Image formation method and correction method

Abstract

To provide an image forming method with which a corrected image can be more appropriately formed.SOLUTION: An image forming method includes: an image for correction forming step of forming an image for correction by sequentially picking up images of a workpiece by a camera while moving a holding table holding the workpiece and the camera relatively not along a first direction and a second direction; a correction value acquisition step of acquiring a plurality of correction values based on the brightness of a plurality of pixels constituting the image for correction; a pre-correction image forming step of forming a pre-correction image by sequentially picking up images of the workpiece by the camera while moving the holding table and the camera relatively along a direction perpendicular to a linear imaging area of the camera; and a post-correction image forming step of forming a post-correction image by correcting the brightness of a plurality of pixels constituting the pre-correction image by using the plurality of correction values.SELECTED DRAWING: Figure 7

Description

Technical Field

[0001] The present invention relates to an image forming method capable of forming an image suitable for inspection of a workpiece, and a correction method used when correcting an image.

Background Art

[0002] In electronic devices typified by mobile phones and personal computers, a device chip having devices such as electronic circuits is an essential component. A device chip is obtained, for example, by partitioning the surface side of a wafer made of a material such as silicon (Si) into a plurality of regions by a linear street (division planned line), forming devices in each region, and then dividing the wafer along this street.

[0003] When dividing a wafer into a plurality of device chips, for example, a disk-shaped tool called a cutting blade in which abrasive grains are dispersed in a bonding material is used. By rotating the cutting blade at high speed and cutting into the street of the wafer while supplying a liquid such as pure water, the wafer is cut along the street, and a plurality of device chips are obtained (see, for example, Patent Document 1).

[0004] By the way, when a wafer is divided into a plurality of device chips, there is a possibility that a breakage called chipping occurs in the device chip due to contact with a tool or the like. When the number of chippings occurring in the device chip increases, the strength of the device chip is significantly reduced, and the quality required for the product cannot be obtained. Therefore, after the wafer is divided into a plurality of device chips, the quality is inspected based on an image obtained by imaging the wafer.

[0005] In this quality inspection, a camera is typically used that collects light reflected from the wafer being inspected using a lens and places it on an image sensor to form an image. However, with this type of camera, if the ambient light level is insufficient, for example, uneven brightness can appear in the image, easily reducing the accuracy of the inspection. Therefore, to compensate for the image quality degradation caused by uneven brightness, a correction called shading correction is performed (see, for example, Patent Document 2).

[0006] In shading correction, for example, a brightness correction value is calculated from an image obtained by imaging the top surface of a table used to hold wafers, in order to compensate for the reduced image quality caused by variations in brightness. By applying this correction value to an image obtained by imaging the wafer to be inspected, a certain degree of variations in brightness are removed from the image, maintaining a high level of inspection accuracy. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 3-198363 [Patent Document 2] Japanese Patent Publication No. 2020-57985 [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] However, with the method described above, the correction value is calculated from the image obtained by imaging the top surface of the table, so images obtained by imaging a wafer with different optical properties than the table are not necessarily properly corrected. In order for the image obtained by imaging the wafer to be inspected to be properly corrected, the correction value needs to be calculated from the image obtained by imaging this wafer.

[0009] On the other hand, when imaging a wafer to form an image, a camera containing a one-dimensional image sensor, such as a line sensor, is sometimes used. In this case, for example, by continuously imaging the wafer with the camera while moving the camera's linear imaging area and the wafer to be inspected relatively and linearly in a direction perpendicular to the linear imaging area, a strip-shaped image showing the desired range can be obtained.

[0010] However, as mentioned above, the wafer being inspected has linear processing marks along the street. Therefore, when the wafer being inspected is imaged with the camera while the camera's imaging area is moved relatively and linearly, it is sometimes impossible to distinguish between the light and dark areas and the processing marks in the resulting image, making it impossible to calculate an appropriate correction value.

