KR20090122266A - Plane substrate auto-test system and the method thereof - Google Patents

Abstract

A plane substrate auto-test system and the method thereof are disclosed. The system comprises an optical module, which comprises a lighting assembly (301, 302, 303) and lenses. Said lenses lead the lighting beam to the partial area of the substrate, and the lenses comprise a fresnel lens at least. The optical module comprises a camera (206) which receives the reflected light generated by the reaction of the lighting light sources and the substrate, and the camera comprises a time delay integral (TDI) sensor. A telemetric imaging lens (205) leads the reflected light which comes from the substrate to the camera (206). The lighting assembly (301, 302, 303) comprises a controller (330) which connects to the multiunit LED light sources (301, 302, 303), and each LED (301, 302, 303) light source can transmit light beams with different wavelengths. The controller (330) controls each LED light source (301, 302, 303) independently.

Description

Plane substrate auto-test system and the method

FIELD OF THE INVENTION The present invention belongs to the field of sensing systems, and more particularly relates to systems and methods for detecting defects in an image imprinted on a planar substrate.

Defect management is a very important process in the production and processing of large substrates, which are flat displays. Flat panel displays are mainly used for personal notebooks, computer flat panel displays, liquid crystals in mobile phones and digital devices, car driving systems, cameras, projection televisions and liquid crystal televisions, and large and small liquid crystals on various devices. Flat displays consist of two transparent substrates (usually glass). The control circuit and the optical color filter are engraved on the substrate, and the liquid crystal is filled between the two substrates. The processing of flat panel display products is very complex and must be carried out in a very clean environment, otherwise manufacturers are forced to repair or discard defective displays, since various defects are easily generated during processing. Since this increases defect rates and increases costs, the manufacturer's success is largely determined by the management and detection of board defects.

Various types of defects occur during the processing of flat panel displays: dust (drops of foreign matter on the substrate), open and short circuits, chemical residues (with chemicals remaining on the surface of the substrate), and through holes (upper and lower layers). Holes, which can lead to short circuits), but are not limited thereto. These defects can cause partial pixels of the substrate to become inoperable or the entire substrate to be inoperative.

Substrate detection during the production process aids in quality control and production process control and can reduce the loss of defective materials. Flat display sensing requires a special technical challenge, because the substrate is made of transparent materials, the lines of the surface are multi-layered, the patterns being engraved are dense, the size of the defect is very small (micro Meter level), the substrate area is too large and the detection time is very short. Automatic optical sensing systems are used in the processing of flat panel transistor liquid crystal display substrates and semiconductor integrated circuits to help diagnose product quality, reduce defect rates, and reduce production costs.

Traditional automatic optical sensing systems take pictures of a substrate with a camera and analyze the images in the photo to find defects on the substrate.The images can be used to determine the defect's coordinates (x, y coordinates, longitude, latitude, area, etc.) to provide. Analyzing the image not only reveals the location and type of defects, but also the numerical flow of defects. This information helps manufacturers manage systems to lower their failure rates.

The performance of an automatic optical sensing system depends primarily on the speed and sensitivity of the detection. Due to the constant upgrade of processing technology, the production speed has been greatly improved, the substrate is constantly growing, and the width of the pattern imprinted on the substrate is getting smaller. These requirements make automated optical sensing systems with faster sensing speeds and higher sensitivity. Nowadays, there is a need for an automatic detection system and method that can analyze large flat substrates (flat liquid crystal display substrates, semiconductor wafers, etc.) at high speed. I want to be able to improve.

SUMMARY OF THE INVENTION The present invention for solving the above problems is a kind of automatic planar substrate sensing system and method, which detects a large planar substrate at high speed, has an accurate sensitivity, and improves an image analysis rate.

It is a kind of flat substrate automatic sensing system, and its features are as follows. Lighting component: The lens set directs the luminous flux of the lighting component to a specific area of the substrate, the lens set has at least one Fresnel lens, the camera receives the illumination light and the reflected light after the substrate surface action, and the camera receives the time delay integration ( TDI) sensitive signal device.

The automatic optical sensing system and method described below is used for the detection, determination and classification of image defects imprinted on a substrate of a flat panel display. Flat panel displays include liquid crystal displays, organic light emitting diode substrates, masks and semiconductor wafers. Automated optical sensing systems and methods, collectively referred to herein as AOI systems, consist of optical modules. The optical module includes an illumination light source and a lens array to project the light of the light source onto the substrate surface. The lens array of the present invention includes at least one Fresnel lens. The optical module also includes a camera to receive the reflective, scattering or projecting portion of the light projected onto the substrate. The camera has a time delay integral sensitive signal device or linear scan charge coupled device (CCD). Telecentric lenses project reflected, scattered or projected light from the substrate onto the camera.

In the lighting section, a controller controls a number of light emitting diode (LED) light sources, all LEDs emit light of each wavelength, and the controller alone controls all the LED light sources. The illumination portion includes a front illumination light source and a back illumination light source, and all light sources include bright field and dark field light sources. Another automatic optical sensing system described herein includes a single LED light source for generating respective wavelength light waves. The feedback system is connected to the output terminal of the camera, which controls the light intensity and the gain of the camera. The camera is connected to an image acquisition processing system in one cable or optical path.

