KR101408673B1 - Method and apparatus for detecting foreign materials on transparent substrates using image processing - Google Patents

Abstract

A method for detecting foreign matters on a transparent substrate using image processing according to the present invention comprises a step of adjusting a distant between a photographing device and a substrate at a predetermined distance; a step of radiating light to the substrate; a step of photographing the substrate with the photographing device hence the depth of focus includes only foreign matters on the substrate; a step of removing the intermediate foreign matters and the rear foreign matters from the photographed image; a step of converting the processed image to a binary image through binary coding; and a step of measuring the distribution and size of the foreign matters by extracting the attributes of the foreign matters from the binary image. Also, a device for detecting foreign matters on a transparent substrate using image processing according to the present invention comprises the substrate; a table for fixing the substrate; an radiation device for radiating the light to the substrate; the photographing device for photographing the substrate; a photographing device controller for controlling the depth of focus of the photographing device; and an image processor for processing the photographed image to determine whether the foreign matters exist. The present invention can photograph the substrate while the depth of focus includes only the foreign matters on the substrate using the photographing device controller.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for detecting foreign substances on a transparent substrate using image processing,

The present invention relates to a method for manufacturing a semiconductor substrate or a substrate in a semiconductor manufacturing process using a glass substrate used for an FPD (Flat Panel Display) such as an LCD (Liquid Crystal Display) or a PDP (Plasma Display Panel) And more particularly, to a foreign matter detection method and apparatus for detecting only a foreign matter on a surface excluding a foreign matter on the back surface.

A foreign matter may be generated for various reasons in a process using a thin transparent substrate or a laminated transparent substrate, and therefore, there is a step of detecting such foreign objects in the middle of the manufacturing process or after the process. However, there are times when it is necessary to carry out the process including the impurities such as impurities and foreign matters on the inside of the substrate except for the surface foreign matter, and the foreign matter such as the back surface foreign matter. However, since the detection apparatus using the conventional image processing detects all the foreign matter regardless of the position, there has been a problem that only the surface foreign matter can not be separated and a defect judgment is made. Especially in recent years, the thickness of the substrate used mainly in the display field has been reduced to a few hundreds of micrometers, and the above-mentioned false eyewash has occurred more frequently.

A device for inspecting the surface foreign matter of an LCD, which is a typical FPD replacing a conventional display such as a cathode ray tube, is shown in Fig. The LCD has a liquid crystal 3 between the glass substrates 1 and 2. The liquid crystal 3 is disposed between the two glass substrates and a TFT (Thin Film Transistor) 4 for driving liquid crystal and an Indium-Tin (ITO) Oxide wiring 5 and the like. In addition, the polarizing films 6 and 7 may be attached to both surfaces of the glass after the panel manufacturing process. On the surface of such an LCD, there is a case where the foreign matter 8 exists due to the process, which is detected through the optical detector 9.

As shown in FIG. 2, the position of the foreign object may be present on the surface, interior, or back side of the glass substrate. In the case of a laminated film or an attached film, In some cases. When using an existing optical detector, not only the surface of the substrate but also the inside or the back of the substrate is photographed, so it is judged as defective after the detection. Even when only the foreign substance on the surface needs to be detected, There is a problem that the image is judged to be defective after the photographing.

Japanese Unexamined Patent Application Publication No. 2000-171227 discloses an apparatus for inspecting a foreign substance on the surface of a wafer, which comprises a light projecting means for projecting dark night illumination, an image pickup means for picking up a portion to be inspected, an image storing means for creating an image, And a defect detecting means for judging whether or not the defect has occurred.

Although the above technique has the advantage of automatically detecting a foreign substance on a substrate, it is disadvantageous in that the equipment is complicated by separately providing the light emitting means and the image sensing means, and the surface and internal foreign matter can not be separately detected when the transparent substrate is used.

Japanese Unexamined Patent Application Publication No. 5-196579 discloses an apparatus for inspecting a foreign substance on a glass substrate used in a liquid crystal panel manufacturing process. In this method, only a surface foreign object is inspected using light having a wavelength that does not transmit through a glass substrate or light having a low transmittance The device is being released.

Although the above technology has an advantage of detecting only the foreign object on the surface of the glass substrate, there is a disadvantage in that it requires an expensive laser generating device such as an Eximer laser or a CO ₂ laser which does not pass through the glass. There is a problem in that the foreign object can not be detected.

