JP7021886B2 - Board inspection equipment, board processing equipment, board inspection method and board processing method - Google Patents

Description

The present invention relates to a substrate inspection device for inspecting a substrate, a substrate processing apparatus, a substrate inspection method, and a substrate processing method.

In the substrate processing apparatus, the substrate horizontally supported by the spin chuck is rotated. In this state, a coating liquid such as a resist liquid is discharged to the central portion of the upper surface of the substrate to form a coating film on the entire surface of the substrate. After the coating film is exposed, it is developed to form a predetermined pattern on the coating film. Here, if the surface of the substrate is in a non-uniform state, the state after exposure varies depending on the portion of the substrate, and processing defects of the substrate occur. Therefore, the surface condition of the substrate may be inspected.

Patent Document 1 describes an inspection device for macro-inspecting a sample such as a semiconductor wafer. In the inspection device, the illumination light extending linearly in the X direction parallel to the surface of the sample is irradiated toward the sample placed on the stage, and the light reflected from the linear region on the surface of the sample is emitted. An image is formed on the light receiving surface of the detector (line sensor camera) by the imaging lens. By moving the stage in the Y direction, which is orthogonal to the X direction and parallel to the surface of the sample, the light reflected by the plurality of linear regions on the surface of the sample is imaged by the detector. This produces an entire image of the surface of the sample. The quality of the sample is judged based on the brightness value of the generated image.

Japanese Unexamined Patent Publication No. 2015-127653

According to the above-mentioned inspection device, a visual inspection can be performed using an image of a sample generated by imaging. However, in an image of a sample produced by imaging, a plurality of parts that should actually be represented by the same color, brightness or density may be represented by different colors, brightness or densities from each other. In this case, it is difficult to determine the presence or absence of defects in the surface state of the sample on the image.

An object of the present invention is to provide a substrate inspection apparatus, a substrate processing apparatus, a substrate inspection method, and a substrate processing method that enable a user to easily and accurately determine the presence or absence of a defect in a visual inspection using an image of a substrate. It is to be.

(1) The substrate inspection apparatus according to the first invention is an imaging unit that captures an image of a holding portion that holds a substrate and one surface of the substrate held by the holding portion, and generates real image data representing an image of one surface of the substrate. And, for each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and a plurality of pixels are calculated. A smoothing unit that performs smoothing processing on the average pixel value of the unit area of the above, and an average pixel value after smoothing processing of a predetermined unit area among a plurality of unit areas and after smoothing processing of each unit area. The difference from the average pixel value of the It is provided with a correction unit for adding.

In the substrate inspection device, one surface of the substrate held by the holding portion is imaged to generate real image data representing an image of one surface of the substrate. For each of the plurality of unit areas, the average value of the plurality of pixels constituting the unit area is calculated as the average pixel value, and the smoothing process is performed on the average pixel value of the plurality of unit areas.

The difference between the average pixel value of the plurality of unit areas after the smoothing process of the predetermined unit area and the average pixel value after the smoothing process of each unit area is calculated. The difference corresponding to the unit region is added to the value of each pixel in the band-shaped region extending parallel to the second radial direction from each unit region. As a result, in the image based on the corrected actual image data, the average value of the plurality of pixels in each band-shaped region becomes equal to or substantially equal to the average pixel value of the predetermined unit region. Therefore, in the image based on the corrected actual image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate in the first diameter direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate in a visual inspection using an image based on the corrected actual image data.

(2) The smoothing process may be a smoothing process by the mobile median method.

In this case, the smoothing process by the moving median method appropriately reduces the variation in the local average pixel value.

(3) A film having a periodic pattern in the first diameter direction is formed on one surface of the substrate, and the width used in the smoothing process by the moving median method is larger than the period of the pattern in the first diameter direction. It may be large.

In this case, the variation in the average pixel value due to the pattern of the film formed on one surface of the substrate is appropriately reduced.

(4) Each pixel constituting each of the plurality of unit areas includes R pixel, B pixel and G pixel, the smoothing unit performs smoothing processing for each pixel type, and the correction unit is a pixel type. A plurality of differences may be calculated for each pixel type, and a plurality of differences may be added for each pixel type.

In this case, it is suppressed that the average hues of the plurality of portions on one surface of the substrate in the first diameter direction are visually recognized as different on the image based on the corrected actual image data.