[0011] Therefore, the object of the present invention is to provide an image forming method that can form a more appropriately corrected image, and a correction method that can correct an image more appropriately. [Means for solving the problem]

[0012] According to one aspect of the present invention, an image forming method is provided for forming an image of a plate-shaped workpiece by imaging the surface side of the workpiece, which has been processed along one or both of a linear first division line along a first direction and a linear second division line along a second direction intersecting the first direction, using a camera including a one-dimensional image sensor, comprising: a holding step of holding the workpiece on a holding table; a corrective image forming step after the holding step, in which the holding table and the camera are moved relative to each other so as not to be along the first and second directions, while sequentially imaging the workpiece with the camera to form a corrective image; and the brightness of a plurality of pixels constituting the corrective image An image forming method is provided, which includes: a correction value acquisition step of acquiring a plurality of correction values ​​used to correct the brightness of a plurality of pixels constituting an image formed by the camera based on a degree; a pre-correction image forming step of forming a pre-correction image by sequentially imaging the workpiece with the camera while moving the holding table and the camera relatively along a direction perpendicular to the linear imaging area of ​​the camera after the holding step; and a post-correction image forming step of forming a post-correction image by correcting the brightness of a plurality of pixels constituting the pre-correction image using the plurality of correction values ​​after the correction value acquisition step and the pre-correction image forming step.

[0013] According to another aspect of the present invention, a correction method is provided for use when forming an image of a plate-shaped workpiece by imaging the surface side of the workpiece, which has been processed along one or both of a linear first division line along a first direction and a linear second division line along a second direction intersecting the first direction, with a camera including a one-dimensional image sensor, the correction method comprising: a holding step of holding the workpiece with a holding table; a correction image forming step of forming a correction image by sequentially imaging the workpiece with the camera while moving the holding table and the camera relatively after the holding step so as not to be along the first direction and the second direction; a correction value acquisition step of acquiring a plurality of correction values ​​used when correcting the brightness of a plurality of pixels constituting an image formed by the camera based on the brightness of a plurality of pixels constituting the correction image; and a correction step of correcting the brightness of a plurality of pixels constituting an image formed by the camera using the plurality of correction values ​​after the correction value acquisition step.

[0014] In each aspect of the present invention, the corrective image forming step involves moving the holding table and the camera relative to each other, for example, so that the imaging area of ​​the camera moves across the surface of the workpiece in a linear trajectory that intersects the first and second directions. Alternatively, the corrective image forming step involves moving the holding table and the camera relative to each other so that the imaging area of ​​the camera moves across the surface of the workpiece in a spiral trajectory. [Effects of the Invention]

[0015] In the image forming method according to one aspect of the present invention, and the correction method according to the other aspect, a correction image is formed by sequentially imaging the workpiece with the camera while moving the holding table and the camera relatively so as not to follow the first and second directions. Therefore, when forming this correction image, the first and second division lines are not imaged using only a part of the camera's linear imaging area, nor are only the first and second division lines imaged using the entire camera.

[0016] That is, since the first division planned line and the second division planned line do not appear only in a narrow area of a part of the correction image, or only the first division planned line and the second division planned line appear in the correction image, correction values suitable for correcting the luminance of a plurality of pixels constituting the image can be obtained from this correction image. Thus, according to each aspect of the present invention, an image forming method capable of forming a more appropriately corrected image and a correction method capable of more appropriately correcting an image are provided.

Brief Description of the Drawings

[0017] [Figure 1] FIG. 1 is a perspective view showing an example of an inspection apparatus. [Figure 2] FIG. 2 is a plan view showing an example of a plate-shaped workpiece. [Figure 3] FIG. 3 is a diagram schematically showing the arrangement of a plurality of light receiving elements in an image sensor. [Figure 4] FIG. 4 is a plan view schematically showing how an image is formed by a camera. [Figure 5] FIG. 5 is a diagram schematically showing the luminance of each pixel of the formed pre-correction image. [Figure 6] FIG. 6 is an example of the formed pre-correction image. [Figure 7] FIG. 7 is a plan view schematically showing how a correction image is formed. [Figure 8] FIG. 8 is a diagram schematically showing the luminance of each pixel of the formed correction image. [Figure 9] FIG. 9 is a diagram schematically showing the bright and dark spots removed by applying correction values. [Figure 10] FIG. 10 is a diagram schematically showing the luminance of each pixel of the formed post-correction image. [Figure 11] FIG. 11 is an example of the formed post-correction image.

