JP6987097B2 - Control device and control method - Google Patents

Description

The present invention relates to a dicing apparatus and a dicing method, and more particularly to a dicing apparatus and a dicing method for photographing a processing target using an autofocus camera.

In the dicing apparatus, when detecting the surface height of the work and positioning the processed portion, autofocus or pattern matching using an image of a microscope camera is performed.

For example, in Patent Document 1, the distance c from the upper surface of the silicon wafer is measured by a camera, the distance d between the lower end of the blade and the upper surface of the silicon wafer is calculated from this distance c, and a predetermined amount of cut is set in the distance d. A dicing method for lowering the blade by the added distance is described.

Further, in Patent Document 2, the focus data for focusing the visible light generated by the observation light source on the surface is calculated based on the image pickup data output from the image pickup element, and the focus data is used as the basis. Described is a laser processing apparatus having an autofocus unit that adjusts the focus of visible light to a surface by controlling the movement of a Z-axis stage.

Japanese Unexamined Patent Publication No. 7-40335 Japanese Unexamined Patent Publication No. 2012-192459

On-axis / magnification chromatic aberration exists in the optical system of the microscope camera, and chromatic aberration is reduced by optimally designing an optical system such as a lens. However, if the aberration cannot be reduced sufficiently due to the limited degree of freedom in design, or if the aberration cannot be corrected on the objective side due to restrictions on applications other than the observation optical system, high-precision requirements are met. , There is a problem that the specifications cannot be satisfied.

Further, although image processing using a grayscale image is generally performed, a monochrome sensor that captures a grayscale image generally has sensitivity characteristics over the entire visible light range (380 nm to 780 nm), and is a monochrome image. The processing is clearly affected by chromatic aberration. Therefore, there is a problem that the focus position shifts in the autofocus and the pattern contour becomes unclear in the alignment, which leads to a decrease in the detection ability or a decrease in the accuracy.

In addition, monochromatic light illumination is often used to remove chromatic aberration, but when monochromatic light illumination is used when various surface reflectances are present, filters such as band pass filters are physically used. It was necessary to replace the optical axis, and it was necessary to adjust the optical axis of the observation optical system.

The present invention has been made in view of such circumstances, and an object of the present invention is to provide a dicing apparatus and a dicing method for performing high-precision machining by performing autofocus and positioning by eliminating the influence of chromatic aberration. ..

In order to achieve the above object, one aspect of the dying device is from a holding portion for holding the work, an objective lens, a first light source that irradiates the work through the objective lens, and the work through the objective lens. An observation optical unit having an image pickup element that receives reflected light, and acquiring an image signal of a plurality of colors including an image signal of a reference color and an image signal of a color other than the reference color from the image pickup element. , The selection unit that selects at least one of a plurality of colors according to the color of the surface of the work, and the relative position of the objective lens with respect to the work, the focusing position that maximizes the contrast of the image signal of the selected color. The auto-focus unit is set by the auto-focus unit, the storage unit that stores the difference between the in-focus position of the reference color in the observation optical unit and the in-focus position of colors other than the reference color for each color. A correction unit that corrects the corrected focusing position based on the difference of the selected color, which is a difference stored in the storage unit, and a processing unit that processes the work based on the corrected focusing position information. And equipped with.

According to this aspect, at least one of a plurality of colors is selected according to the color of the surface of the work, and the relative position of the objective lens with respect to the work is focused so that the contrast of the image signal of the selected color is maximized. Set to a position and correct the set in-focus position based on the difference between the in-focus position of the reference color and the in-focus position of colors other than the reference color in the observation optical unit stored in the storage unit. Therefore, it is possible to eliminate the influence of chromatic aberration and perform autofocus and positioning, and it is possible to perform high-precision processing.

It is preferable to include an input unit in which the user inputs the color of the surface of the work. As a result, the color of the surface of the work can be appropriately obtained.

Further, an identification unit that identifies the color of the surface of the work based on the image signals of a plurality of colors acquired by the observation optical unit may be provided. Further, the identification unit may identify the hue angle on the surface of the work. As a result, the color of the surface of the work can be appropriately obtained.

The plurality of colors are three colors of red, green, and blue, and the reference color is preferably green. This makes it possible to use an image pickup device in which a primary color filter is arranged as an observation optical unit.