1. Japanese Unexamined Patent Application Publication No. 2000-171227 2. Japanese Unexamined Patent Application Publication No. 5-196579

SUMMARY OF THE INVENTION The present invention has been conceived to solve the above-mentioned problems, and it is an object of the present invention to maintain the distance from a substrate using a displacement sensor and to use a depth of focus (DOF) ), And after shooting a picture that focuses only on the surface of the glass substrate, the image captured by the image processor is focused out of focus using automatic detection algorithms such as High Pass Filter, contrast ratio contrast, There is provided a method and apparatus for measuring the size and distribution of a foreign object by removing an image (middle layer and back surface foreign object), converting it into a binary image, and extracting attributes by processing a blob (Binary Large Object).

According to an aspect of the present invention, there is provided a method of detecting foreign objects on a surface of a transparent substrate using image processing, the method comprising: adjusting a distance between a photographing device and a substrate to a predetermined distance; irradiating light onto the substrate; The method comprising the steps of photographing the substrate so that the depth of focus includes only the substrate surface foreign object, removing middle layer foreign objects and back surface foreign objects from the photographed image, binarizing the processed image to convert it into a binary image, And a step of extracting the attribute and measuring the distribution and size of the foreign object.

In addition, in the step of removing the middle layer foreign objects and the back side foreign objects from the photographed image, the surface foreign matter detection algorithm uses a high pass filter to remove blurred images of the unfocused middle layer and back layer.

In addition, in the step of removing the intermediate layer and the back layer from the photographed image, the surface image detection algorithm removes low contrast images of the middle layer and the back layer through contrast emphasis.

Also, in the step of removing the middle layer foreign objects and the back side foreign objects from the photographed image, the surface foreign matter detection algorithm is characterized by removing low brightness images of the middle layer and the back layer using the difference in brightness between the interface and the center.

The step of extracting attributes from the binary image is characterized by using a binary image object.

According to an aspect of the present invention, there is provided an apparatus for detecting a surface of a transparent substrate using image processing, the apparatus comprising: a substrate; a table for fixing the substrate; an irradiation device for irradiating the substrate with light; A photographing apparatus controller for adjusting the depth of focus of the photographing apparatus; and an image processor for processing the photographed image to judge a foreign object part, wherein, through the photographing apparatus controller, the substrate is photographed such that the depth of focus includes only the foreign object .

Further, the table is movable in x, y and z directions.

Further, the irradiation device is characterized by being coaxial illumination or indirect illumination.

Further, the photographing apparatus is provided with a photosensor and a plurality of lenses.

Further, the photo sensor is a CCD or CMOS type.

In addition, the surface image detection algorithm of the image processor is characterized by using a high-pass filter to remove blurred images of an out-of-focus middle layer and a back surface.

In addition, the surface image detection algorithm of the image processor is characterized by removing the low contrast image of the middle layer and the back layer through contrast emphasis.

Also, the surface image detection algorithm of the image processor is characterized by removing low brightness images of the intermediate layer and the background layer using the brightness difference between the interface and the center.

The image processor may convert the intermediate layer and the background image into a binary image by binarizing the image.

The image processor may convert the intermediate layer and the background image into a binary image by binarizing the image.

According to the present invention, foreign matter such as the upper surface, middle layer, and back surface of the substrate can be inspected faster and more accurately than before. More specifically, it is possible to shorten the depth of focus adjustment time by using a simpler apparatus configuration, and to save inspection time. In addition, it is possible to detect both the middle layer and the back layer inside the substrate separately or separately by adjusting the depth of focus, thereby improving the accuracy and accuracy of the inspection.

1 is a schematic diagram for schematically showing a conventional foreign body detecting apparatus for an LCD panel.
2 is a schematic diagram for schematically showing the position of a foreign object present on a substrate;
3 is a view schematically showing an object detecting apparatus according to an embodiment of the present invention.
4 is a view schematically showing a foreign object detecting apparatus of the present invention
5A to 5C are diagrams schematically showing changes of photographed images according to the image processing procedure of the present invention
FIG. 6 is a graph showing an image processing process by emphasizing the contrast ratio of the present invention
7 is a view schematically showing a foreign object detecting process according to an embodiment of the present invention.

4 is a view schematically showing a foreign matter detecting apparatus according to the present invention.

4, the apparatus for detecting foreign materials according to the present invention includes a table 100 for fixing a substrate 110 including foreign substances, an irradiating device 200 for irradiating light to the substrate for radiography, And a photographing apparatus (300) for photographing a foreign object on the substrate.