Modes for Carrying Out the Invention

[0018] Embodiments of the present invention will be described below with reference to the attached drawings. Figure 1 is a perspective view showing an example of an inspection apparatus 2 used in the image forming method and correction method according to this embodiment. In Figure 1, some components of the inspection apparatus 2 are represented by functional blocks. Also, the X direction (front-back direction), Y direction (left-right direction), and Z direction (vertical direction) used in the following description are directions perpendicular to each other.

[0019] As shown in Figure 1, the inspection apparatus 2 includes a base 4 that supports various components. A table moving mechanism 6 is positioned on the upper surface of the base 4. The table moving mechanism 6 has, for example, a pair of X-axis guide rails 8 fixed to the upper surface of the base 4 and generally parallel to the X direction. An X-axis moving table 10 is mounted on the X-axis guide rails 8 in such a manner that it can slide along the X direction.

[0020] A nut portion (not shown) that constitutes a ball screw is provided on the underside of the X-axis moving table 10, and a screw shaft 12, which is generally parallel to the X-axis guide rail 8, is connected to this nut portion in a manner that allows it to rotate. A rotational drive source 14, such as a motor, is connected to one end of the screw shaft 12. By rotating the screw shaft 12 with the rotational drive source 14, the X-axis moving table 10 moves along the X-axis guide rail 8, that is, along the X direction.

[0021] A holding table 16 capable of holding a plate-shaped workpiece to be inspected is positioned above the X-axis moving table 10. Figure 2 is a plan view showing an example of a plate-shaped workpiece 11. As shown in Figure 2, the workpiece 11 is a disc-shaped wafer made of a semiconductor material such as silicon. That is, this workpiece 11 has a circular surface 11a and a circular back surface (not shown) opposite to the surface 11a.

[0022] The surface 11a of the workpiece 11 is divided into multiple small regions by a plurality of linear first streets (first division lines) 13a along a first direction D1 that is generally parallel to the surface 11a, and a plurality of linear second streets (second division lines) 13b along a second direction D2 that is generally parallel to the surface 11a and generally perpendicular to the first direction D1. A device 15 such as an integrated circuit (IC) is formed in each small region.

[0023] Furthermore, the workpiece 11 is cut along a plurality of linear first streets 13a and a plurality of linear second streets 13b, dividing it into a plurality of device chips corresponding to each device 15. Therefore, when inspecting the workpiece 11, it is desirable that a support member, such as a dicing tape made of a material such as resin, be attached to the back side of the workpiece 11 so that the entire workpiece 11 can be handled together.

[0024] In this embodiment, a disc-shaped wafer made of a semiconductor material such as silicon is used as the workpiece 11, but the material, shape, structure, size, etc. of the workpiece 11 are not limited to this embodiment. For example, substrates made of other semiconductors, ceramics, resins, metals, etc. can be used as the workpiece 11. Similarly, the type, quantity, shape, structure, size, arrangement, etc. of the devices 15 are not limited to the above embodiment. The workpiece 11 does not need to have devices 15 formed on it.

[0025] Furthermore, the first street 13a and the second street 13b do not necessarily have to be orthogonal as long as they intersect with each other. In other words, the first direction D1 and the second direction D2 do not need to be orthogonal to each other as long as they intersect. Moreover, the workpiece 11 does not necessarily have to be cut on both the first street 13a and the second street 13b as long as it is processed on either the first street 13a or the second street 13b.

[0026] The holding table 16 of the inspection apparatus 2 shown in Figure 1 is connected to a rotational drive source (not shown), such as a motor, and rotates around a rotation axis that is generally parallel to the Z direction. The holding table 16 also moves in the X direction together with the X-axis moving table 10 by the table moving mechanism 6 described above. A portion of the upper surface 16a of the holding table 16 is configured to be porous, for example, and an external suction source (not shown) is connected to this porous portion of the upper surface 16a via a flow path or the like provided inside the holding table 16.

[0027] Therefore, when the workpiece 11 is placed on the upper surface 16a of the holding table 16 and the negative pressure from the suction source is applied to the porous portion of the upper surface 16a, the workpiece 11 is held in place by the suction force associated with this negative pressure. As the suction source, for example, a vacuum pump combining an air supply source and an ejector is used. However, other vacuum pumps such as rotary pumps may also be used as the suction source.