The processing unit is provided integrally with the observation optical unit, and is provided with a second light source for processing the work by irradiating the work with laser light via an objective lens. When the work is irradiated with laser light at the in-focus position, the laser light is emitted. Is focused inside the work. This aspect is suitable for a laser dicing device that processes a work by condensing laser light inside the work.

One aspect of the dying method for achieving the above object is a holding step of holding the work in the holding portion, irradiating the work through the objective lens, and receiving the reflected light from the work through the objective lens. An observation step of acquiring an image signal of a plurality of colors including an image signal of a reference color and an image signal of a color other than the reference color by an observation optical unit, and a plurality of colors according to the color of the surface of the work. A selection step of selecting at least one color, an autofocus step of setting the relative position of the objective lens with respect to the work to the in-focus position where the contrast of the image signal of the selected color is maximized, and a reference in the observation optics section. The storage process of storing the difference between the in-focus position of the color to be the focus and the in-focus position of the color other than the reference color in the storage unit for each color and the in-focus position set by the autofocus process are stored. It is provided with a correction step of correcting the difference based on the difference of the selected color, and a processing step of processing the work based on the corrected focus position information.

According to this aspect, at least one of a plurality of colors is selected according to the color of the surface of the work, and the relative position of the objective lens with respect to the work is focused so that the contrast of the image signal of the selected color is maximized. Set to a position and correct the set in-focus position based on the difference between the in-focus position of the reference color and the in-focus position of colors other than the reference color in the observation optical unit stored in the storage unit. Therefore, it is possible to eliminate the influence of chromatic aberration and perform autofocus and positioning, and it is possible to perform high-precision processing.

The dicing method is configured as a program for realizing each of the above steps in a computer, and a non-temporary recording medium such as a CD-ROM (Compact Disk-Read Only Memory) in which the program is stored can also be configured. Is.

According to the present invention, high-precision machining can be performed by performing autofocus and positioning by eliminating the influence of chromatic aberration.

Configuration diagram showing the outline of the laser dicing device Block diagram showing the configuration of the control unit Flow chart showing an example of the procedure of the dicing method Flow chart showing an example of the procedure of the dicing method Flow chart showing an example of the procedure of the dicing method The figure which showed an example of the focus intensity of each color image at the height in each Z direction. The figure which shows the color image acquired at the height in each Z direction, and an example of each color image. The figure which showed an example of the focus intensity of each color image at the height in each Z direction.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[Construction of laser dicing device]
FIG. 1 is a configuration diagram showing an outline of a laser dicing apparatus 10 according to an embodiment of the present invention. As shown in the figure, the laser dicing device 10 includes a stage 12 for moving the work W, an optical unit 40 in which the processing optical unit 20 and the observation optical unit 30 are integrally provided, and between the optical unit and the work W. A drive unit 50 for changing the distance and a control unit 60 are provided.

The stage 12 (an example of the holding portion) is configured to be movable in the XYZθ direction, and sucks and holds the work W placed on the stage 12.

The optical unit 40 is arranged at a position facing the work W, and irradiates the work W with a processing laser beam L1 for forming a modified region by multiphoton absorption inside the work W.

The control unit 60 controls the operation of each unit of the laser dicing device 10 and stores data necessary for processing the work W.

In addition to this, the laser dicing device 10 includes a work transfer means (not shown).

Next, the detailed configuration of the optical unit 40 will be described.

As shown in FIG. 1, the processing optical unit 20 of the optical unit 40 includes a processing laser light source 22, a collimating lens 24, a mirror 26, and a lens 28.

The processing laser light source 22 (an example of the second light source) emits the processing laser light L1 for forming a modified region inside the work W. The processing laser beam L1 has, for example, a wavelength of 1.
.. 1 [mu] m, the pulse width is not more 1μs or less, the peak power density at the converging point is ^{1 × 10 8 (W / cm} 2) or more laser beams.

A collimating lens 24, a mirror 26, and a lens 28 are arranged in this order from the processing laser light source 22 side on the optical path of the processing laser light L1.

The collimating lens 24 adjusts the incident laser light to a parallel state. The mirror 26 is arranged so as to change the direction of the optical axis of the laser beam by 90 degrees. The lens 28 emits the incident laser light from the processing optical unit 20.