The present invention has high sensitivity and high image analysis rate, which improves defect detection sensitivity by adjusting the output intensity of each LED lamp of each wavelength, and reduces weight and volume by thinning the lens thickness by using Fresnel lens. I was. The use of TDI cameras allows for relatively short, large amounts of signal acquisition within sample time, providing a high scan rate linear scan compared to other image scanning methods.

Below is an embodiment of a sensing system and method for specific explanation and understanding. Those skilled in the art can implement this embodiment even if one or several details are missing or separate parts are used. In other words, all well known structures and operations have not been shown or described in detail.

1 is a schematic embodiment of an optical module 605 of an automatic optical sensing system.

The optical module 605 is used to detect the substrate, determine and locate the defect, and through detection, the incident intensity and integration time can be multiplied by a directly proportional defect signal to achieve this goal. The optical module 605 has a plurality of LED light sources, all of which emit light of each wavelength onto the sensed planar substrate. Although the optical module 605 in this example generates three wavelengths of light, three LED light sources can use three quantities of LEDs or wavelengths in practical applications as well. The LED 301 included in the optical module 605 of the multi-light source produces wavelength λ1 light, the wavelength of the LED 302 is λ2 light, and the wavelength of the LED 303 is λ3 light. The output luminous flux of the LED is collimated through lenses 311, 312 and 313 and merged by dichroic ray splitters 314 and 315. Dichroic ray splitter 314 reflects light at wavelength λ 2 and projects light having wavelength λ 1. The dichroic ray splitter 315 projects light with wavelengths λ 1 and λ 2 and reflects light with a wavelength λ 3.

The control unit 330 independently adjusts the intensity of the three wavelength lights. The formula of the strength I of the specimen surface is as follows:

I = I1 + 12 + I3

Where I1, 12, and I3 are the output intensities of the LED light sources whose wavelengths are λ1, λ2, and λ3, respectively.

The intensities I1, 12, and I3 of the LED light sources 301-303 are adjusted from 0 to 100% alone, so that the sensitivity of defect detection can be excellent for each sample surface.

For example, light of some wavelengths may have higher defect detection sensitivity than light of other wavelengths because the thin film optical properties and thickness plated on the substrate can affect the light reflectance of each wavelength. The control of the relative intensity of light of each wavelength can make the defect detection sensitivity excellent, but the benefits are very difficult to realize in traditional fiber optic light sources.

In addition, because of the relative weight of light at each wavelength, adjustments can be made continuously so that the optical module 605 can compensate for the nonuniformities projected by the respective spectra of the optical system and the nonuniformities of the optical response of the CCD sensor. As such, the optical module 605 can provide a truly flat illumination spectrum, which is very important for the detection of many noises on the substrate due to the uneven thickness of the surface plating film. Another source of light can be increased to achieve lighting and imagery in dark fields (not shown), just like lasers illuminate surfaces at tilt angles.

Between the filter 315 and the light splitter 307 on the illumination light path in the optical module 605 is a columnar lens 202 and a spherical lens 203. The columnar lens 202 and the spherical lens 203 through the arrangement and positioning, and controls the shape formed by the light of the light source output from each LED is projected on the substrate, in the present invention, the illumination region collected on the surface of the substrate is a narrow line shape It is a zone of. The illumination zone still needs to be further refined, at least with the aspect ratio, size, and image sensor of the pixels matched to one or all of the field of view on the substrate. The optical module 605 also includes an imaging lens 205 positioned on the beam splitter 307 reflection path. The beam splitting surface 308 of the beam splitter 307 faces the image lens 205 such that the light beam from the sample surface 204 or substrate forms an image without passing through the beam splitter 307. The optical module 605 through this method can specifically eliminate the aberration caused by the ray cracker thickness. Light from the imaging lens 205 is directed to a linear scan charge coupled device (CCD) or time delay integrated (TDI) camera 206 as described below.

2 is yet another implementation method of the optical module 202 schematic of the automatic optical sensing system.

The optical module 200 of the automatic optical sensing system includes a plurality of LED light sources, and all the LED light sources output light of each wavelength through the installation. Although the optical module 200 of the present example has three LED light sources and outputs light of three wavelengths each, the device may further arrange each LED light source and / or wavelength. The light source wavelength of the LED 301 included in the optical module 200 having a plurality of light sources is lambda 1, the light source wavelength output from the LED 302 is lambda 2, and the light source wavelength output from the LED 303 is lambda 3. The output light source is corrected through lenses 311, 312 and 323, and then merged by dichroic ray splitters 314 and 315. The dichroic ray splitter 314 reflects a ray of wavelength lambda 2 and projects a ray of wavelength lambda 1. The dichroic ray splitter 315 projects light rays having wavelengths λ 1 and λ 2 and reflects light rays having a wavelength λ 3.

The control unit 330 independently controls the light source intensities of the three wavelengths. Calculation of the specimen surface illumination intensity is as shown in FIG. 1 above. Intensities I1, 12, and I3 of the LED light sources 301-303 can be adjusted independently within the 0 to 100% intensity range, thereby improving the sensitivity of defect detection on the sample surface, respectively.