[0028] A camera movement mechanism 18 is positioned next to the table movement mechanism 6 along the Y direction. The camera movement mechanism 18 has, for example, a pair of Y-axis guide rails 20 fixed to the upper surface of the base 4 and generally parallel to the Y direction. A Y-axis moving table 22 is mounted on the Y-axis guide rails 20 in such a manner that it can slide along the Y direction.

[0029] A nut portion (not shown) constituting a ball screw is provided on the underside of the Y-axis moving table 22. A screw shaft 24, which is generally parallel to the Y-axis guide rail 20, is connected to this nut portion in a manner that allows it to rotate. A rotational drive source 26, such as a motor, is connected to one end of the screw shaft 24. By rotating the screw shaft 24 with the rotational drive source 26, the Y-axis moving table 22 moves along the Y-axis guide rail 20, that is, along the Y direction.

[0030] A support structure 28 is provided on the upper surface of the Y-axis moving table 22, with sides positioned approximately parallel to the Z-direction. A pair of Z-axis guide rails 30, also approximately parallel to the Z-direction, are fixed to the sides of this support structure 28. A camera holder 32 is attached to the Z-axis guide rails 30 in such a manner that it can slide along the Z-direction.

[0031] A nut portion (not shown) constituting a ball screw is provided on the support structure 28 side of the camera mounting device 32. A screw shaft (not shown), which is generally parallel to the Z-axis guide rail 30, is connected to this nut portion in a manner that allows it to rotate. A rotational drive source 34, such as a motor, is connected to one end of the screw shaft. By rotating the screw shaft with the rotational drive source 34, the camera mounting device 32 moves along the Z-axis guide rail 30, that is, along the Z direction.

[0032] A camera 36 is fixed to the camera fixture 32 for imaging the workpiece 11 and the like held on the holding table 16. When the Y-axis moving table 22 is moved along the Y direction by the camera moving mechanism 18, the camera 36 also moves along the Y direction. Also, when the camera fixture 32 is moved along the Z direction by the camera moving mechanism 18, the camera 36 also moves along the Z direction.

[0033] This camera 36 includes a one-dimensional image sensor, also known as a line sensor, and a light-gathering lens. It forms a one-dimensional image by collecting light reflected from the workpiece 11 onto the one-dimensional image sensor using the lens. Figure 3 is a schematic diagram showing the arrangement of multiple light-receiving units (pixels) within the one-dimensional image sensor 38 that constitutes the camera 36, ​​and Figure 4 is a schematic plan view showing how an image is formed by the camera 36.

[0034] The one-dimensional image sensor 38 that constitutes the camera 36 is typically a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor that is sensitive to visible light. As shown in Figure 3, this one-dimensional image sensor 38 has a structure in which n light-receiving parts, numbered from 1 to n (where n is a natural number), are arranged linearly. Note that the numbers in Figure 3 are merely assigned to the light-receiving parts for convenience and do not actually number the light-receiving parts.

[0035] Furthermore, as shown in Figure 4, the camera 36 of this embodiment has a linear imaging area 17 realized on the workpiece 11 by the one-dimensional image sensor 38, which is arranged along the X direction. Therefore, when the workpiece 11 held by the holding table 16 is imaged by the camera 36, ​​at a certain timing, a one-dimensional image is obtained in which n pixels are linearly arranged along the X direction.

[0036] Then, as shown in Figure 4, when the workpiece 11 is continuously imaged with the camera 36 while the holding table 16 and the camera 36 are moved relative to each other in a direction perpendicular to the linear imaging area 17, that is, along the Y direction, a band-shaped image is obtained in which multiple one-dimensional images are linked together along the Y direction. There is no theoretical limit to the length of the band-shaped image obtained in this way along the Y direction.

[0037] A control unit 40 is connected to the components of the inspection device 2, such as the table movement mechanism 6, camera movement mechanism 18, and camera 36. The control unit 40 is composed of a computer, for example, a processing unit 42 and a storage device 44, and controls the operation of each of the above-mentioned components so that the workpiece 11 is properly imaged by the camera 36.