Further, the observation optical unit 30 of the optical unit 40 includes an observation laser light source 32, a collimating lens 34, a half mirror 36, and an SD (Standard Definition) color camera 38.

The observation laser light source 32 (an example of the first light source) is composed of, for example, an LD (Laser Diode) light source, an SLD (Super Luminescent Diode) light source, or the like, and has a wavelength different from that of the processing laser light L1 and is of the work W. An observation laser beam L2 having a wavelength that can be reflected on the surface is emitted.

The collimating lens 34 adjusts the incident laser light to a parallel state. The half mirror 36 is arranged at a position where the optical axis of the processing laser light L1 and the optical axis of the observation laser light L2 intersect, and the processing laser light L1 incident from the processing optical unit 20 is transmitted and incident. The observation laser beam L2 is reflected to change the direction of the optical axis by 90 degrees.

The SD color camera 38 includes CMOS (Complementary Metal Oxide Semiconductor) or C.
This is a digital camera that captures a subject (work W in this case) using an image pickup device such as a CD (Charge-Coupled Device). The image pickup device includes a large number of photoelectric conversion elements (photodiodes) that generate electric charges according to the amount of light received. Each photodiode is evenly arranged in the row direction and the column direction, and is separated for each pixel by a pixel separation region provided around the photodiode. The electric charge generated by the photodiode is stored in a capacitor connected to each photodiode, and the electric charge stored in the capacitor is read out as an image signal.

A color filter is formed on the front surface of the photodiode that constitutes the pixel. The color filter is composed of red (R), green (G), and blue (B) primary color filters arranged in a predetermined arrangement structure such as a Bayer arrangement, and any one of them is used for each photodiode. Color filters for colors are arranged to correspond.

The SD color camera 38 configured in this way can select only the signals from the respective pixels of R, G, and B even if the illumination light having a wide wavelength is used, so that the SD color camera 38 can be of R, G, and B, respectively. It is possible to acquire three images, an R channel image, a G channel image, and a B channel image, which are equivalent to the images acquired by using the illumination light having a wavelength in a narrow band. These images have chromatic aberration characteristics close to those of R, G, and B monochromatic light, respectively.

In the optical unit 40 configured as described above, the processing laser light L1 emitted from the processing laser light source 22 is parallelized by the collimating lens 24, the optical axis is changed by 90 degrees by the mirror 26, and then the lens is formed. It is output from the processing optical unit 20 via 28.

Further, the observation laser light L2 emitted from the observation laser light source 32 is parallelized by the collimating lens 34, the optical axis is changed by 90 degrees by the half mirror 36, and is output from the observation optical unit 30. Here, the processing laser light L1 output from the processing optical unit 20 is incident on the observation optical unit 30, passes through the half mirror 36, and is output from the observation optical unit 30 in the same optical path as the observation laser light L2. ..

Further, the optical unit 40 includes a lens 42 and an objective lens 44. The lens 42 transmits the incident laser light and causes it to enter the objective lens 44. The objective lens 44 is a condensation lens (condensing lens) and condenses the incident laser light.

The processing laser light L1 and the observation laser light L2 output from the observation optical unit 30 are incident on the objective lens 44 via the lens 42 arranged on the optical path. The processing laser light L1 and the observation laser light L2 incident on the objective lens 44 are focused at different distances in the Z direction (work thickness direction). When the height of the optical unit 40 in the Z direction (the distance between the objective lens 44 and the surface of the work W, an example of the relative position of the objective lens with respect to the work) is appropriately adjusted, the processing laser beam incident on the objective lens 44 L1 is focused by the objective lens 44 to a desired depth inside the work W, and the observation laser beam L2 incident on the objective lens 44 is focused on the surface of the work W by the objective lens 44.

The reflected light of the observation laser beam L2 reflected on the surface of the work W returns to the objective lens 44, passes through the objective lens 44 and the lens 42, and is incident on the half mirror 36. The reflected light incident on the half mirror 36 is changed in the direction of the optical axis by 90 degrees by the half mirror 36, passes through the collimating lens 34, and is incident on the SD color camera 38. That is, the SD color camera 38 receives the reflected light via the objective lens 44.

