit 4 and indicates the location of the black spot defect G or the white spot defect on the image data. The evaluation result of the defect G or white spot defect is displayed on the display 17.

FIG. 10 shows the image data Q of the output phosphor screen 1 displayed on the display 17 and black spot defects G displayed on the image data Q, for example, four black spot defects G1 to G4. The display 17 also displays enlarged data B1 to B4 of the four black spot defects G1 to G4.
FIG. 11 shows an evaluation result column J including the final edge strength Fa of the black spot defect G displayed on the display 17. The evaluation result column J displays, for example, the size, area, and final spot strength of the circumscribed rectangle of the four black spot defects G1 to G4. The size of the circumscribed rectangle is the length in the XY direction of the rectangle when each black spot defect G1 to G4 is surrounded by a rectangular frame in contact with the outer periphery of each black spot defect G1 to G4. The area is the area of each of the four black spot defects G1 to G4. The respective areas of the black spot defects G1 to G4 are calculated by the CPU 11 counting the number of pixels present in the black spot defects G1 to G4, respectively. The final spot strength is the final edge strength Fa acquired by the defect evaluation unit 23.
For example, the four black spot defects G1 to G4 displayed on the image data Q may be distinguished and displayed by a color corresponding to the final edge strength Fa.

Next, the operation of the apparatus configured as described above will be described according to the defect inspection flowchart shown in FIG.
The electron beam source 2 emits an electron beam. The deflection yoke 3 raster scans the electron beam emitted from the electron source 2 over the entire surface of the output phosphor screen 1 as shown in FIG. When the electron beam is raster scanned on the output phosphor screen 1, the phosphor on the output phosphor screen 1 emits fluorescence. If there is a spot defect portion on the output phosphor screen 1, the fluorescence becomes non-uniform at the spot defect portion.
In step # 1, the imaging unit 4 captures an optical image generated on the entire surface of the output phosphor screen 1 when the electron beam is raster scanned over the entire surface of the output phosphor screen 1, and outputs the image signal. The image signal output from the imaging unit 4 is sent to the image processing apparatus 10.

The image input unit 12 of the image processing apparatus 10 receives the image signal output from the imaging unit 4 and acquires image data including the entire surface of the output phosphor screen 1 from the image signal.
The defect extraction unit 21 extracts a spot defect on the output phosphor screen 1 from the image data acquired by the imaging of the imaging unit 4. That is, in step # 2, the defect extraction unit 21 binarizes image data including the entire surface of the output phosphor screen 1 using the threshold value for black spot defect extraction or the threshold value for white spot defect extraction. In # 3, for example, a black spot defect G is extracted. In the image data D1 shown in FIG. 13, the black spot defect G exists on the image data Q of the portion of the output phosphor screen 1. The defect extraction unit 21 acquires the image data D2 from which the black spot defect G is extracted by binarizing the image data D1 with a threshold for black spot defect extraction.

The contour acquisition unit 22 acquires a contour line Lg along the contour of, for example, the black spot defect G extracted by the defect extraction unit 21 in step # 4.
Specifically, the contour acquisition unit 22 scans the contour extraction area EC shown in FIG. 5A in the X direction with respect to the binarized image data. For example, as shown in FIG. When the binarization level “1” is read in the area e5, the pixel position S1 corresponding to the pixel area e5 is determined as the contour portion of the black spot defect G.
The contour acquisition unit 22 has, for example, each pixel area e2, e1, e4, e7, e8, e9, e6 around the pixel area e5 with the pixel area e5 (pixel position S1) read from the binarization level “1” as the center. , E3 read the binarization level “1” or “0” of each pixel.

  The contour acquisition unit 22 reads each of the surrounding pixel areas e2, e1,..., E3, for example, when the binary level “1” is read in the pixel area e8, the pixel position S2 corresponding to the pixel area e8 is displayed as a black spot. The outline of the defect G is determined.

  The contour acquisition unit 22 moves the pixel area e5 at the center of the contour extraction area EC to the pixel position S2, and after this movement, the pixel area e2, e1,. , E3 read the binarization level “1” or “0” of each pixel.

Thereafter, the contour acquisition unit 22 reads the binarization level “1” or “0” of each pixel area e2, e1,. When the conversion level “1” is read, it is repeatedly determined that the pixel positions S3, S4,..., Sk corresponding to the pixel area are the contour portions of the black spot defect G.
As a result, the contour obtaining unit 22 obtains the contour line Lg of the black spot defect G by continuing each pixel position S1, S2,..., Sk determined as the contour part of the black spot defect G as shown in FIG. To do. The contour acquisition unit 22 stores the position information of the pixel positions S1, S2,..., Sk forming the contour line Lg of the black spot defect G, for example, in the data memory 14 as contour information.

  In step # 5, the defect evaluation unit 23 determines the magnitude of the change in all shade levels on the contour line Lg acquired by the contour acquisition unit 22, that is, the edge portion of the black spot defect G corresponding to the contour line Lg. The magnitude of the change in the shade level is acquired, and the black spot defect G is evaluated based on the magnitude of the change in the shade level.