[0038] The processing unit 42 is typically a CPU (Central Processing Unit) and performs various processes necessary to control the components described above. The storage device 44 includes, for example, a main memory such as DRAM (Dynamic Random Access Memory) and an auxiliary storage device such as a hard disk drive or flash memory. The functions of this control unit 40 are realized, for example, by the processing unit 42 operating according to software such as programs stored in the storage device 44.

[0039] In the image forming method according to this embodiment, an image of the workpiece 11 is formed by the inspection device 2 described above. At that time, the image of the workpiece 11 is corrected by the correction method according to this embodiment. In other words, the correction method according to this embodiment is a part of the image forming method according to this embodiment.

[0040] In the image forming method according to this embodiment, first, the workpiece 11 to be inspected is held by the holding table 16 of the inspection device 2 (holding step). That is, the workpiece 11 is placed on the upper surface 16a of the holding table 16. In this embodiment, the workpiece 11 is placed on the upper surface 16a of the holding table 16 with its surface 11a facing upwards. Then, the control unit 40 applies negative pressure from a suction source to the porous portion of this upper surface 16a. As a result, the workpiece 11 is held on the holding table 16 by the suction force associated with the negative pressure from the suction source.

[0041] In this embodiment, after the workpiece 11 to be inspected is held on the holding table 16, for example, the control unit 40 rotates the holding table 16 with a rotary drive source so that the first direction D1 is parallel to the X direction and the second direction D2 is parallel to the Y direction. That is, as shown in Figure 4, the first street 13a of the workpiece 11 becomes parallel to the X direction and the second street 13b of the workpiece 11 becomes parallel to the Y direction.

[0042] However, the orientation of the workpiece 11 in the present invention is not limited to this embodiment. For example, the control unit 40 may rotate the holding table 16 with a rotary drive source so that the first direction D1 is parallel to the Y direction and the second direction D2 is parallel to the X direction. Alternatively, the control unit 40 may rotate the holding table 16 with a rotary drive source so that the first direction D1 and the second direction D2 are tilted at any angle with respect to the X direction and the Y direction, respectively.

[0043] Subsequently, a pre-correction image used for inspecting the workpiece 11 is formed (pre-correction image formation step). Specifically, first, the control unit 40 operates the table movement mechanism 6 and the camera movement mechanism 18 to adjust the positional relationship between the holding table 16 and the camera 36 so that the area of ​​the workpiece 11 to be imaged is captured by the camera 36.

[0044] Then, the control unit 40 moves the camera 36 along the Y direction using the camera movement mechanism 18 while continuously imaging the workpiece 11 with the camera 36. That is, the control unit 40 moves the holding table 16 and the camera 36 relative to each other along a direction perpendicular to the linear imaging area 17 of the camera 36, ​​and sequentially images the workpiece 11 with the camera 36. This forms a strip-shaped pre-correction image.

[0045] Figure 5 schematically shows the brightness of each pixel in the formed strip-shaped uncorrected image. In Figure 5, the strip-shaped uncorrected image is schematically shown by concatenating m (where m is a natural number) one-dimensional images (images with n pixels arranged in the X direction) that are repeatedly acquired along the Y direction.

[0046] If the above operation is performed at a different position shifted along the X direction by the amount of the imaging area 17, an additional band-shaped uncorrected image corresponding to this different position is obtained. Then, by concatenating the obtained multiple band-shaped uncorrected images in the X direction, an uncorrected image that captures a wide area along both the X and Y directions is obtained. Figure 6 is an example of an uncorrected image formed by concatenating multiple band-shaped uncorrected images.

[0047] As shown in Figures 5 and 6, at both ends of the strip-shaped uncorrected image along the X direction, the ambient light is insufficient, making it easy for pixel brightness to decrease. Figure 5 shows that the brightness of the first, second, (n-1)th, and nth pixels located at both ends along the X direction has decreased, resulting in patches of light and dark. Note that in Figure 5, the decrease in brightness of the first and nth pixels in the X direction is greater than the decrease in brightness of the second and (n-1)th pixels in the X direction.