A columnar lens 202 and a spherical lens 203 are included between the filter 315 and the beam splitter 307 on the illuminated light path of the optical module 200. The columnar lens 202 and the spherical lens 203 are disposed and positioned to control the shape of the light emitted from the LED by projecting onto the substrate, and the illumination region gathered on the substrate surface is a narrow line-shaped region. Illumination zones must be further refined, and at least pixel aspect ratios, sizes, and image sensors must match one or all of the fields of view on the substrate. The optical module 200 includes an imaging lens 205 positioned on the ray path between the ray resolver 307 and the sample surface 204. The ray resolver plane of the ray resolver 307 faces the camera 206 and reflects the light 308 on the sample surface 204 directly onto the camera 206 with a linear scan charge coupled device (CCD) or time delay integrator (TDI), so that only the sample surface The light beam from 204 or the substrate does not pass through the ray resolver 307. The optical module 200 of the automatic optical sensing system that has passed this method can specifically realize the removal of aberrations caused by the ray resolver thickness.

The light module 200 is located between the light source on the light path and the light splitter 307 and includes lenses 311, 312, 313, 316, 202 and 203. All lenses 311, 312, 313, 316, 202 and 203 are arranged to collect and direct light from light sources 301-303 to the substrate surface 204. All lenses 311, 312, 313, 316, 202 and 203 are all Fresnel lenses.

3 is an embodiment of a Fresnel lens 300 included in the optical module 200.

The optical module 200 replaces the traditional lens with the polypropylene Fresnel lens 300. The Fresnel Lens 300 is a lightweight, thin piece of plastic made by direct pressing that can replace the bulk of materials required by traditional lenses. The Fresnel Lens 300 has large apertures and short focal lengths, but it doesn't have to consume heavy, bulky materials like other lenses. Compared to other types of lenses, the Fresnel Lens 300 is much lighter and has a higher light transmittance. As such, Fresnel lenses are so thin that there is very little or very little light loss due to absorption. Fresnel lens 300 consists of the same central annular structure called Fresnel, reducing the amount of material required by traditional spherical lenses. For all Fresnel spheres, the significant reduction in the overall thickness of the lens has the same effect as cutting into multiple surfaces that have the same curvature as a standard lens with continuous surfaces but are not continuous. This can greatly reduce the thickness of the lens (light weight and reduce volume), but at the cost of poor image quality. The arrangement of the light module 200 enables the Fresnel lens 300 to be used, since the TDI sensitive signal device does not read only one pixel each time in the scanning direction and read the image. As a result, the illumination lines of the optical module 200 are much wider than the image region, so that only one very small, relatively uniform line-shaped region on the surface is used. As such, poor image quality produced by the Fresnel lens does not affect the use of the Fresnel lens in the optical illumination system of the automatic optical sensing system of the present invention.

The reflected light 308 from the substrate 204 is perpendicular to the TDI camera 206 because the movement in the image direction coincides with the charge transfer direction (sensor passing through the camera 206). Referring to Figure 1, the placement of the optical module satisfies this requirement, causing incident light to shine through the light splitter 307 onto the substrate surface 204. The same beam splitter directs the scattered and / or projected light reflected back from the substrate 204 to the TDI camera 206. The imaging lens 205 is located in the light path and directs light from the substrate surface 204 to the TDI camera 206.

The beam module 200 has a more compact design that places the image lens 205 between the ray resolver 307 and the substrate surface 204, which allows the image lens 205 to collect and guide the reflected light from the substrate 204. The imaging lens 205 in the optical module 200 has a relatively small working distance. When the magnification of the optical module 200 is fixed, the distance between the imaging lens 205 and the TDI lens 206 is also shortened.

Based on the above reason, the image lens 205 included in the optical module 200 uses a telecentric lens. In general, the placement of telecentric lenses makes the principal axis of light of all points from an object or image parallel. Telecentric lenses provide coaxial video light.

Since the telecentric lens used by the automatic optical sensing system of the present invention has a fixed magnification and a geometric shape, the telecentric lens is not affected by the position of the object image whose object size is in view, and even However, it does not matter if there is a change in the distance between the object and the lens. At the same time, telecentric lenses and LED light sources were used to make the telecentric effect excellent because the LEDs can be illuminated through the telecentric lens to create parallel luminous flux. As such, the incident light and the reflected light from the substrate surface 204 pass through the image lens 205 in the same light path.

The difference with traditional lenses is that telecentric lenses have the same magnification for all distance objects in the distance lens, so telecentric lenses produce images of the same size for all distance objects and within the full field of view. Hold a fixed view. If an object is too close or too far from the telecentric lens, the focusing will be blurry, but even if the focusing image is blurry, the focusing will be the same size as the clear image.

Telecentric lenses applied to the instrument's vision system provide images of fixed size and geometry within a certain distance and within the entire field of view. Automatic optical detector vision systems using telecentric lenses have therefore solved the problems often encountered in instrument vision systems using traditional lenses, which include changes in the appearance size caused by changes in object distance, in the central field of view. Non-uniform deformations, such as, but not limited to, the use of traditional lenses (the time when the object is at the edge and the time when the object is in the center of the field of view is different).

Under normal circumstances, a substrate that needs sensing has two surfaces. It is a figure structure of the substrate and the upper layer of the bottom layer. Telecentric lenses do not receive a shadow image that is reflected from the bottom layer (eg, the substrate below the upper geometry we are interested in). Because when the incident light is incident on the substrate, the shaded image is directly under the figure structure. The image obtained by TDI is a bird's eye view of the substrate. All shadows in the lower layer are blocked by the geometry above them, and the shaded image becomes background noise during image processing. Therefore, using telecentric imaging lenses can reduce the background noise of the optical module 200 of the automatic optical sensing system.