3 is a view showing the depth of focus shown in the particle detection apparatus and the substrate of the present invention. The substrate 110 may include a surface foreign object 120, an intermediate layer foreign material 130 and a back surface foreign substance 140. As shown in FIG. 3, the depth of focus d of the photographing apparatus 300 of the present invention is The upper and lower limits thereof are adjusted so as to include only the surface foreign matter 120.

A method for detecting foreign matter on the surface of a transparent substrate according to the present invention is as follows. First, the distance between the photographing apparatus 300 and the substrate 110 is adjusted to a predetermined distance to obtain a necessary depth of focus (d). Thereafter, the substrate 110 is irradiated with light for photographing the foreign object, and then the substrate 110 is photographed by the photographing apparatus 300. Before shooting, the depth of focus d of the imaging device 300 must be adjusted to include only foreign objects on the substrate surface, since the intermediate layer 130 and the backside foreign substance 140 of the substrate are blurred by shifting out of the depth of focus d It is for this reason. If the blurred portion of the photographed image is deleted, the intermediate layer 130 and the back side foreign material 140 are automatically removed. After that, the preprocessed image is converted into a binary image, and the attribute is extracted to determine the presence or absence of foreign matter. Distribution and size can be measured.

The processing of the image photographed by the photographing apparatus 300 is automatically performed by an image processor (not shown) connected to the photosensor of the photographing apparatus 300, and the image processor processes the image photographed in the digital image photographed by the photosensor The blurred portion outside the portion is deleted and converted into a binary image to facilitate data processing. Since only the position corresponding to the foreign object has a different value in this converted image, it is easy to judge the foreign object and measure the size.

 The distance between the substrate 110 for adjusting the depth of focus (d) and the photographing apparatus 300 is performed by a photographing apparatus controller (not shown). The substrate is fixed to the table 100, and adjusts the distance between the substrate on which the photographing apparatus controller is fixed and the photographing apparatus, adjusts the distance between the lens and the sensors inside the photographing apparatus 300, (d). When inspecting a substrate having the same thickness in the inspection process, there is no need to adjust the distance 300 between the substrate 110 and the image pickup apparatus and the depth of focus (d), and therefore the inspection time does not increase due to the adjustment.

The depth of focus (d) refers to a range in which the image can be seen clearly when the photographing surface of the photographing apparatus 300, that is, the photosensor is moved in the axial direction of the photographing surface and the subject. As a result, it is the same as the range in which the subject (foreign object) is in focus (clear image) when the subject is moved in the axial direction, and often called the depth of field. When the surface foreign object 120 is located on the surface of the substrate 110 and the distance between the photographing apparatus 300 and the substrate 110 is v, the P ₁ 'plane closer to the photographing apparatus than the substrate surface and the P ₂ ' , The distance between P ₁ 'and P ₂ ' is called depth of field (depth of field). If the distance between P ₁ 'and P ₂ ' is the depth of focus (d), d, then d has the following relationship.

d = cv / D = cvF / f

c is the diameter of the blur that is assumed to have a diameter of zero. The allowable limit can be adjusted as needed. It is related to the threshold value in the image processing stage. Is used. D is the size of the photosensor (lens diameter), f is the focal length, and F is the number of F's.

According to the above equation, as the magnitude of the image sensing plane, that is, the size of the photosensor, the F value, the brightness of the lens, and the focal length of the lens become farther away ), And the closer the distance from the subject is, the smaller the depth of focus (d) becomes. Also, in the presence of a diaphragm, since the depth of focus d becomes narrower as the diaphragm is opened, it is preferable to add a diaphragm to the photographing apparatus 300 as needed.

In the present invention, it is preferable to adjust the values so that only the object (surface foreign matter) on the surface is focused, thereby keeping the depth of focus d small. Since the depth of focus d is preferably on the order of the average size of the foreign object, the size of the foreign object differs depending on the type of process and the use environment. Therefore, in the present invention, , The focus depth (d) is automatically calculated and applied. In recent years, since the glass substrate of several hundreds of micrometers is used in the display process, the depth of focus d does not fall below the center of the substrate thickness when the focus (center of the depth of focus) is accurately located on the substrate surface So that the maximum value of the depth of focus value is about the thickness of the substrate. Also, in the present invention, it is possible to detect both the intermediate layer 130 and the back surface foreign matter 140 separately or separately from each other through the adjustment of the depth of focus (d), if necessary.

