JP3311135B2 - Inspection range recognition method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for recognizing an inspection area such as a surface inspection apparatus having a visual recognition mechanism.

[0002]

2. Description of the Related Art Conventionally, this type of recognition method is disclosed in Japanese Unexamined Patent Publication No.
Japanese Patent Application Laid-Open No. 3-112145 (first conventional example) discloses a method for matching the coordinate system of a wafer to be inspected with the coordinate system of an inspection apparatus.
Is known. In the first conventional example,
The position recognition mark provided on a part of the inspection target is moved to three positions, and the coordinates of the device and the inspection target are corrected by recognizing the respective coordinates. In the second conventional example, from two points on an orientation flat provided on a wafer,
Move the wafer so that the orientation flat matches the X axis, calculate the wafer center and radius from the three points on the outer edge of the wafer, and specify the wafer position by positioning the Y axis in contact with the outer circumference I do.

[0003]

SUMMARY OF THE INVENTION However, such a first
In the related art, there is a problem that it is impossible to specify an inspection area for an inspection target to which a mark serving as a mark is not attached. In the second conventional example, there is a problem that an end cannot be specified for an inspection object whose surface is not a mirror surface. SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and has as its object to provide an inspection range recognition method capable of recognizing an inspection range even if the inspection target does not have a site such as a polygonal mark.

[0004]

Accordingly, the present invention is a method for recognizing an inspection range of an inspection target to be inspected together with a substrate, and imaging the inspection target and determining an average density at a central portion of the obtained captured image. Obtaining, and setting a continuous rectangular area from the center so that the boundary of the inspection range is included in any of the areas; obtaining the average density of each rectangular area; A step of comparing the average density of the central portion with the average density of the center portion, a step of determining a rectangular area in which a difference equal to or more than a predetermined value is first detected as a boundary area, and a step of obtaining reference point coordinates in the boundary area, The above problem is solved by providing an inspection range recognizing method characterized by comprising a step of recognizing an inspection range based on a plurality of reference point coordinates obtained by repeating.

[0005]

The inspection object is moved by the transport unit, and an image of the vicinity of the center of the inspection object is taken. Then, the captured image data is input to the image processing unit, an average density value of the image data is calculated, and a reference density area value having an allowable range around the average density value is set. Then, the inspection object is moved,
The vicinity of one end of the inspection target is imaged and input to the image processing unit. A part of the input image data is divided into a plurality of areas continuous from the center side of the inspection target to the end, and an average density value of the image data in the area is obtained. The average density value in the area and the set reference density area value are compared, and if the average density value in the area is out of the standard to outside the standard, the coordinates of the area are compared with points on one side of the inspection target. Be recognized. When a point on one side is recognized, the vicinity of the other end of the same side is imaged, and the above operation is repeated, and points on a plurality of sides are recognized for one side. A regression line is determined and recognized from a plurality of points on the side that is recognized for one side of the inspection target. Similarly, the operation of recognizing the regression line is repeated for all sides of the inspection target, and a region surrounded by all the recognized regression lines is recognized as the inspection range.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic longitudinal sectional view of the embodiment, FIG. 2 is an explanatory view showing an imaging state, FIG. 3 is an explanatory view showing an imaging state near an end portion, and FIG. FIG. 5 is an explanatory diagram showing the processing of the image data represented by 3; FIG. 5 is a graph showing a change in the density of the captured image data;
Is a graph showing the average density change of each area of the captured image data.

Reference numeral 1 denotes a surface inspection device. Surface inspection device 1
As shown in FIG. 1, is composed of a transport unit 2, a glass plate 3, an illuminating unit 4, an imaging unit 5, and an image processing unit 6. The surface inspection device 1 images the surface state of the inspection target W, and inspects the surface state of the inspection target W based on the captured image. Transport unit 2
Can place the glass plate 3 as a substrate and move the glass plate 3 in the plane direction. Further, the transport unit 2 outputs a signal indicating the movement position to the control unit 3. In this embodiment, the glass plate 3 is placed in the horizontal direction, and as shown in FIG.
Is the origin O of the XY coordinates, and the X axis and the Y axis, respectively. The inspection target W is attached to one surface of the glass plate 3 and is placed on the transport unit 2. In this embodiment, the inspection object W is a polarizing film having a rectangular shape, which is generally of a reflection type and has an aluminum sheet adhered to one surface, has a relatively high reflectance, and is previously adhered to the glass plate 3. Heat treatment. Inspection object W on glass plate 3
Is always kept at substantially the same position by an inspector or the like, but the relative position between the inspection target W and the glass plate 3 is shifted at the time of attachment.