The optical module 200 includes a TDI camera 206 for capturing a substrate image, as described above. TDI cameras are a type of linear scan camera with a TDI sensitive signaling device. Normally, TDI cameras effectively increase the integration time of the collected incident light by accumulating multiple exposures to the same object. When the object making up the image is in motion, the movement of the object and the exposure of the TDI must proceed simultaneously to ensure the quality of the image.

Because TDI cameras collect a large amount of signals within a relatively short sampling time, they can provide a sort of higher response speed linear scan compared to other video scanning methods. TDI cameras can scan at higher speeds under the same incident intensity, or at the same speed under low light conditions.

The TDI signal sensitive device includes a multiline optoelectronic probe or sensor (eg, the optoelectronic probe of line 96 to line 4). When all the photoelectron probes in a line of photoelectron probes collide with photons, a charge is generated that is directly proportional to the photon quantity. Because systems based on TDI cameras collect images of moving objects in a time-delayed multi-exposure method, an automatic optical sensing system moves the sensed substrate so that it progresses simultaneously with the linear images collected by the TDI camera. This movement causes the substrate to pass through the field of view of the TDI lens line by line each time, just as the document passes through the scanner.

As the substrate motion passes through the TDI camera, the collected linear images (part of the substrate) are moved from one line of probe to the next line of probe in sequence. At the same time, the TDI camera moves the stored charge to keep it in line with the kinetic image. Thus, as the substrate motion passes through the TDI camera, the charge in the substrate image also flows to the adjacent optoelectronic probes in order and accumulates constantly. TDI sensitive signal devices that pass this method slightly accumulate (integrate) the linear image on the sensor, allowing the image to obtain more light. The TDI camera sends the information of the linear image onto the image acquisition card, matching the information of the pixels into one complete image.

The integrated image signal has a good signal to noise ratio and a dynamic range. More integration time allows for faster acquisition of dynamic images. Furthermore, since the TDI camera can effectively and uniformly manipulate the intensity fluctuations of the DC light source, the LED light source can replace the high power, high consumption, high temperature DC halogen lamp, and lower the maintenance cost of the system. This automatic optical sensing system uses a single TDI camera 206, but other sensitive detectors are included in the optical module 200 of the automatic optical sensing system, for example, an enhanced charge coupled device (ICCD), an optoelectronic Multiplier tubes (PMT), linear scan charge-coupled devices, and complementary metal oxide semiconductors (CMOS).

The automatic optical sensing system described above senses a substrate on a bright field sensing principle. The bright field method obtains an image of the substrate surface directly from the reflected light of the substrate, with some defects (scratches, granules, etc.) having strong dark field optical response, and some defects having strong bright field optical response. Therefore, in order to effectively detect each type of defect, another automatic optical sensing system is designed to detect the substrate by supplementing the bright field method using the dark field method.

4 is a connection diagram in which the automatic optical sensing system uses a dark field sensing method.

In the dark field sensing method, incident light 601 is obliquely incident on the substrate at one or several places. Incident light 601 generates reflected light 602 and scattered light 603 after working with the substrate. Lens set 604 collects scattered light 603 from the substrate surface and directs it to the probe.

5 is a connection diagram of the automatic optical sensing system optical module 500, which uses a bright field and a dark field detection method.

The arrangement and function of the bright field sensing portion of the optical module 500 is shown in FIG. 2 (optical module 200). The optical module 500 also has a dark field light source 701 in addition to including the bright field LED light sources 301-303 described above. The dark field light source 701 includes one or more LEDs, light bulbs, fiber optic lights and laser light sources, and the like. Lens 702 collects incident light 703 from dark field light source 701 on substrate surface 204, where the dark field illumination zone overlaps the bright field illumination zone. Incident light 703 works with 204 on the substrate surface to produce reflected light 704 and scattered light 705. Image lens 205 collects scattered light 705 and images it on a TDI camera 206.

The automatic optical sensing system also uses a backlight or backside illumination to obtain the substrate image. Backlight installation places a light source underneath the substrate that requires an image. Backlighting allows certain types of defects (image defects on glass substrates) to achieve very high contrast. If the detected specimen is a glass substrate, the back light can detect defects on the glass surface and inside the glass.

6 is a connection diagram of an automatic optical sensing system using backlight illumination.

The backlight enhances the detection rate of defects that the front light source is difficult to detect. The optical module 600 of the automatic optical sensing system includes a front light source 804 and a back light source 803. The front light source 804 includes all the light sources described above. The light source is a single LED, a bulb or an optical fiber light source. The numerical aperture of the back light source 803 should match the front light source 804 to ensure that as much light as possible enters the TDI camera. Both light sources illuminate the same area of the substrate surface. The optical module 600 includes two gas-lift flotation bases 603 in vacuum, with the area between the bases 603 being the back light source 803. The pedestal includes a compressed air inlet 801 and a vacuum outlet 802, wherein the vacuum sucks down the substrate so that it can stably maintain high speed motion in gas support.

Another method of providing a backlight light source is to place a mirror of each reflectance under the glass substrate.

7 is a connection diagram of an optical module 700 of an automatic optical sensing system in which mirror surface 903 is a backlight illumination.