[0048] Therefore, if a correction can be performed to cancel out the decrease in brightness of each pixel, it is thought that the unevenness of light and dark can be removed from the image and the accuracy of the inspection can be maintained at a high level. However, as described above, the workpiece 11 is processed in the first street 13a and the second street 13b, and processing marks are formed in the first street 13a and the second street 13b.

[0049] Therefore, when an image is acquired using the general imaging procedure described above, the first street 13a and the second street 13b may only be visible in a narrow area of ​​the image. For example, in the procedure described above, the holding table 16 and the camera 36 are moved relative to each other along the second street 13b, so the second street 13b may only be visible in a limited area in the X direction. In Figure 5, the second street 13b is visible only in the third and fourth pixels in the X direction.

[0050] Furthermore, in such images, the light and dark areas resulting from processing marks on the first street 13a and the second street 13b cannot be clearly distinguished from the light and dark patches that should be removed. In other words, it was extremely difficult to extract information such as brightness correction values ​​suitable for correction from images acquired using the general imaging procedure described above.

[0051] It is also conceivable to acquire a correction image by moving the holding table 16 and the camera 36 relatively along the X direction. However, for example, if the holding table 16 and the camera 36 are moved relatively along the first street 13a, the first street 13a and the imaging area 17 may coincidentally overlap, resulting in only the first street 13a being captured in the entire image.

[0052] Furthermore, in such images, the light and dark areas resulting from processing marks on the first street 13a and the second street 13b cannot be clearly distinguished from the light and dark patches that should be removed. Thus, the method of acquiring images by relatively moving the holding table 16 and the camera 36 along the first street 13a and the second street 13b did not always yield images suitable for extracting correction information.

[0053] Therefore, in the image forming method according to this embodiment, after the workpiece 11 to be inspected is held on the holding table 16, a correction image used for correcting the image is formed (correction image forming step). Specifically, the control unit 40 continuously captures images of the workpiece 11 with the camera 36, ​​moves the holding table 16 along the X direction with the table moving mechanism 6, and moves the camera 36 along the Y direction with the camera moving mechanism 18.

[0054] In other words, the control unit 40 moves the holding table 16 and the camera 36 relative to each other so as not to follow the first direction D1 and the second direction D2, and sequentially images the workpiece 11 with the camera 36. This forms a band-shaped correction image. Figure 7 is a schematic plan view showing how the correction image is formed.

[0055] As shown in Figure 7, in this embodiment, the control unit 40 moves the holding table 16 and the camera 36 relative to each other so that the imaging area 17 of the camera 36 moves along the surface 11a of the workpiece 11 in a linear trajectory that intersects the first direction D1 and the second direction D2. Figure 8 is a schematic diagram showing the brightness of each pixel in the formed correction image. In the correction image acquired in this procedure, the first street 13a and the second street 13b are tilted with respect to the contour of the image. For example, the image in Figure 8 shows the tilted second street 13b.

[0056] Thus, in this embodiment, the holding table 16 and the camera 36 are moved relative to each other so as not to follow the first direction D1 and the second direction D2. Therefore, the first street 13a and the second street 13b are not captured by only a part of the imaging area 17, nor are only the first street 13a and the second street 13b captured by the entire camera 36.

[0057] In other words, the first street 13a and the second street 13b do not appear in only a narrow area of ​​the correction image, nor do only the first street 13a and the second street 13b appear in the correction image. Therefore, correction information suitable for image correction can be extracted from this correction image.

[0058] Furthermore, this correction image may be formed before the pre-correction image is formed, or it may be formed after the pre-correction image is formed. Also, in this embodiment, the holding table 16 and the camera 36 move relative to each other so that the imaging area 17 traces a linear trajectory that intersects the first direction D1 and the second direction D2, but the manner in which the relative movement of the holding table 16 and the camera 36 occurs is not limited to this.

[0059] The holding table 16 and the camera 36 only need to move relative to each other so as not to follow the first direction D1 and the second direction D2. For example, the control unit 40 may move the holding table 16 and the camera 36 relative to each other so that the imaging area 17 of the camera 36 moves along the surface 11a of the workpiece 11 in a spiral trajectory.