(5) The substrate inspection apparatus according to the second invention is an imaging unit that captures an image of a holding portion that holds the substrate and one surface of the substrate held by the holding portion, and generates actual image data representing an image of one surface of the substrate. And, for each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the plurality of units are calculated. The smoothing unit that performs the first smoothing process by the moving median method, the average pixel value before the first smoothing process of each unit area, and the average pixel value after the first smoothing process A deviation calculation unit that calculates the deviation from the average pixel value of A plurality of deviations after the third smoothing process are obtained by performing a third smoothing process by the movement minimum method for the deviations of the plurality of unit areas and the deviation maximum value calculation unit that calculates the deviation as a plurality of maximum deviation values. The deviation minimum value calculation unit that calculates as the deviation minimum value of, and the difference maximum value calculation unit that calculates the sum of the average pixel value and the deviation maximum value after the first smoothing process of each unit area as the difference maximum value. And the difference minimum value calculation unit that calculates the sum of the average pixel value and the deviation minimum value after the first smoothing process of each unit area as the difference minimum value, and the difference corresponding to the predetermined unit area. The reference range determination unit that determines the range from the minimum value to the maximum difference value as the reference range, and the range from the minimum difference value to the maximum difference value corresponding to each unit area are corrected so as to match the reference range, and each It is provided with a correction unit that corrects the value of each pixel in each band-shaped region extending parallel to the second radial direction orthogonal to the first radial direction from the unit region so as to fit the corrected range.

In the substrate inspection device, one surface of the substrate held by the holding portion is imaged to generate real image data representing an image of one surface of the substrate. For each of the plurality of unit areas, the average value of the plurality of pixels constituting the unit area is calculated as the average pixel value, and the average pixel value of the plurality of unit areas is subjected to the first smoothing process by the moving median method. Is done. In this case, the variation in the local average pixel value is appropriately reduced.

The difference between the average pixel value before the first smoothing process and the average pixel value after the first smoothing process in each unit region is calculated as a deviation. A second smoothing process is performed on the deviations of the plurality of unit regions by the moving maximum method, and the plurality of deviations after the second smoothing process are calculated as a plurality of maximum deviation values. Further, a third smoothing process is performed on the deviations of the plurality of unit regions by the movement minimum method, and the plurality of deviations after the third smoothing process are calculated as the plurality of deviation minimum values.

The maximum difference value is calculated by adding the average pixel value after the first smoothing process of each unit area and the maximum deviation value, and the average pixel value after the first smoothing process of each unit area is calculated. And the minimum deviation value are added to calculate the minimum difference value.

The range from the minimum difference value to the maximum difference value corresponding to the predetermined unit area is determined as the reference range. The range from the minimum difference value to the maximum difference value corresponding to each unit area is corrected so as to match the reference range, and each pixel in each band-shaped area extending parallel to the second radial direction from each unit area. The value is corrected to fit the corrected range. As a result, in the image based on the corrected actual image data, the average value of the plurality of pixels in each band-shaped region becomes equal to or substantially equal to the average pixel value of the predetermined unit region. Further, the value of each pixel in each band-shaped region is represented so as to conform to the reference range. Therefore, in the image based on the corrected actual image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate in the first diameter direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate in a visual inspection using an image based on the corrected actual image data.

(6) A film having a periodic pattern in the first radial direction is formed on one surface of the substrate, and the first smoothing treatment by the moving median method, the second smoothing treatment and the movement by the moving maximal method are performed. The width used in the third smoothing process by the minimum method may be larger than the period of the pattern in the first diametrical direction.

In this case, variations in the average pixel value, the maximum deviation value, and the minimum deviation value due to the pattern of the film formed on one surface of the substrate are appropriately reduced.

(7) Each pixel constituting each of the plurality of unit areas includes R pixel, B pixel, and G pixel, the smoothing unit performs the first smoothing process for each pixel type, and the deviation calculation unit performs the first smoothing process. , Multiple deviation calculation processes are performed for each pixel type, the deviation maximum value calculation unit performs a plurality of difference maximum value calculation processes for each pixel type, and the deviation minimum value calculation unit performs each pixel type. The calculation process of a plurality of difference minimum values is performed, the difference maximum value calculation unit performs a plurality of difference maximum value calculation processes for each pixel type, and the difference minimum value calculation unit performs a plurality of difference minimum values for each pixel type. The value calculation process is performed, the reference range determination unit performs the reference range determination process for each pixel type, and the correction unit performs the difference minimum value to the difference maximum value corresponding to each unit area for each pixel type. A correction process for matching the range with the reference range and a correction process for matching the value of each pixel in each band-shaped region to the corrected range may be performed.

In this case, it is suppressed that the average hues of the plurality of portions on one surface of the substrate in the first diameter direction are visually recognized as different on the image based on the corrected actual image data.

(8) The smoothing unit is an average to be calculated for each of the plurality of unit regions located at one end in the first radial direction of the substrate on the image based on the actual image data, which is a part of the plurality of unit regions. The target pixel value is estimated based on a plurality of average pixel values calculated for a plurality of unit regions located in a portion adjacent to the one end portion, and is divided into a plurality of unit regions located in the one end portion based on the estimation result. A plurality of average pixel values corresponding to each may be determined.