Light output from the front light source 804 is reflected on the mirror surface 903, and the reflected light of the mirror surface 903 provides a backlight light source. Mirror 903 is, but is not limited to, a triangular prism structure. All surfaces of the triangular prism 903 are plated with thin films having different reflectances. The backside light source mirror of different reflectivity is selected for the substrate of each property, passing through the turns or appropriately placing a triangular prism face to provide the back light source of the respective intensity. The reflected light from both the substrate surface and the mirror surface enters the TDI camera 206. The mirrored area under the substrate is much smaller than other light source (LED, light bulb, laser light, etc.) areas. Another type of automatic optical sensing system is designed using different types of illumination light sources, such as arbitrary geometry and any reflectance and / or type of film and diffuse light sources.

8 is a connection diagram of the automatic optical sensing system optical module 800 using the front light source and the back light source and the bright field and dark field detection methods.

The automatic optical sensing system optical module 800 includes a front light source 804 and a back light source 803 used in the bright field, as described above. The front light source 804 and / or the back light source 803 can be one or multiple LEDs, bulbs, or fiber optic lights. The automatic optical sensing system 800 includes a front light source 701 and a back light source 1001 used in the dark field, as described above. The dark field light source 1001 includes one or more LEDs, light bulbs, fiber optic lights, or laser light sources. The dark field light source illumination zone overlaps the bright field light source illumination zone. The front light source structure and manipulation in the optical module 800 is similar to the optical module 200 (FIG. 2) and the optical module 500 (FIG. 5) described above, respectively, when used for bright field and dark field sensing. The back light source structure and operation of the optical module 800 are similar to the optical module 600 (FIG. 6) and the optical module 700 (FIG. 7) described above.

No matter what kind of light source type or structure described above, the LED light source intensity during substrate sensing operation must remain stable. LED light sources have a relatively long service life, but their strength declines with the aging response of the LED or semiconductor. The design of the automatic optical sensing system uses a feedback system that checks and compensates for LED intensity decay by passing standard samples. Standard LEDs are used to measure LED intensity on a regular basis, and if there is a change in intensity (falling light), the feedback system adjusts the LED working current to provide the intensity required for the automatic optical sensing system.

9 is a connection diagram of an optical module 900 of an automatic optical sensing system with a feedback system.

The optical module 900 in this example includes the optical module 200 described in FIG. 2 above, but may also include all the optical modules described in the present patent. The optical module 900 includes a feedback system, which has a control unit 1102, which controls the output gain of the TDI camera 206 and the input of the LED power source 1103. Since the output of the power source 1103 is connected to the LED light source 301-303 in one, the current supplied to the LED light source can be controlled.

During the calibration operation, the optical module 900 uses the mirror surface 1101 as a reference plane, and the reflected light from the mirror surface 1101 becomes a standard for automatic calibration. The light reflected back from the mirror surface 1101 enters the sensing device of the TDI camera 206, passes through a survey, and is input to the control unit 1102. If the value of the reflected light is less than the set value, the control unit drives the power source 1103 to generate a signal or command to increase the current of the LED light source to the intensity value set value on the TDI camera 206.

Standard mirrors can also be used to calibrate TDI cameras. When the reference mirror 1101 is used, the outputs of all TDI-sensing device pixels may not be the same. Differences in pixel output may occur due to optical response of all pixels, non-uniformity of image lenses and light sources. The first step in TDI camera calibration is to measure the output of every pixel after receiving light from reference mirror 1101. Next, the output is surveyed under different intensities, which can be accomplished by lowering the intensity of the illumination or by using mirrors of different reflectivity. Survey results are used to determine the coefficients of both calibrations and the slope and deviation of every pixel or set of pixels. In surveying an actual substrate, the output of all pixels is first corrected for nonuniformity between the pixels by subtracting the deviation and then multiplying the gradient of this pixel.

In addition, the feedback system controls the gain of the TDI camera 206 sensitive signal device to maximize the dynamic range of the sensitive device. The feedback system is divided into two processes to maximize or control the dynamic range of the TDI sensitive signaling device. First, check the saturation time numeric output of the TDI sensitive signal device. 1, and then the numeric output of the image signal maximum. 2. The feedback system sets the TDI gain value to approximately the numeric output 1 divided by the numeric output 2. Keep it. This gain maximizes the dynamic range of the sensitive signal device.

The modular system of the automatic optical sensing system described in this patent can be applied to a substrate or sensing operation that is sensed through a new structure or sensitization.

10 is a connection diagram of the module 1000.

Module 1000 includes one or more components or combinations of components, for example, at least one light source 1201, two lenses 1202 and 1203 included in the module direct light flux to the substrate, and one ray resolver 1204 Reflect light rays from the TDI camera 1206. The image lens 1205 creates an image of the substrate in the TDI camera 1206, and the cable or optical fiber 1209 connected to the lens transmits the image information from the TDI camera 1206 to the image collection card 1207. The image collection card 1207 collects and analyzes the image values and provides the image processing calculator 1208 with the image values. All modules 1000 are all independent and the quantity and / or type installed on the system are all adjusted based on the specific substrate and sensing process.

For example, FIG. 12 shows three optical modules 605 included in an automatic optical sensing system.

11 is a connection diagram of an automatic optical sensing system 1100, which includes an optical sensing module 801 and a measurement module 802-806 for re-inspecting and reproducing the photograph.