The photographing apparatus controller controls the distance between the table 100 and the photographing apparatus 300 and the distance between the photosensor and the lens in the photographing apparatus 300 And the photographing condition is determined by the value of the depth of focus (d) most suitable for the environment. It is preferable to use a displacement sensor to measure and maintain the distance between the photographing apparatus 300 and the substrate 110. [

The photographed image includes an image of the intermediate layer 130 and the backside particle 140 that are out of focus in addition to the surface foreign object 120 that is in focus. In order to achieve the object of the present invention, images related to the intermediate layer 130 and the back side foreign material 140 should be removed. It is also desirable to remove them through an image processing algorithm before converting the photographed images into binary images, since there may be defective pixels in the photosensor or noise 150 by the imaging device.

The image processing algorithms that can be used at this time include using a high pass filter, using contrast emphasis, and using brightness differences between the boundary and the center.

The high-pass filter processing means a filter processing by a digital filter which passes only a high frequency (a part where a luminance change or a luminance is high). First, the high pass filter measures the luminance level by dividing the photographed image into image areas of a predetermined size. Thereafter, an image of a low frequency component (low luminance level) is removed, and the threshold value is determined by a predetermined threshold value c. In addition to the threshold value (c), since the result may vary depending on the size of the image divided into the predetermined size, it is necessary to set the size of the foreign object and the thickness of the substrate.

If necessary, the outline can be emphasized through processing such as sharpening before the high-pass filter processing. If the outline is emphasized before the high pass filter processing, since the surface foreign matter is clearly emphasized due to the focus, the image of the surface foreign matter is partially or entirely removed by the high pass filter processing, Can be reduced.

It is also possible to use dynamic thresholding. In an inspection apparatus using general visible light, unevenness of the entire image may occur due to the uniformity of the light source or the difference in intensity of light depending on the position in the sensing plane. Since the high pass filter processing using the dynamic threshold method performs noise reduction for each region It is possible to increase the detection accuracy even for an image having luminance unevenness or the like.

5A shows an image photographed first by the photographing apparatus 300. Fig. 5A, the originally photographed image falls within the depth of focus d, and the blurred image of the intermediate layer 130 or the back side foreign matter 140, the noise 150 The image of the photographed image is filtered through a high pass filtering process so that signals in a region corresponding to a low frequency, that is, a region related to the foreign object in the middle layer 130 and the backside 140, and a region corresponding to the noise 150 are filtered, By passing the image for the area of the focused foreign object 120 that is in focus, image processing is performed in which only the area of the foreign object 120 is emphasized clearly. As a result, the images of the intermediate layer 130 and the back side foreign material 140 which are blurred due to out of focus are eliminated, and the general noise 150 is also removed. 5B shows an image processed by the high pass filter. As shown in FIG. 5B, since the area outside the surface foreign object 120 is deleted after the processing, It makes it easy.

The photographed image can also remove portions related to the intermediate layer 130 and the back side foreign object 140 through the contrast emphasis algorithm. The contrast ratio emphasis basically uses a principle similar to that of the high pass filter. After reading all the brightness values of the image data through the image processor, a value lower than the threshold value (c) according to a predetermined threshold value (c) (0) and the high value is converted to the name (1).

Referring to FIG. 6, a brightness value in a photographed image by the surface 120, the middle layer 130, and the backside foreign material 140 located on the substrate 110 is indicated by a dotted line in the graph have. The image processor displays a value obtained by processing the luminance value of the largest luminance value 1 in the image photographed by the photographing apparatus 300 and relatively processing the luminance values smaller than the luminance value in a solid line graph. The surface foreign object 120 photographed in focus with a focus depth has a relatively large luminance value and the intermediate layer 130 and the back surface foreign object 140 photographed outside the depth of focus have a relatively low luminance value . In the image processor, the value smaller than the specific threshold value (c) is converted to 0 regardless of its size, and the value larger than the threshold value (c) is also changed to 1 regardless of the size, As shown by a thick solid line in FIG. 3B, all values other than the image corresponding to the surface foreign object 120 are removed, and the surface foreign matter has the same value regardless of the central portion and the peripheral portion. It makes it easy.

The photographed image can remove the portions related to the middle layer 130 and the backside foreign object 140 through the use of the brightness difference between the interface and the center. Since the single threshold value is used when the high pass filter or the contrast ratio emphasis is used, an error may occur when the brightness or brightness value of the specific region is not significantly different. When the above algorithm is used, . In an algorithm that uses a brightness change, it is preferable that the image processor first divides the photographed image and removes noise through an average filter. Thereafter, brightness values of a specific region (or pixel) and a surrounding region (or pixel) are compared with each other, and when the difference exceeds a specific threshold value, it is judged to be a foreign object. Since the algorithm using the brightness difference includes complex mathematical expressions, further explanation is omitted, but conventional brightness difference image processing can be utilized.