The SD color camera 38 acquires an image of the surface of the work W from the reflected light of the incident observation laser beam L2.

FIG. 2 is a block diagram showing the configuration of the control unit 60. As shown in the figure, the control unit 60 includes a CPU (Central Processing Unit) 62, a memory 64, an input / output circuit 66, a user interface 68, an image processing unit 70, an autofocus control unit 72, a processing control unit 74, and a bus. It has 76.

The CPU 62 controls each unit of the control unit 60 in an integrated manner. The memory 64 is composed of a ROM (Read Only Memory) in which an operation program is stored, a RAM (Random Access Memory) which is a work area when executing a program, and a frame memory in which image data is temporarily stored. ing. The input / output circuit 66 is an interface for inputting / outputting a control signal to / from the stage 12, the processing optical unit 20, the observation optical unit 30, and the drive unit 50.

The user interface 68 includes display devices such as switches, a television monitor, and an indicator lamp for which the user operates the operation of each part. The TV monitor displays an image of the work W captured by the SD color camera 38, or displays program contents, various messages, and the like. The indicator lamp displays the operating status of the laser dicing apparatus 10 during processing, processing completion, emergency stop, and the like.

The image processing unit 70 performs various image processing on the image captured by the SD color camera 38. The autofocus control unit 72 (an example of the autofocus unit) captures images in the SD color camera 38 by changing the height of the optical unit 40 in the Z direction by the drive unit 50 based on the image captured by the SD color camera 38. The in-focus position where the contrast of the image is maximized, that is, the condensing point of the observation laser beam L2 is moved to a position corresponding to the surface of the work W (autofocus operation).

The machining control unit 74 positions the optical unit 40 and the work W in the XYZθ direction, feeds the stage 12 in the X direction while irradiating the work W with the laser beam L1 for machining, and feeds the index in the Y direction. Let me. The bus 76 connects each unit of the control unit 60 to each other to transmit data.

The control unit 60 configured in this way causes the optical unit 40 to perform an autofocus operation by the autofocus control unit 72 based on the image of the surface of the work W acquired by the SD color camera 38. As described above, the condensing point of the processing laser light L1 and the condensing point of the observation laser light L2 are at different positions in the Z direction, and the condensing point of the observation laser light L2 is one on the surface of the work W. At this time, the condensing point of the processing laser beam L1 is set to a desired distance in the thickness direction of the work W.

The optical unit 40 (an example of the processing unit) moves the stage 12 in the XY direction in a state where the condensing point of the observation laser beam L2 is aligned with the surface of the work W, so that the processing laser beam L1 causes the work W. A modified region along the scribing line can be formed at the desired depth.

[Method of using image information: first embodiment]
As described above, the processing laser light L1 and the observation laser light L2 are incident on the work W via the objective lens 44. Here, since the objective lens 44 is optimized for the processing laser light L1, chromatic aberration occurs at the wavelength of the observation laser light L2. If the autofocus operation is performed with chromatic aberration, the measurement accuracy will deteriorate.

Therefore, when processing the work W, it is necessary to perform autofocus that reduces the influence of chromatic aberration. The laser dicing apparatus 10 according to the first embodiment is appropriate because the user specifies the R / G / B color channel of the image used for autofocus and positioning according to the color of the surface of the work W. Use images of different colors for image processing.

FIG. 3 is a flowchart showing an example of the procedure of the dicing method according to the first embodiment.

In step S1, the user inputs the surface color of the work W using the user interface 68 (an example of the input unit, see FIG. 2).

Next, in step S2, it is determined whether or not the color of the surface of the input work W is red. If it is red, the process proceeds to step S3, and the R channel image is used for image processing. That is, the image processing unit 70 creates an R channel image from the image captured by the SD color camera 38, and the autofocus control unit 72 autofocuses the optical unit 40 based on the created R channel image for processing. The control unit 74 controls the stage 12 based on the created R channel image to position the work W, feed the work W in the XY direction, and form a modified region by the processing laser beam L1. After that, the process related to this flowchart is terminated.

If it is not red, the process proceeds to step S4, and it is determined whether or not the input work surface color is green. If it is green, the process proceeds to step S5, the G channel image is used for image processing, and the processing related to this flowchart is completed.