The automatic optical sensing system 1100 includes multiple (e.g. three) optical sensing modules 801, and at the same time also includes an optical photo reexamination module 802, linewidth and overlay measurement module 803, thin film thickness measurement module 804, numerical integrated defect measurement module 805, regeneration Module 806 includes laser cutting and chemical vapor deposition (CVD) (or similar techniques) to reproduce short circuits and open circuit-like defects. Line width / overlay measurements require 500mm or better accuracy. Any small vibration of the substrate can affect beyond the accuracy range of the measurement result. The gas-lift vacuum pedestal of the automatic optical sensing system can ensure the measurement accuracy of the CD / Overlay because it provides suction force down, allowing the glass to move steadily and rapidly over the gas-lift. Therefore, various functions on the automatic optical sensing system 1100 can be realized. Each combination of modules described above can be used to make other designs.

Raw values generated by the TDI camera of the above-described automatic optical sensing system are sent to an image acquisition card for use in image processing. Traditional automatic optical sensing systems transmit a number of lenses by connecting a lens to a cable connected to a TDI camera and an image acquisition card. Lens connections are a type of numerical transmission agreement that require specific cable connections. Cables connected to the lens are subject to stringent restrictions on the tension, shielding and length of twisted pair cables. In practical automatic optical sensing applications, the cable length between the TDI camera (e.g. on the rail on the platform) and the image acquisition card (located on the image processing calculator) may even exceed 10 meters, using repeaters to Integrity must be maintained. In addition, because of the large volume of cables connected to the lens (approximately 15 millimeters in diameter), the wiring of these cables can be wired back and forth in many places, making them very complex and difficult. These cables can also interfere with each other due to antenna action, which can destroy the numerical signal.

Automatic optical sensing systems can use fiber optics to replace the lens and cable connected between the TDI camera and the image acquisition card. First, the electronic signal of the TDI camera is converted into an optical signal, and then the optical signal is transmitted to the optical fiber. At the image acquisition card terminal, the optical signal is converted back into an electronic signal. Fiber optics are very thin, light and soft, allowing you to connect remote components without the need for repeaters, allowing image acquisition cards or other numerical processors to be remote from the automatic optical sensing system. In addition, the optical signal is not affected by electronic disturbances and has a wider bandwidth compared to the cable connected to the lens (so higher numerical flow rates are possible).

12 is an automatic optical sensing system 1200 used for sensing large size substrates.

Platform 601, which includes the automatic optical sensing system 1200, includes a gas lift 602 and a gas lift vacuum preload bracket 603. The automatic optical sensing system 1200 includes at least one optical module 605 (or three optical modules 605), which are used to detect large substrates such as liquid crystal display (LCD) glass substrates 606, but only It is not limited. The optical module 605 includes an illumination light source and an image component, and the detailed description is as described above (the optical module 200 in FIG. 2, the optical module 500 in FIG. 5, the optical module 600 in FIG. 6, the optical module 700 in FIG. 7, and FIG. 8 optical modules 800, 9 optical modules 900)

During the sensing operation, a pixel wide substrate image is produced each time the line contacts the TDI. The long side of the linear image region of the pixel width is parallel to the linear motor axis 604 (x axis) that drives the optical head motion. The glass 606 edges move in a direction parallel to the linear motor axis 607 (y axis) and are supported by a gas lift 602 and a vacuum preload gas lift 603. The vacuum preload flotation base 603 is a glass that moves in a thin layer of gas flotation to control stringent flight altitudes. The linear motor axis 604 edge x axis drives the optical module until the entire glass is complete sensing.

After the board is loaded into the sensing system 1200 each time, the board must be precisely aligned with the coordinates of the platform, requiring a high speed, high analysis rate, high accuracy and long travel distance. Aligning high-speed substrates accurately ensures high throughput (or shorter turnaround times) and higher production capacity, while long travel distances can accommodate large load errors. The high analytical and high accuracy drivers in the automatic optical sensing system allow precise aiming under the microscope, enabling accurate aiming and position registration.

13 is a connection diagram of a sensing system using a high speed driver.

The high speed driver 1301 includes, but is not limited to, an air driver, a voice coil driver, a linear motor driver and a spiral tube. The high speed driver 1301 includes a pre- aiming 1304 and is used for aiming or positioning the substrate 1305. The substrate is loaded above 1304 prior to aiming by the action of the spring or its own gravity. Before aiming 1304 is in contact with the substrate 1305 and driven by the high speed driver 1301. The position strong stop 1302 before aiming is controlled by a high analysis rate driver 1303. The driving force generated in the high speed driver 1301 is less than the tension in the opposite direction generated in the high analysis rate driver 1303. The position strong stop 1302 controlled by the high analysis rate driver 1303 determines the final position before aiming or the accuracy and analysis rate of the aiming.

The sensing systems and methods described herein are all possible in all programmable circuits, field programmable gate arrays (FPGAs), programmable array logic (PAL) devices, electronic programmable logic and memory, and devices based on batteries as standard. And programmable semiconductors (PLDs), which are dedicated to the application of integrated circuits.

In addition, methods for enabling sensing systems and methods include embedded microprocessors (electronic erase and programmable read memory (EEPROM)), embedded microprocessors, firmware, software, and the like. Further, this sensing system and method can be realized by software-based circuit simulators, discrete logic elements (in combination with continuous ones), user-configurable elements, fuzzy logic (in a neural net), a mixture of quantum elements and various other elements. Can be. Of course, there are various types of combinations in the device technology that is currently being developed, for example, complementary metal oxide semiconductor (CMOS), emitter coupling logic (ECL), polymer technology (bonded polymer and metal-bonded polymer-metal structure). , Metal oxide semiconductor field effect transistor (MOSFET) technology, such as simulation and numerically mixed devices.