The preprocessed image is converted into a binary image. FIG. 5C shows an example in which an image preprocessed by the high-pass filter is converted into a binary image.

A binary image is a binary image, typically a digital image with various colors or contrast values, including pixels with only one of two values, black and white. Thus, a binary image can be stored in memory as a binary image because the values of all pixels are stored as a single bit (0 or 1). Such a binary image has merits such that data processing is easy because it has only values of 0 and 1, and compression and the like are easy. When the image is expressed in the form of an image, as shown in FIG. 5C, the binary-converted image can be represented only by black (0) and white (1), and when stored in a database or memory, . The image processor displays the presence or size of the foreign object to the operator in the form of display or the like through the processed database.

FIG. 7 is a flow chart schematically illustrating a foreign object detection process according to an embodiment of the present invention. In step S110, light is irradiated onto the substrate fixed on the table. The depth of focus is adjusted by the imaging device controller in step S120, and the surface of the substrate is imaged using the imaging device in step S130. In step S140, the image processor removes an area other than the focused surface foreign object in the photographed image. In step S150, the image processor converts the preprocessed image into a binary image. In step S160, the binary image is converted into a database by blob extraction. In step S170, the presence / absence of the foreign object, the size, and the like are output.

The configuration of a transparent substrate surface foreign object detecting apparatus using the image processing of the present invention will be described in more detail with reference to FIGS. 3 and 4. FIG.

The apparatus for detecting only the surface foreign object 120 among the foreign objects generated during the process or washing process includes a substrate 100 for fixing the substrate 110 to be processed and a table 100 for fixing the substrate 110. As the substrate 110, a glass substrate used for a typical FPD or a semiconductor wafer transmitting light can be used. The apparatus for detecting a foreign object according to the present invention includes an irradiating device 200 including a light source for irradiating light for photographing the substrate 110, Device 300 as shown in FIG. The photographing apparatus 300 includes a photosensor (not shown) and a plurality of lenses (not shown). The photographing apparatus controller 300 adjusts the relative distance between the photosensor, the lens, and the substrate Not shown). The image pickup device controller includes an image processor (not shown) for processing the image photographed by the photographing apparatus 300 to determine the presence or absence of the surface foreign object 120, So that only the foreign object can enter the focus depth (d) area. Further, since the depth of focus d of the present invention can be adjusted, it is possible to separately detect both the middle layer 130 and the back surface foreign matter 140 inside the substrate or separately if necessary.

The table 100 is movable in the x, y, and z directions with adjustment of the photographing apparatus controller while the substrate 110 is fixed, and is movable in the z direction (axial direction of the photographing apparatus and the substrate) It is preferable to use a servo motor or an actuator to precisely control the speed and displacement. The distance between the photographing apparatus 300 and the substrate 110 is preferably measured precisely using a displacement sensor or the like.

The irradiating device 200 can use coaxial illumination or indirect illumination, and it is advantageous to provide a radiating device in the radiographing device 300 so that coaxial illumination that vertically illuminates the foreign object obtains a clear image and secures illumination. In the case of indirect lighting, there is an advantage of reducing the size and weight of the photographing device by providing illumination on the table to illuminate the side surface of the substrate, but it is necessary to use a strong light source in order to secure sufficient illumination.

The photographing apparatus 300 includes a photosensor and a plurality of lenses. The photographing apparatus controller adjusts the relative distance between the photosensor and the lens and the substrate 110 so that the depth of focus is adjusted to include only the surface of the substrate . The number and arrangement of the lenses can be changed, and a detailed description of the configuration is omitted because it is a conventional optical technology. The photographing apparatus 300 has a general machine vision structure, and has a CCD or CMOS type photosensor and an objective lens. If necessary, an iris or a condenser lens can be added. The diaphragm can have a great influence on adjusting the depth of focus (d), and the depth of focus (d) becomes narrower as the diaphragm is opened.

CCD (Charge Coupled Device), which is a type of photo sensor, is also called a charge-coupled device (charge-coupled device). It is a recording device that uses charge accumulation and transfer and can be integrated with high density, Device. The basic structure of a CCD is a simple structure in which a thin oxide film is formed on a silicon substrate and a plurality of electrodes are arranged thereon. However, due to the high integration structure, there is a disadvantage that there is a lot of noise in the image obtained by the CCD.