The illuminating section 4 includes a light generating section 41 for generating light for irradiating the inspection target W, and an irradiation port 4 for irradiating the inspection target W.
2. It comprises an optical fiber 43 for guiding illumination light to an irradiation port 42, and irradiates the inspection object W. The irradiation port 42 has a donut shape having a light emission port on the side of the inspection target W, and can irradiate the inspection target W approximately at an average. The light emitted from the irradiation port 42 passes through the glass plate 3, is reflected on the inspection target W, passes through the glass plate 3 again, and proceeds in the irradiation direction. The imaging unit 5 captures light emitted from the irradiation port 42 and reflected on the inspection target W,
The image recognized as the shade is output to the image processing unit 6 as image data. The imaging unit 5 is installed in the center gap of the irradiation port 42 in the direction of the inspection target W, and as illustrated in FIG. 2, a part of the inspection target W is defined as an imaging region p, q _{1 to} q ₁₂ as illustrated in FIG. 3. Image data can be captured. The magnification for determining the imaging region p of the imaging unit 5 can be changed to a magnification suitable for confirming the inspection target W and can be changed to the most efficient inspection magnification depending on the type of the inspection target W and the like. Be composed. In this embodiment, the imaging unit 5 is composed of a CCD camera, and sequentially outputs the density of each pixel in the imaging area p to the image processing unit 6 as image data.

[0009] The image processing unit 6 outputs an image output from the imaging unit 5.
Enter the data. Further, the image processing unit 6 includes an XY plane
The movement amount and position of the inspection target W in the
The signal is recognized and detected as an XY coordinate by the symbol. And the picture
The image processing unit 6 moves the inspection target W by the transport unit 2.
Output signal. Therefore, under the control of the image processing unit 6,
The sending unit 2 places the inspection target W at the position of the designated XY coordinate.
Can be moved. The transport set in advance by the image processing unit 6
The moving position of the feeding unit 2 is the inspection target attached to the glass plate 3
A position p corresponding to substantially the center of W and a right side l₁Position q₁~ Q
_Three, Left side l_TwoPosition q_Four~ Q₆, Upper side l_ThreePosition q₇~ Q₉,
And lower side l_FourPosition q_Ten~ Q₁₂The position of each.
Each position q₁~ Q₁₂Each image data captured by the imaging unit 5 at
The data is stored in a memory (not shown). And the middle position
From the image data taken at the position p
Determine the density and give a preset density width to the reference density area
Set as a value. The position q of the side shown in FIG.₁From the figure
The position q of the side stored as represented in FIG.₁Imaged at
Part of the image data, the X direction centered on the Y = y1 position
The pixels of the strip-shaped region 7 continuous in the direction are divided in the Y direction, respectively.
It is divided into a book area 8. And the strip on the center position p side
The average density value is obtained sequentially from the shape area 81, and
The value is compared with the obtained reference density area value. Flat of strip area 8
The average density value changes from within the reference density area value to outside the reference density area value.
XY coordinates of the center of the strip-shaped area 8i
Right side l₁Is stored as an edge point which is a point on the side of. same
Thus, position q_Two, Q_ThreeXY coordinates of the inspection area
Right side l₁Are set as edge points. Then set
XY coordinates by statistical processing from the coordinates of the multiple edge points
Find the regression line on the benchmark, right side l of the inspection area₁Set as
In this embodiment, the least squares are calculated from the XY coordinates of each edge point.
Find a regression line by the method. Left side l_TwoPosition q_Four~ Q ₆,
Upper side l_ThreePosition q₇~ Q₉, And lower side l_FourPosition q_Ten~ Q
₁₂Of the right side l₁Position q₁~ Q_ThreeSet similarly
It is. The average density value of all strip-shaped areas is the reference density area
If the value is within the value, the imaging position is set to the position q by the transport unit 2.₁Or
Position q in the direction opposite to the center position p₁With the imaging screen next to
Move to position q₁The same operation as in the strip area
Until the average density value becomes equal to or less than the reference density area value.
The average density value of all strip-shaped areas 8 is out of the reference density area value
The position q₁To the transport section 2 in the center position p direction
Twist position q₁Move so that it is the imaging screen adjacent to
q₁The same operation as above is based on the average density value of the strip-shaped area 8.
Repeat until the value falls within the density area value.

The position and direction of the strip-shaped area 7 which is a part of the image data, and the dividing direction of the strip-shaped area 8 are preset in the image processing section 6 for each side of the inspection object W. In this embodiment, in the right side l ₁ and the left side l ₂ , the band-shaped region 7 continuing left and right at the center of the image region is divided into a plurality of strip-shaped areas 8 extending horizontally in the same direction. In the upper side l ₃ and the lower side l ₄ , a vertically continuous strip-shaped area 7 at the center of the image area is divided into a plurality of vertically-long strip-shaped areas 8. In this embodiment, the position of the band-shaped region 7 is set at the center of the imaging screen, but another position may be set as the band-shaped region 7. Furthermore, the strip-shaped area 8
The area is 50 pixels × 10 pixels, but the pixel ratio of each side and the number of pixels in the strip-shaped area can be changed depending on the inspection target or the like. As described above, the surface inspection unit 1 recognizes a region surrounded by the regression line of the set side as an inspection region,
The inspection of the surface state of the inspection target W is performed efficiently.