It should be noted that the combinations of the various sensing systems and methods of the present invention have been described using a calculator aided design tool, and are shown as a medium that can be read by a variety of calculators in numerical and / or command manner. Transitions, logic combinations, transistors, plate designs, and / or other features. The calculator reads, but is not limited to, numbers and / or instruction set words in the media formatting. For example, a fixed number of storage media (such as optical, electromagnetic or semiconductor storage media) and formatted values that can be transmitted by radio, optical or cable or combinations thereof and / or carriers of a set of instructions.

The transmission of formatted values and / or command sets over a carrier is, by way of example and not limited to, transmitted via a connection or combination of some or several types of numerical transfer agreements (HTTP, FTP, SMTP, etc.). (Upload, download, email, etc.) Coupling or connection methods include, but are not limited to, wired, wireless, and wired / wireless hybrid connections. Further, the connection scheme may include, but is not limited to, various nets and / or net combinations (not shown) that provide communication services. The mentioned net and corresponding net combinations include, but are not limited to, local area networks (LANs), metropolitan area networks (MANs), wide area networks (WANs), private networks, terminal networks and the Internet. Expressions resulting from these numbers and / or instructions of the systems and methods described above, when passed to a readable medium through one or more calculators within the calculator, are passed through the processor unit of the executable program and the calculator associated with it ( One or more arithmetic processing units).

Unless otherwise specified, the meanings included in all descriptions similar to “include” and “include” are the opposites of “only” or “thoroughly,” including but not limited to: It is also an interpretation. Other pronouns similar to “here,” “under”, “above”, and “below” are all citations of the present invention and do not refer to any part of the invention. If you apply a list of two or more items with “or”, the meaning means any item in the list, every item and every combination of each item.

The specific description and methods of the sensing system described above are not intended to be exhaustive or to limit the described systems and methods. For specific realizations, for example, sensing systems and methods are described here for purposes. Each of these modifications is also possible in a range of other sensing systems and methods, as acknowledged by well-known technology. The sensing systems and methods provided herein may be utilized in other processing systems and methods, and are not the only sensing systems and methods described above.

Various examples, elements, and actions described above can be combined to provide more advanced examples. Developing on the basis of the above detailed description, various developments are possible for the detection system and method.

1 is a connection diagram of an optical module embodiment of the present inventors automatic optical sensing system.

Figure 2 is a connection diagram of another embodiment of an optical module of the present inventors automatic optical sensing system.

3 is a structural diagram of a sample embodiment of a Fresnel lens of the optical module of the present inventors automatic optical sensing system.

4 is a connection diagram of an embodiment of the present invention using the dark field method in an automatic optical sensing system.

5 is a connection diagram of an embodiment in which the optical module of the present invention employs a bright field and a dark field.

6 is a connection diagram of an embodiment of an optical module using backlight illumination of the present inventors automatic optical sensing system.

FIG. 7 is a connection diagram of an embodiment of an optical module illuminated by a backlight using a reflector of the present invention.

8 is a connection diagram of an embodiment in which the optical module of the present inventors automatic optical sensing system uses a combination of front light, backlight illumination, bright field and dark field detection.

9 is a connection diagram of an optical module embodiment of a feedback system possessed by the present inventors' automatic optical sensing system.

10 is a connection diagram of the present inventors optical optical sensing system module.

Fig. 11 is a connection diagram of the present invention's automatic optical sensing system, in which the system is composed of respective optical modules, each having monitoring and surface surveying functions.

12 is a connection diagram of an embodiment of the present invention used in an automatic optical sensing system for sensing a large substrate.

Figure 13 is a connection diagram of a high speed driver embodiment of the present inventor's sensing system.

Claims (45)