If it is not green, the process proceeds to step S6, and it is determined whether or not the input work W surface color is blue. If it is blue, the process proceeds to step S7, the B channel image is used for image processing, and the processing related to this flowchart is completed.

If it is not blue, the process proceeds to step S8, and it is determined whether or not the input work W surface color is purple. If it is purple, the process proceeds to step S9, the R channel image and the B channel image are combined and used for image processing, and the processing related to this flowchart is completed.

If it is not purple, the process proceeds to step S10, and it is determined whether or not the input work W surface color is yellow. If it is yellow, the process proceeds to step S11, the R channel image and the G channel image are combined and used for image processing, and the processing related to this flowchart is completed.

If it is not yellow, the process proceeds to step S12, the G-channel image and the B-channel image are combined and used for image processing, and the processing according to this flowchart is completed.

As described above, an image having a color corresponding to the color of the surface of the work W can be used for image processing. As a result, the influence of chromatic aberration of the optical system to be used can be reduced, the accuracy of autofocus can be improved, and the detection ability of positioning can be improved.

[Second Embodiment]
The laser dicing apparatus 10 according to the second embodiment automatically recognizes the color of the surface of the work W, automatically calculates the image of the optimum color channel, and uses it for image processing.

FIG. 4 is a flowchart showing an example of the procedure of the dicing method according to the second embodiment.

In step S21, the work W is imaged by the SD color camera 38 and a color image is acquired.

Next, in step S22, the acquired color image is converted into an L * a * b color system, and hue angle data for each pixel is collected. The L * a * b color system is a color system using a uniform color space composed of a lightness L * and chromatic exponents a * and b *. In addition, L * u * v * color system and H
Hue angle data may be collected by converting to an SI (Hue, Saturation, Intensity) color system.

Subsequently, in step S23, the image processing unit 70 (an example of the selection unit and an example of the identification unit) identifies the color of the surface of the work W based on the hue angle data for each collected pixel.

Specifically, the hue angle of each pixel is voted in the hue angle voting space by Hough transform, and the hue angle with the largest number of votes is recognized as the surface color of the work W. In the case of achromatic color, the hue angle is also indefinite because the saturation information cannot be obtained. Therefore, this process is performed on pixels that exceed a certain saturation threshold.

Next, in step S23, the coefficient _A R according to the hue angle of the color of the surface of the workpiece W, to calculate the _{A G,} and _{A B.} Coefficient, for example if the hue angle of the center of R and _{G, A R = 0.5, A} G = 0.5, which is _A B = 0. Moreover, R: G = 2: In the case of the first _angle, which is _{A R = 2/3, A} G = 1/3, A B = 0.

Finally, in step S24, the calculated coefficients A _R, A _G, and synthesizes the image data corresponding to A _B, utilizing the synthesized image data to the image processing. Image data, for example, synthesized as image data = R channel image × _A R + G channel image × _A G + B channel image × _{A B.}

Based on this image data, the autofocus control unit 72 controls the drive unit 50 to operate the optical unit 40 in the autofocus operation, and the machining control unit 74 controls the stage 12 to position the work W and perform the XY direction. Send. After that, the process related to this flowchart is terminated.

As described above, the color of the surface of the work W can be recognized from the acquired image, and the image of the color corresponding to the recognized color can be used for image processing. As a result, the influence of chromatic aberration of the optical system to be used can be reduced, the accuracy of autofocus can be improved, and the detection ability of positioning can be improved.

[Third Embodiment]
The laser dicing apparatus 10 according to the third embodiment acquires in advance the offset due to the chromatic aberration of red and blue (an example of a color other than the reference color) when green (an example of a reference color) is used as a reference. Then, the in-focus position is corrected using this offset data.

FIG. 5 is a flowchart showing an example of the procedure of the dicing method according to the third embodiment.

First, in step S31 (an example of a storage process), chromatic aberration data of the optical unit 40 (observation optical unit 30, lens 42, objective lens 44) of the laser dicing device 10 is acquired and stored. This process is performed in advance at the time of adjustment such as at the time of shipment of the apparatus.