Next, the operation of the embodiment will be described. The glass plate 3 on which the inspection target W is adhered by the inspector is placed on the transport unit 2. Next, the transport unit 2 moves to the preset position p.
Is moved to the inspection target W. When the inspection target W moves to the position p, the imaging unit 5 captures an image, and the image processing unit 6 receives a signal of the XY coordinate position of the position p, and inputs image data at the position p captured by the imaging unit 5. Then, the average density is obtained and a reference density area value is set. Then, the conveyance unit 2 moves the inspected W to the position q _1, the image processing section 6 the image data of the same positioned q ₁ receives. At this time, the vertical axis is the density value,
The horizontal axis is the X coordinate, and as shown in FIG. 5 showing the change in density at Y = y1 in FIG. 3, the density value near the edge portion is affected by the glue 10 which has protruded from the inspection target W attached. Is scattered.

[0012] The image processing unit 6, among the pixels of the image data of the position q _1, the data of the band-shaped area 7 strip areas 8
An average density value of 1 is obtained, and the previously set reference density area value and the average density value of the strip-shaped area 81 are compared. When the average density value of the strip-shaped area 81 is within the reference density area value,
The average density value of the next strip-shaped area 82 is obtained and compared in the same manner, and the process is repeated until the average density value of the strip-shaped area becomes equal to or less than the reference density area value. That is, the vertical axis is the density, the horizontal axis is the strip-shaped area 8, the reference density area value is V, and as shown in FIG. 6 showing changes in the average density values 91 to 9n of the strip-shaped areas 81 to 8n. setting the X-Y coordinate of the strip-shaped area 8i center of the average density value 9i became less than the reference density area value and edge point to Q ₁ right l _1. Looking at the transition of the average density value at this time, as shown in FIG. 6, the average density value does not show a rapid change unlike the density value at each position shown in FIG. 5 in the average density value. Similarly, the edge point Q at the positions q ₂ and q ₃
Setting the _2, Q _3. Then, to calculate the regression line from the set edge points Q ₁ to Q _3, it sets the right-hand side l _1. Similarly, work is performed on the left side l ₂ , the upper side l ₃ , and the lower side l ₄ of the inspection target W, and the obtained edge points Q _{4 to} Q ₆ , Q ₇ to
A regression line is calculated from Q ₉ , Q _{10 to} Q ₁₂ , and each side l _{2 to} l ₄
Is set, and an area closed by each side l _{1 to} l ₄ is recognized as an inspection area.

[0013]

Therefore, according to the present invention, it is possible to obtain the points on the sides while avoiding the adverse effects of noise, and to cancel the errors in obtaining the points on each side to accurately determine the side to be inspected. The inspection area can be recognized even for an inspection target having no mark as a mark, and a point on a side can be specified even for an inspection target other than a mirror surface. Therefore, the inspection area can be easily recognized.

[Brief description of the drawings]

FIG. 1 is a schematic longitudinal sectional view illustrating an embodiment.

FIG. 2 is an explanatory diagram showing an imaging state.

FIG. 3 is an explanatory diagram showing a state in which an image of the vicinity of an end is captured.

FIG. 4 is an explanatory diagram showing processing of image data shown in FIG. 3;

FIG. 5 is a graph showing a change in density of captured grayscale image data.

FIG. 6 is a graph showing an average density change of each area of captured grayscale image data;

[Explanation of symbols]

1 surface inspection unit 2 transport unit 3 glass plate 4 illumination unit 41 the light generator 42 irradiation port 43 optical fibers 5 imaging section 6 the image processing section 7 band region 81~8n strip area 91~9n average density value 10 glue l _1, l _2, l _3, l ₄ sides p position W inspected center Q ₁ to Q ₁₂ edge points q ₁ to q ₁₂ side

Claims (1)

(57) [Claims] 1. A method for recognizing an inspection range of an inspection object to be inspected together with a substrate, comprising: a step of imaging the inspection object and calculating an average density at a central portion of an obtained captured image; Setting a continuous rectangular area from the center so as to be included in any one of the areas; and calculating the average density of each rectangular area and comparing the average density of the rectangular areas closer to the center with the average density of the center in order. And a step of determining a rectangular area in which a difference equal to or more than a predetermined value is detected for the first time as a boundary area as a result of the comparison and obtaining coordinates of a reference point in the boundary area. Recognizing an inspection range based on point coordinates.