In a flat substrate automatic sensing system, Lighting components; A lens set for guiding the luminous flux of the lighting component to a particular sphere of the substrate, the lens set comprising at least one Fresnel lens, A camera for receiving reflected light after illumination and substrate assertion, the camera comprising a time delay integration (TDI) sensitive signal device. In the planar substrate automatic detection system of claim 1, The lighting component includes one multi-set LED light source, wherein all LED light sources emit light of respective wavelengths. In the planar substrate automatic detection system of claim 2, The controller is coupled to a multi-set LED light source, independently controlling all the LED light source, the planar substrate automatic detection system. In the planar substrate automatic detection system of claim 1, And a beam splitter is between the lens set and the substrate, the beam splitter being a dichroic beam splitter. In the planar substrate automatic detection system of claim 1, And the imaging lens is between the beam splitter and the substrate, wherein the imaging lens is a telecentric lens. In the planar substrate automatic detection system of claim 1, And the lens set comprises a plurality of Fresnel lenses. In the planar substrate automatic detection system of claim 1, And said lighting component comprises a bright field light source and a dark field light source. In the planar substrate automatic detection system of claim 7, And the bright field light source is a first front light source on a first side of the substrate. In the planar substrate automatic detection system of claim 8, And wherein the bright field light source has a back light source on a second side of the substrate, the second side opposite the first side. In the planar substrate automatic detection system of claim 9, And the first rear-sided light source is a mirror on one side, and reflects the light source emitted by the lighting component. In the planar substrate automatic detection system of claim 7, And the dark field light source comprises a second front light source on a first side of the substrate. In the planar substrate automatic detection system of claim 11, And the dark field light source includes a second backside light source on the second side of the substrate, the second side opposite the first side. In the planar substrate automatic detection system of claim 1, And a feedback system coupled with the camera output. In the planar substrate automatic detection system of claim 13, And the feedback system controls the lighting component. 13. The planar substrate automatic sensing system of claim 13, And the feedback system controls the gain of the camera. In the planar substrate automatic detection system of claim 1, Image acquisition card: A planar substrate auto-sensing system connected to one optical fiber, camera and said image acquisition card. In the planar substrate automatic detection system of claim 1, A gas lift pedestal transmits the substrate, and a movement mechanism causes relative movement of the substrate. The flat substrate automatic sensing system of claim 17, Vacuum preload gas flotation base for stable support of the saik substrate. In the planar substrate automatic detection system of claim 1, Planar substrate auto-sensing system including high speed, high analysis rate, high accuracy and long distance drive. A flat substrate automatic sensing system according to claim 1. Wherein the camera comprises one or more linear scan charge coupled devices (CCDs), enhanced CCDs, photomultiplier (PMT) arrays or complementary metal-oxide-semiconductor (CMOS) probes. In the planar substrate automatic detection system of claim 1, And said lighting component comprises an LED light source, said LED light source emitting light of various wavelengths. In the planar substrate automatic detection system of claim 1, The substrate may include one or more liquid crystal display (LCD) panels, flat panel displays (FPDs), organic light emitting diode (OLED) substrates, and curtain plates. Or a semiconductor wafer. In a flat substrate automatic sensing system, Lighting parts: A lens set for guiding the luminous flux of the lighting component to a specific area of the substrate, the lens set including at least one Fresnel lens; And An imaging lens positioned between the lens set and the substrate, wherein the imaging lens is a telecentric lens. 24. The planar substrate automatic sensing system of claim 23, wherein The fixed camera receives an illumination light source and reflected light after substrate assertion, and the camera includes one or more time delay integrated (TDI) sensitive signaling devices, a linear scan CCD, an ICCD, a PMT array and a CMOS detector. Detection system. 24. The planar substrate automatic sensing system of claim 23, wherein The controller included in the lighting component is connected to a multi-set LED light source, all the LED light source emits light of each wavelength, the controller independently controls all the LED light source. 24. The planar substrate automatic sensing system of claim 23, wherein And said lighting component comprises a bright field light source and a dark field light source. 27. The planar substrate automatic sensing system of claim 26, Wherein said lighting component comprises at least one front light source and at least one back light source. 28. The planar substrate automatic sensing system of claim 27, wherein And the mirror surface included by the at least one back light source reflects the light rays of the illumination system. 24. The planar substrate automatic sensing system of claim 23, wherein A feedback system coupled to the camera output, the feedback system controlling gain of the camera with one or more lighting components. The planar substrate automatic sensing system of claim 23, wherein: And an optical fiber coupled to the camera and at least one image processor. 24. The planar substrate automatic sensing system of claim 23, wherein And a transfer device transmits the substrate. 24. The planar substrate automatic sensing system of claim 23, wherein Planar substrate auto-sensing system, including drivers of high speed, high analysis rate, high accuracy and long distance travel. 24. The planar substrate automatic sensing system of claim 23, wherein And said illumination component comprises an LED light source, said LED light source emitting light of various wavelengths. In a kind of flat substrate automatic detection method, Lighting occurs: Using a Fresnel lens, the beam is directed to a section of the substrate: Receiving incident light and reflected light after the substrate has been applied using a time delay integrated (TDI) sensitive signal device: and And collecting the reflected light signal to produce an image of the substrate. The method of claim 34, wherein the automatic detection of the planar substrate And using the imaging lens to guide the reflected light to the TDI sensitive signal device, wherein the imaging lens is a telecentric lens. The method of claim 34, wherein the automatic detection of the planar substrate Generation of the illumination light source includes generation of light rays having multiple wavelengths. The method of claim 34, wherein the automatic detection of the planar substrate Independently controlling multiple light sources corresponds to the respective multiple wavelengths, flat substrate automatic detection method. The method of claim 34, wherein the automatic detection of the planar substrate The generation of the illumination light source comprises the generation of a bright field illumination light source and a dark field light source. The method of claim 34, wherein the automatic detection of the planar substrate The generation of the illumination light source includes the generation of one or more front and back illumination light sources. The method of claim 34, wherein the automatic detection of the planar substrate The generation of the backside illumination light source includes a method of reflection of the frontside light source. The method according to claim 34, wherein the planar substrate automatic detection method, Generating an image optical signal from an electronic signal of the image; And And automatically transmitting the optical signal. The method of claim 34, wherein the automatic detection of the planar substrate And a method for detecting the substrate defect using an image value. In a type of flat substrate automatic detection method, Generation of an illumination source; Directing light rays to the substrate using a Fresnel lens; An image lens is used to guide the reflected light to an image sensor, the reflected light being the result after an illumination light source and the substrate assertion, the image lens being a telecentric lens, and using the reflected light to produce the substrate image Board automatic detection method. The method according to claim 43, wherein the planar substrate automatic detection method And detecting said substrate defect using said image. The method according to claim 43, wherein the planar substrate automatic detection method The lens sensor comprises one or more time delay integrated (TDI) sensitive signaling devices, a linear scan CCD, an ICCD, a PMT array and a CMOS probe.

Images