  Specifically, the defect evaluation unit 23 causes the pixel areas f2, f8,..., F7 of the surrounding points to follow the pixel area f5 of the point of interest in the shade extraction area H as shown in FIG. Of each of the surrounding points facing each other through the pixel area f5 of the target point, for example, the absolute value | f2−f8 | of each gray level of each of the pixel areas f2 and f8 existing in the vertical direction according to the above formula (1) The absolute value | f4−f6 | of each gray level of the pixel areas f4 and f6 existing in the left-right direction and the absolute value | f1−f9 | of each gray level of the pixel areas f1 and f9 present in the left diagonal direction. Then, absolute values | f3−f7 | of the gray levels of the pixel areas f3 and f7 existing in the diagonally right direction are respectively obtained, and the maximum value of these absolute values is set as the edge strength F of the target point.

  As shown in FIG. 6, the defect evaluation unit 23 sequentially uses the pixel positions S1, S2,..., Sk on the contour line Lg as points of interest, and acquires the edge strengths F at these points of interest. The defect evaluation unit 23 averages the edge strengths F at the pixel positions S1, S2,..., Sk on the contour line Lg, and acquires the averaged value as the final edge strength Fa of the black spot defect G. .

  In step # 6, for example, the defect evaluating unit 23 compares the final edge strength Fa with a plurality of predetermined edge strengths for determination, and determines the level of the final edge strength Fa. The defect evaluation unit 23 compares the final edge strength Fa with a predetermined edge strength for determination. If the final edge strength Fa is higher than the determination edge strength, the defect evaluation unit 23 compares the final edge strength Fa with the black value having the final edge strength Fa. The spot defect G may be determined as a black spot defect G having an intensity that cannot be used as the output phosphor screen 1 of the X-ray image intensifier.

  In step # 7, the notification unit 24 displays on the display 17 the evaluation result of the black spot defect G and the white spot defect acquired by the defect evaluation unit 23, that is, the final edge strength Fa. For example, as shown in FIG. 10, the notification unit 24 displays image data Q of the output phosphor screen 1 on the display 17 and black spot defects G displayed on the image data Q, for example, four black spot defects G1 to G1. G4 and the enlarged data B1 to B4 of the black spot defects G1 to G4 are displayed. The notification unit 24 displays the evaluation result column J on the display 17 as shown in FIG. 11, and the size and area of the circumscribed rectangle of, for example, the four black spot defects G1 to G4 are displayed in the evaluation result column J. And the final spot intensity Fa are displayed. For example, the four black spot defects G1 to G4 displayed on the image data Q may be distinguished and displayed by a color corresponding to the final edge strength Fa.

  As described above, according to the one embodiment, the black spot defect G or the like in the output phosphor screen 1 is extracted from the image data acquired by the imaging unit 4, and the contour of the black spot defect G or the like is extracted. The contour line Lg is acquired, and the magnitudes of changes in all the light and shade levels on the contour line Lg, that is, the magnitudes of changes in the light and shade levels in the edge portion such as the black spot defect G corresponding to the contour line Lg are obtained. The final spot intensity Fa of the black spot defect G or the like is obtained based on the magnitude of the change in the shade level, and the black spot defect G or the like is evaluated. Thereby, the black spot defect G and the like can be displayed on the display 17 as the final spot intensity Fa obtained by quantification. The determination result of the black spot defect G or the like does not differ depending on the inspector, and the inspection result of the black spot defect G or the like can be obtained stably.

  In order to extract the magnitude of the change in the light and shade level at the edge portion of the black spot defect G or the like, for example, the light and shade extraction area H as shown in FIG. Let the maximum value be the edge strength F of the point of interest. In this grayscale extraction area H, for example, the absolute value | f4−f6 | of each gray level of each pixel area f4, f6 existing in the left-right direction represents the steepness of each gray level of each pixel area f4, f6. Therefore, the strength of the black spot defect G or the like can be obtained by obtaining the edge strength F of the point of interest. The intensity of the black spot defect G or the like is not related to the size of the area of the black spot defect G or the like, but if the intensity is high even if the area is small, the black spot defect G or the like is output fluorescence of the X-ray image intensifier. Indicates that it cannot withstand use as surface 1.

  Therefore, as shown in FIG. 8, when the density level change Gb on one line Ga of the black spot defect G is compared with the density level change Gc, the absolute value | f4−f6 indicating the magnitude of the density level change Gb. | Is larger than the absolute value | f4−f6 | indicating the magnitude of the change Gc of the light and shade level. As a result, the edge strength F of the black spot defect G having the density level change Gb is higher than the edge strength F of the black spot defect G having the density level change Gc. A black spot defect G having a change Gb in gray level indicates that it cannot be used as the output phosphor screen 1 of the X-ray image intensifier.