According to the above configuration, the plurality of average pixel values of the plurality of unit regions located at one end of the substrate on the image based on the actual image data are the plurality of unit regions located at the portions adjacent to the one end. Estimated based on multiple average pixel values. Based on the estimation result, a plurality of average pixel values corresponding to each of the plurality of unit regions located at one end are determined. As a result, in the image based on the corrected actual image data, it is suppressed that the average color of one end of the substrate is visually recognized as being significantly different from the average color of the portion adjacent to the one end.

(9) The image pickup unit has a linear image pickup region extending parallel to the first radial direction on one surface of the substrate held by the holding portion, and the substrate inspection device holds the image pickup region by the holding portion. A moving unit that relatively moves the image pickup unit and the holding unit may be further provided so as to pass through one surface of the substrate in the second diameter direction.

In this case, the relative movement between the image pickup unit and the holding unit generates real image data representing the entire image of one surface of the substrate. Therefore, the increase in size of the image pickup unit is suppressed.

(10) The substrate processing apparatus according to the third invention inspects a coating treatment unit that forms a film on one surface of the substrate by supplying a treatment liquid to one surface of the substrate, and a substrate on which the film is formed by the coating treatment unit. The above-mentioned substrate inspection device and a transfer device for transporting a substrate between the coating processing unit and the substrate inspection device are provided.

In the substrate processing apparatus, the surface condition on one surface of the substrate on which the film is formed is inspected by the above-mentioned substrate inspection apparatus. This makes it possible to easily and accurately determine the presence or absence of defects in the film formed on one surface of the substrate in a visual inspection using an image showing one surface of the substrate. As a result, the occurrence of processing defects on the substrate is reduced.

(11) The substrate inspection method according to the fourth invention includes a step of capturing an image of one surface of the substrate held by the holding portion and generating real image data representing an image of one surface of the substrate, and an image based on the actual image data. For each of the plurality of unit regions arranged in the first radial direction of the substrate, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the average pixel value of the plurality of unit regions is calculated. Calculates the difference between the step of performing the smoothing process and the average pixel value after the smoothing process of a predetermined unit area among a plurality of unit areas and the average pixel value after the smoothing process of each unit area. It also includes a step of adding the difference corresponding to the unit region to the value of each pixel in the band-shaped region extending parallel to the second diametrical direction orthogonal to the first diametrical direction from each unit region.

In the substrate inspection method, one surface of the substrate held by the holding portion is imaged to generate real image data representing an image of one surface of the substrate. For each of the plurality of unit areas, the average value of the plurality of pixels constituting the unit area is calculated as the average pixel value, and the smoothing process is performed on the average pixel value of the plurality of unit areas.

The difference between the average pixel value of the plurality of unit areas after the smoothing process of the predetermined unit area and the average pixel value after the smoothing process of each unit area is calculated. The difference corresponding to the unit region is added to the value of each pixel in the band-shaped region extending parallel to the second radial direction from each unit region. As a result, in the image based on the corrected actual image data, the average value of the plurality of pixels in each band-shaped region becomes equal to or substantially equal to the average pixel value of the predetermined unit region. Therefore, in the image based on the corrected actual image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate in the first diameter direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate in a visual inspection using an image based on the corrected actual image data.

(12) The substrate inspection method according to the fifth invention includes a step of capturing an image of one surface of the substrate held by the holding portion and generating real image data representing an image of one surface of the substrate, and an image based on the actual image data. For each of the plurality of unit regions arranged in the first radial direction of the substrate, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the average pixel value of the plurality of unit regions is calculated. The difference between the step of performing the first smoothing process by the moving median method and the average pixel value before the first smoothing process and the average pixel value after the first smoothing process of each unit region is used as a deviation. A step of calculating, a step of calculating a plurality of deviations after the second smoothing process as a plurality of deviation maximum values by performing a second smoothing process by the moving maximum method for deviations of a plurality of unit regions, and a plurality of steps. A step of calculating a plurality of deviations after the third smoothing process as a plurality of deviation minimum values by performing a third smoothing process by the movement minimum method for the deviation of the unit area of The step of calculating the sum of the average pixel value and the maximum deviation value after the smoothing process as the maximum difference value, and the addition of the average pixel value and the minimum deviation value after the first smoothing process of each unit area. The step of calculating the value as the minimum difference value, the step of calculating the range from the minimum difference value corresponding to the predetermined unit area to the maximum difference value as the reference range, and the difference from the minimum difference value corresponding to each unit area. After correcting the range up to the maximum value so that it matches the reference range, and correcting the value of each pixel in each band-shaped region extending parallel to the second diametrical direction orthogonal to the first diametrical direction from each unit area. Includes steps to correct to fit the range of.