Obtaining chromatic aberration data, first places the workpiece W _S as a reference to the stage 12, and the Z-direction height of the workpiece W _S optical unit 40 by the driving unit 50 (the objective lens 44 and the workpiece W _S of the surface while changing the distance) imaging the workpiece W _S by SD color camera 38, acquires the R channel image of the workpiece W _S, G channel image, and the B-channel image at each Z-direction height. Then, the focus intensity is extracted from the contrast of these images, and the height in the Z direction of the optical unit 40 at which the focus intensity of the R channel image, the G channel image, and the B channel image peaks is acquired, respectively.

FIG. 6 is a diagram showing an example of the focus intensity of the R channel image, the G channel image, and the B channel image at each height of the optical unit 40 in the Z direction, and the horizontal axis is the height of the optical unit 40 in the Z direction. The vertical axis represents the focus intensity. In the example shown in the figure, the height of the optical unit 40 at which the focus intensity of the R channel image peaks is 1 μm lower than the height of the optical unit 40 at which the focus intensity of the G channel image peaks. , The height in the Z direction of the optical unit 40 at which the focus intensity of the B channel image peaks is +1.2 μm higher. Therefore, the offset due to red chromatic aberration when green is used as a reference (an example of difference) is +1 μm, and the offset due to blue chromatic aberration when based on green is -1.2 μm, and the memory 64 (an example of a storage unit) is set for each color. ).

After storing the chromatic aberration data, the work W of the product can be processed. To process the work W of the product, first, in step S32, the work W of the product is placed on the stage 12 (an example of the holding process), and the work W is changed by the SD color camera 38 while changing the height of the optical unit 40 in the Z direction. Is imaged, and a color image at each height in the Z direction is acquired (an example of the observation process).

Subsequently, in step S33, from the color image captured at each Z-direction height acquired in step S32, an R-channel image, a G-channel image, and a B-channel image at each Z-direction height (an image signal of a reference color). An example of an image signal of a plurality of colors including an image signal of a color other than the reference color and an image signal of a color other than the reference color) is created.

_{FIG. 7 shows the color images 100, 102 and 104 acquired at different heights h 1} , h ₂ and h ₃ in the Z direction of the optical unit 40, and the R channel image 100R and G channel obtained by decomposing the color image 100 into each RGB color. The image 100G, the B channel image 100B, the R channel image 102R obtained by decomposing the color image 102 into each RGB color, the G channel image 102G, and the B channel image 102B, and the R channel image 104R obtained by decomposing the color image 104 into each RGB color. It is a figure which shows an example of G channel image 104G, and B channel image 104B. Here, the alignment mark on the surface of the work W is imaged.

Next, in step S34, the image processing unit 70 (an example of the selection unit) calculates the focus intensity from the R channel image, the G channel image, and the B channel image, respectively, and determines the color to be used for the image processing. The in-focus position is measured using a color image.

Here, the focus intensities of the R channel images 100R, 102R, and 104R, the G channel images 100G, 102G, and 104G, the B channel images 100B, 102B, and 104B are compared, and the image having the highest contrast is extracted and the image thereof. Let the color of be used for image processing. As shown in FIG. 7, since the R channel images 100R, 102R, and 104R have the highest contrast of the alignment mark, the color used for image processing is determined to be red (an example of the selection process). ). That is, here, the color of the surface of the work W is the closest to red among red, green, and blue.

Further, among the R channel images 100R, 102R, and 104R, the R channel image 102R is captured with the highest contrast of the alignment mark, so that the Z direction of the optical unit 40 when the color image 102 is captured is taken. The height h ₂ is set to the in-focus position (an example of the autofocus process).

Finally, in step S35 (an example of the correction step), the autofocus control unit 72 (an example of the correction unit) adds an offset amount from the focusing position, and the height of the optical unit 40 in the Z direction is focused on the green reference. Correct to the position. In this embodiment, the color used for the autofocus operation is red (an example of the selected color). Then, the offset "+1 μm" due to the chromatic aberration of red when green is used as a reference is stored in the memory 64. Therefore, the offset due to red chromatic aberration is added to the in-focus position to set the height of the optical unit 40 in the Z direction to (h ₂ + 1 μm).