[0060] In this modified example, the relative direction of movement between the holding table 16 and the camera 36 momentarily becomes parallel to the first direction D1 and the second direction D2. However, this mode of relative movement is not "relative movement along the first direction D1 and the second direction D2." In other words, the mode of relative movement described in the above modified example is included in the "relative movement that does not follow the first direction D1 and the second direction D2" of the present invention.

[0061] After the correction image is formed, information on multiple correction values ​​used to correct the brightness of multiple pixels that make up the image formed by the camera 36 is obtained based on the brightness of multiple pixels that make up the correction image (correction value acquisition step). Specifically, the control unit 40 calculates the average brightness of all pixels in the same column, compares this with a reference value, and obtains the brightness correction value for the corresponding column.

[0062] For example, when a correction image of m rows and n columns, as shown in Figure 8, is formed, the control unit 40 calculates the average value of the luminance of m pixels (where k is a natural number less than or equal to n) that make up the k column. That is, the control unit 40 calculates the sum of the luminances of the k pixel present in each row from the 1st row to the mth row, and divides this by m. The average value is calculated for all columns. This gives n average values ​​corresponding to each column.

[0063] Furthermore, the control unit 40 obtains a reference value by any method. For example, the control unit 40 calculates the average brightness of all pixels constituting the correction image as the reference value. In other words, the control unit 40 calculates the sum of the brightness of all pixels constituting the m x n correction image and divides this by m x n. This gives the reference value.

[0064] The control unit 40 then calculates a correction value for the brightness of each column from the average brightness of each column and a reference value. For example, the control unit 40 calculates the correction value for the brightness of each column by subtracting the reference value from the average brightness of each column. This gives n correction values ​​corresponding to each column. Figure 9 schematically shows the patches of light and dark that are removed by applying the correction values ​​obtained in this way. In other words, Figure 9 schematically represents the correction value for the brightness of each pixel.

[0065] In this embodiment, since the first street 13a and the second street 13b do not appear in only a narrow area of ​​the correction image, the average value of the m pixels constituting the k column is not affected by the first street 13a and the second street 13b, and an abnormal correction value is not calculated.

[0066] After multiple correction values ​​have been acquired and the pre-correction image has been formed, the brightness of multiple pixels in the pre-correction image is corrected using the acquired multiple correction values ​​to form a corrected image (post-correction image formation step). For example, the control unit 40 forms a corrected image by increasing or decreasing the brightness of multiple pixels in the pre-correction image according to the correction values. Figure 10 is a schematic diagram showing the brightness of each pixel in the formed corrected image, and Figure 11 is an example of the formed corrected image.

[0067] As described above, in the image forming method according to this embodiment, a correction image is formed by sequentially imaging the workpiece 11 with the camera 36 while relatively moving the holding table 16 and the camera 36 so as not to follow the first direction D1 and the second direction D2. Therefore, when forming this correction image, the first street (first division planned line) 13a and the second street (second division planned line) 13b are not imaged using only a part of the linear imaging area 17 of the camera 36, ​​nor are only the first street 13a and the second street 13b imaged using the entire camera 36.

[0068] In other words, the first street 13a and the second street 13b are not captured in only a narrow area of ​​the correction image, nor are only the first street 13a and the second street 13b captured in the correction image. Therefore, correction values ​​suitable for correcting the brightness of multiple pixels constituting the image can be obtained from this correction image. Thus, according to this embodiment, an image forming method is provided that can form a more appropriately corrected image.

[0069] As described above, the correction method according to this embodiment corresponds to a part of the image forming method according to this embodiment. That is, in the correction method according to this embodiment, the workpiece 11 to be inspected is held by the holding table 16 of the inspection device 2 (holding step). After the workpiece 11 to be inspected is held by the holding table 16, a correction image used for image correction is formed (correction image forming step).

[0070] Furthermore, after the correction image is formed, information on multiple correction values ​​used to correct the brightness of multiple pixels constituting the image formed by the camera 36 is acquired based on the brightness of multiple pixels constituting the correction image (correction value acquisition step). Then, after the multiple correction values ​​have been acquired, the brightness of multiple pixels constituting an arbitrary image formed by the camera 36 is corrected using the acquired multiple correction values ​​(correction step: corresponding to the post-correction image formation step of the image formation method).