Note that the present invention is not limited to the above embodiment, and may be modified as follows.
The inspected object 1 is, for example, an output fluorescent screen 1 provided in an X-ray image intensifier. However, the present invention is not limited to this, and it can also be applied to inspect uneven defects existing on a liquid crystal panel or a painted surface. .
The defect extraction unit 21 uses the threshold value for binarization to extract a spot defect on the output phosphor screen 1, but is not limited thereto. For example, spot defect extraction uses a local ternary method. As for the spot defect, the black spot defect is divided into three values of black (darkness level 0), the white stain defect is white (darkness level 255), and the non-stained portion is gray (lightness level 128).
Image data D1 including the entire surface of the output fluorescent screen 1 is acquired from the image signal output from the imaging unit 4.
In this image data D1, as shown in FIG. 14, a target pixel N for spot inspection and a target region K including the target pixel are set.
From the target pixel N and the average value M of the target region K, an output value W, that is, three values of black (lightness level 0), white (lightness level 255), and gray (lightness level 128) is calculated. Let L be a fixed value.
M−N <L → W = 255
| M−N | ≦ L → W = 128
M−N> L → W = 0
The target area K is moved to the entire surface of the image data D1, and the entire output value W of the image data D1 is acquired. Thereby, a spot defect can be extracted from the output value W.

  According to the embodiment as described above, the defect inspection apparatus capable of stably obtaining the inspection result of the black spot defect G or the like without the determination result of the black spot defect G or the like being different depending on the inspector. Can be provided.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  Hereinafter, other features of the embodiment will be described.

(1) The defect evaluation unit acquires the magnitudes of changes in all shade levels on the contour line acquired by the contour acquisition unit.

(2) The defect evaluation unit acquires the magnitude of the change in the gray level at the edge portion of the defect portion corresponding to the contour line.

(3) If the defective part is a defect due to a stain, the defect evaluation unit obtains a value obtained by averaging the magnitude of the change in the light and dark levels as the strength of the stain.

(4) The defect evaluation unit has a density extraction area including a target point and a plurality of surrounding points that exist in both the vertical and horizontal directions and the left and right diagonal directions of the target point. As the intensity on the contour line at the target point, the absolute value of the difference between the gray levels of the peripheral points facing each other through the target point among the plurality of peripheral points is made to follow the target point on the contour line. It is characterized in that it is obtained sequentially, and the absolute values of all the points of interest on the contour line are averaged and acquired as the intensity of the stain.

(5) The light and shade extraction area is characterized by having a matrix having a plurality of extraction areas in the vertical and horizontal directions for extracting the light and shade levels of the target point and the plurality of surrounding points.

(6) It has a report part which reports the evaluation result of the defective part acquired by the defect evaluation part.

(7) The notifying unit displays the image data of the object to be inspected acquired by imaging of the imaging unit, indicates the location of the defective portion on the image data, and displays the evaluation result of the defective portion. It is characterized by.

(8) The object to be inspected is an output fluorescent screen provided in the X-ray image intensifier, and the imaging unit captures an optical image generated on the output fluorescent screen when the output fluorescent screen is irradiated with an electron beam. The contour acquisition unit acquires a contour line along the contour of the spot defect portion on the output phosphor screen as the contour information, and the defect evaluation unit acquires all the spot defect portions corresponding to the contour line acquired by the contour acquisition unit. It is characterized in that the magnitude of the change in the light and shade level at the edge portion is acquired.

  1: Inspected object (output phosphor screen), 2: electron beam source, 3: deflection yoke, 4: imaging unit, 10: personal computer (image processing apparatus) for image processing, 11: CPU, 12: image input unit 13: Program memory, 14: Data memory, 15: Output unit, 16: Operation input unit, 17: Display, 18: Keyboard, 21: Defect extraction unit, 22: Outline acquisition unit, 23: Defect evaluation unit, 24: Notification section.

Claims (5)

An imaging unit for imaging the inspected object;
A defect extraction unit that extracts a defect portion in the inspection object based on image data acquired by imaging of the imaging unit;
An outline acquisition unit for acquiring an outline of the defect portion extracted by the defect extraction unit;
A defect evaluation unit that evaluates the defect portion based on a magnitude of a change in a light and shade level in the contour portion acquired by the contour acquisition unit;
A defect inspection apparatus comprising:   The defect inspection apparatus according to claim 1, wherein the contour acquisition unit acquires a contour line along the contour of the defect portion as the contour portion.   The defect inspection unit according to claim 2, wherein the defect evaluation unit averages the magnitudes of changes in all the shade levels on the contour line, and acquires the averaged value as an evaluation of the defect portion. apparatus. The subject is imaged by the imaging unit,
Storing image data acquired by imaging of the imaging unit in a memory by a computer;
Based on the image data stored in the memory, the computer extracts a defective portion in the inspection object,
Obtaining an outline of the extracted defect portion by the computer;
Performing an evaluation of the defective portion by the computer based on the magnitude of the change in the shade level in the acquired contour portion;
A defect inspection method characterized by that. On the computer,
An extraction function for extracting a defective portion in the inspected object based on image data acquired by imaging of the imaging unit;
A contour acquisition function for acquiring a contour portion of the defect portion extracted by the extraction function;
An evaluation function for acquiring an evaluation of the defective portion based on a magnitude of a change in a light and shade level in the contour portion acquired by the contour acquisition function;
Defect inspection program to realize.

Images