FIG. 8 is a diagram showing an example of the focus intensity of the R channel image, the G channel image, and the B channel image at each height of the optical unit 40 when the work W is imaged, and the horizontal axis is the optical unit. The height of 40 in the Z direction and the vertical axis represent the focus intensity. As shown in the figure, R-channel image in the Z-direction height h ₂ of the optical unit 40 is most focused high intensity. Therefore, the optical unit 40 is set as the in-focus position at the position of the _{height h 2 in the Z direction.}

Here, as shown in FIG. 6, in the present embodiment, the focus intensity of the R channel image peaks with respect to the height in the Z direction of the optical unit 40 at which the focus intensity of the reference G channel image peaks. The height of the optical unit 40 in the Z direction is 1 μm lower.

Therefore, as shown in FIG. 8, the height of the optical unit 40 at the in-focus position of the green reference _{needs to be corrected by +1 μm with respect to the set in-focus position to be (h 2} + 1 μm). There is. Therefore, the autofocus control unit 72 corrects the height of the optical unit 40 in the Z direction to (h ₂ + 1 μm).

The work W is machined based on the information on the in-focus position corrected in this way. That is, by irradiating the work W with the processing laser light L1 at the corrected focusing position, the processing laser light L1 can be focused to a desired depth inside the work W (in the processing process). One case).

As described above, the chromatic aberration data of the optical system to be used is measured in advance, and the offset due to the chromatic aberration of red and blue when green is used as a reference is acquired in advance. By adding this as offset data to the result of the in-focus position obtained using image data of colors other than green, the in-focus position can always be used as the green reference, and the influence of chromatic aberration of the optical system is reduced. can do. Therefore, it is possible to improve the accuracy of autofocus and the detection ability of positioning.

In the present embodiment, the colors used for image processing are determined based on the focus intensities of the R channel image, the G channel image, and the B channel image, but as in the first embodiment, the color is determined from the user interface 68. It may be determined based on the input, or it may be determined based on the hue angle data of the image as in the second embodiment.

In each embodiment, a laser dicing device using a laser beam has been described as an example, but the present invention can also be applied to a camera of a dicing device using a blade. In this case, high-precision machining can be performed by controlling the height of the blade in the Z direction based on the corrected focusing position information.

The technical scope of the present invention is not limited to the scope described in the above embodiments. The configurations and the like in each embodiment can be appropriately combined between the embodiments without departing from the spirit of the present invention.

10 ... Laser dicing device, 12 ... Stage, 20 ... Processing optics, 22 ... Processing laser light source, 24 ... Collimating lens, 26 ... Mirror, 28 ... Lens, 30 ... Observation optics, 32 ... Observation laser light source, 34 ... Collimating lens, 36 ... Half mirror, 38 ... SD color camera, 40 ... Optical unit, 42 ... Lens, 44 ... Objective lens 44, 50 ... Drive unit, 60 ... Control unit, 62 ... CPU, 64 ... Memory 64, 66 ... Input / output circuit, 68 ... User interface, 70 ... Image processing unit, 72 ... Autofocus control unit, 74 ... Processing control unit, 76 ... Bus, L1 ... Processing laser light, L2 ... Observation laser light, W ... Work

Claims (4)

An image pickup unit that captures an image of the surface of a work and acquires an image having multiple color components,
A recognition unit that collects hue angle data for each pixel from the image captured by the imaging unit and recognizes the color of the surface of the work based on the hue angle data.
Based on the result recognized by the recognition unit, a composition unit that calculates a coefficient according to the hue angle of the color of the surface of the work and synthesizes image data according to the calculated coefficient, and a composition unit.
A control unit that controls the controlled device based on the image synthesized by the composition unit, and a control unit.
A control device equipped with. The recognition unit votes the hue angle for each pixel in the hue angle voting space for the pixels exceeding the saturation threshold value, and recognizes the hue angle with the largest number of votes as the surface color of the work.
The control device according to claim 1. The control unit executes control of either autofocus or positioning.
The control device according to claim 1 or 2. An imaging step of imaging the surface of a work to acquire an image having multiple color components,
A recognition step of collecting hue angle data for each pixel from the image captured in the imaging step and recognizing the color of the surface of the work based on the hue angle data.
Based on the result recognized in the recognition step, a synthesis step of calculating a coefficient according to the hue angle of the color of the surface of the work and synthesizing image data according to the calculated coefficient, and a synthesis step.
A control step that controls the controlled device based on the image synthesized in the synthesis step, and a control step.
A control method that comprises.

Images