CMOS (Complementary Metal-oxide Semiconductor) photo sensor is a low power consumption type image pickup device, which has a power consumption of about one tenth of that of a CCD, a single power supply, and noise, but has a drawback .

An image processor (not shown) preprocesses an image obtained from the photosensor by using an algorithm such as a high pass filter, a contrast ratio enhancement, a brightness difference, etc. to obtain an intermediate layer and a backside image, Thereby removing stains due to unevenness of the light source. Also, the image processor converts the preprocessed image into a binary image, and forms a database through a method such as blob extraction, and then determines whether or not there is a foreign object and outputs it to an operator.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims. Therefore, the scope of the present invention should not be limited by the illustrated embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

1,2 Glass substrate
3 liquid crystal
4 TFT
5 ITO wiring
6,7 polarized film
8 foreign matter
9 optical detector
100 tables
110 substrate
120 surface foreign matter
130 middle layer
140 back side foreign body
150 Noise
200 irradiation device
300 photographing device

Claims (16)

A method for detecting a foreign substance on a surface of a transparent substrate using image processing,
Adjusting a distance between the photographing apparatus and the substrate to a predetermined distance,
Irradiating the substrate with light,
Photographing the substrate with the imaging device such that the depth of focus includes only the substrate surface foreign object,
Removing middle layer foreign matter and backside foreign matters from the photographed image,
Transforming the preprocessed image into a binary image,
Extracting the attribute of the foreign object from the binary image and measuring the distribution and size of the foreign object
And detecting the foreign matter on the upper surface of the transparent substrate using the image processing.
2. The method according to claim 1, wherein in the step of removing the middle layer foreign objects and the back side foreign objects from the photographed image, the surface foreign matter detection algorithm uses a high pass filter to remove blurred images of the unfocused middle layer and back surface foreign matter A method for detecting a foreign substance on a top surface of a transparent substrate using a process.
2. The method according to claim 1, wherein in the step of removing the intermediate layer and the back layer from the photographed image, the surface foreign matter detection algorithm removes low contrast images of the middle layer and the back layer through emphasis of contrast ratio. A method for detecting foreign matter on a top surface of a transparent substrate.
2. The method according to claim 1, wherein in the step of removing the intermediate layer and the back layer from the photographed image, the surface image detection algorithm removes low-brightness images of the middle layer and the back layer using the difference in brightness between the interface and the center A method for detecting foreign matter on a transparent substrate using image processing.
2. The method of claim 1, wherein extracting attributes from the binary image uses a binary image object.
A transparent substrate surface foreign matter detecting device using image processing,
Board,
A table for fixing the substrate,
An irradiation device for irradiating the substrate with light,
A photographing device for photographing the substrate,
A photographing apparatus controller for adjusting a depth of focus of the photographing apparatus,
An image processor for processing the photographed image to judge a foreign object part
/ RTI >
And the substrate is photographed through the imaging device controller so that the depth of focus includes only the surface foreign matter of the substrate.
7. The apparatus according to claim 6, wherein the table is movable in x, y and z directions.
The apparatus according to claim 6, wherein the irradiation device is coaxial illumination or indirect illumination.
The transparent substrate surface foreign matter detecting apparatus according to claim 6, wherein the photographing apparatus comprises a photosensor and a plurality of lenses.
The transparent substrate surface foreign matter detecting apparatus according to claim 9, wherein the photosensor is CCD or CMOS.
11. The method according to any one of claims 6 to 10, wherein the surface image detection algorithm of the image processor uses a high pass filter to remove blurred images of unfocused middle and back surfaces. A transparent substrate surface foreign matter detecting device.
11. The method as claimed in any one of claims 6 to 10, wherein the surface image detection algorithm of the image processor removes low contrast images of the intermediate layer and the back surface through contrast enhancement, Foreign body detection device.
11. The image processing method as claimed in any one of claims 6 to 10, wherein the surface image detection algorithm of the image processor removes low-brightness images of middle and rear debris using difference in brightness between the interface and the center, A transparent substrate surface foreign matter detecting device using the transparent substrate.
12. The apparatus according to claim 11, wherein the image processor binarizes the image from which the intermediate layer and the backside foreign material have been removed, and converts the binary image into a binary image.
13. The apparatus according to claim 12, wherein the image processor binarizes an image from which the middle layer and the backside are removed, and converts the binary image into a binary image.
14. The apparatus according to claim 13, wherein the image processor binarizes the image from which the intermediate layer and the backside foreign material have been removed, and converts the binary image into a binary image.

Images