[0071] It should be noted that the present invention is not limited to the embodiments described above and can be implemented with various modifications. For example, in the image forming method of the embodiments described above, both a pre-correction image and a correction image are formed by imaging the same workpiece 11. However, for example, when inspecting the same type of workpiece 11 consecutively, the pre-correction image formed by imaging another workpiece 11 may be corrected using the correction image formed by imaging one workpiece 11.

[0072] Furthermore, the structures, methods, etc., of the embodiments and modified versions described above can be modified as appropriate without departing from the scope of the present invention. [Explanation of symbols]

[0073] 11: Workpiece 11a: Surface 13a: 1st Street (1st planned division line) 13b: Second Street (Planned Second Division Line) 15: Device 17: Imaging area 2: Inspection equipment 4: Base 6: Table movement mechanism 8: X-axis guide rail 10: X-axis movement table 12: Screw shaft 14: Rotary drive source 16: Holding Table 16a:Top surface 18: Camera movement mechanism 20: Y-axis guide rail 22: Y-axis moving table 24: Screw shaft 26: Rotary drive source 28 :Support structure 30: Z-axis guide rail 32: Camera mounting device 34: Rotary drive source 36: Camera 38: 1D image sensor 40: Control Unit 42: Processing unit 44:Storage device

Claims

1. An image forming method for forming an image of a plate-shaped workpiece by capturing the surface side of the workpiece, which has been processed along one or both of a linear first division line along a first direction and a linear second division line along a second direction intersecting the first direction, using a camera including a one-dimensional image sensor. A holding step of holding the workpiece on a holding table, A correction image forming step is performed after the holding step, in which the holding table and the camera are moved relative to each other so as not to be aligned with the first direction and the second direction, and the workpiece is sequentially photographed with the camera to form a correction image, A correction value acquisition step, which acquires a plurality of correction values ​​used when correcting the brightness of a plurality of pixels constituting an image formed by the camera, based on the brightness of a plurality of pixels constituting the correction image, A pre-correction image forming step is performed by sequentially imaging the workpiece with the camera while moving the holding table and the camera relative to each other along a direction perpendicular to the linear imaging area of ​​the camera, thereby forming a pre-correction image. An image forming method comprising: a post-correction image forming step, after the correction value acquisition step and the pre-correction image forming step, a post-correction image forming step, in which a post-correction image is formed by correcting the brightness of a plurality of pixels constituting the pre-correction image using the plurality of correction values.

2. The image forming method according to claim 1, wherein in the corrective image forming step, the holding table and the camera are moved relative to each other so that the imaging area of ​​the camera moves across the surface of the workpiece in a linear trajectory that intersects the first direction and the second direction.

3. The image forming method according to claim 1, wherein in the corrective image forming step, the holding table and the camera are moved relative to each other so that the imaging area of ​​the camera moves across the surface of the workpiece in a spiral trajectory.

4. A correction method used when forming an image of a plate-shaped workpiece by capturing the surface side of the workpiece, which has been processed along one or both of a linear first division line along a first direction and a linear second division line along a second direction intersecting the first direction, using a camera including a one-dimensional image sensor, A holding step of holding the workpiece on a holding table, A correction image forming step is performed after the holding step, in which the holding table and the camera are moved relative to each other so as not to be aligned with the first direction and the second direction, and the workpiece is sequentially photographed with the camera to form a correction image, A correction value acquisition step, which acquires a plurality of correction values ​​used when correcting the brightness of a plurality of pixels constituting an image formed by the camera, based on the brightness of a plurality of pixels constituting the correction image, A correction method comprising, after the correction value acquisition step, a correction step of correcting the brightness of a plurality of pixels constituting an image formed by the camera using the plurality of correction values.

5. The correction method according to claim 4, wherein in the correction image forming step, the holding table and the camera are moved relative to each other so that the imaging area of ​​the camera moves across the surface of the workpiece in a linear trajectory that intersects the first direction and the second direction.

6. The correction method according to claim 4, wherein in the correction image forming step, the holding table and the camera are moved relative to each other so that the imaging area of ​​the camera moves across the surface of the workpiece in a spiral trajectory.

Images