In the substrate inspection method, one surface of the substrate held by the holding portion is imaged to generate real image data representing an image of one surface of the substrate. For each of the plurality of unit areas, the average value of the plurality of pixels constituting the unit area is calculated as the average pixel value, and the average pixel value of the plurality of unit areas is subjected to the first smoothing process by the moving median method. Is done. In this case, the variation in the local average pixel value is appropriately reduced.

The difference between the average pixel value before the first smoothing process and the average pixel value after the first smoothing process in each unit region is calculated as a deviation. A second smoothing process is performed on the deviations of the plurality of unit regions by the moving maximum method, and a plurality of deviations after the second smoothing process are generated as a plurality of maximum deviation values. Further, a third smoothing process is performed on the deviations of the plurality of unit regions by the movement minimum method, and a plurality of deviations after the third smoothing process are generated as a plurality of deviation minimum values.

The maximum difference value is calculated by adding the average pixel value after the first smoothing process of each unit area and the maximum deviation value, and the average pixel value after the first smoothing process of each unit area is calculated. And the minimum deviation value are added to calculate the minimum difference value.

The range from the minimum difference value to the maximum difference value corresponding to the predetermined unit area is calculated as the reference range. The range from the minimum difference value to the maximum difference value corresponding to each unit area is corrected so as to match the reference range, and each pixel in each band-shaped area extending parallel to the second radial direction from each unit area. The value is corrected to fit the corrected range. As a result, in the image based on the corrected actual image data, the average value of the plurality of pixels in each band-shaped region becomes equal to or substantially equal to the average pixel value of the predetermined unit region. Further, the value of each pixel in each band-shaped region is represented so as to conform to the reference range. Therefore, in the image based on the corrected actual image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate in the first diameter direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate in a visual inspection using an image based on the corrected actual image data.

(13) The substrate processing method according to the sixth invention is a step of forming a film on one surface of the substrate by supplying a treatment liquid to one surface of the substrate, and a film is formed on one surface by using the above-mentioned substrate inspection method. Includes steps to inspect the substrate.

In the substrate processing method, the surface condition on one surface of the substrate on which the film is formed is inspected by the above-mentioned substrate inspection method. This makes it possible to easily and accurately determine the presence or absence of defects in the film formed on one surface of the substrate in a visual inspection using an image showing one surface of the substrate. As a result, the occurrence of processing defects on the substrate is reduced.

According to the present invention, the user can easily and accurately determine the presence or absence of a defect in a visual inspection using an image of a substrate.

It is external perspective view of the substrate inspection apparatus which concerns on 1st Embodiment. It is a schematic side view which shows the internal structure of the substrate inspection apparatus of FIG. It is a schematic plan view which shows the internal structure of the substrate inspection apparatus of FIG. (A) is a diagram showing an example of an actual image showing one surface of a substrate, and (b) is a diagram showing an example of a conventional image based on image data. FIG. 3 is a schematic plan view showing a state in which an entire surface of a substrate is imaged by an imaging unit in the substrate inspection device of FIG. 1. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 1st Embodiment. It is a block diagram which shows the control system of the substrate inspection apparatus which concerns on 1st Embodiment. It is a flowchart of the defect determination process which concerns on 1st Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 2nd Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 2nd Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 2nd Embodiment. It is a figure for demonstrating the method of generating the image display data which concerns on 2nd Embodiment. It is a block diagram which shows the functional structure of the image data generation part for display which concerns on 2nd Embodiment. It is a flowchart of the defect determination process which concerns on 2nd Embodiment. It is a schematic block diagram which shows the whole structure of the substrate processing apparatus which concerns on 3rd Embodiment.

Hereinafter, the substrate inspection apparatus, the substrate processing apparatus, the substrate inspection method, and the substrate processing method according to the embodiment of the present invention will be described with reference to the drawings. In the following description, the substrate is a semiconductor substrate, a substrate for FPD (Flat Panel Display) such as a liquid crystal display device or an organic EL (Electro Luminescence) display device, a substrate for an optical disk, a substrate for a magnetic disk, a substrate for a magneto-optical disk, and the like. A substrate for a photomask, a ceramic substrate, a substrate for a solar cell, or the like. Further, the substrate used as an inspection target in the present embodiment has one surface (main surface) and the other surface (back surface), and a film having a periodic pattern in two directions orthogonal to each other is formed on the one surface. Has been done. The period of the pattern in each direction is determined according to the size (die size) of the range that can be exposed in one shot in the exposure apparatus that exposes the substrate. Examples of the film formed on one surface of the substrate include a resist film, an antireflection film, and a resist cover film.

[1] First Embodiment (1) Configuration of Board Inspection Device FIG. 1 is an external perspective view of the board inspection device according to the first embodiment, and FIG. 2 is an internal view of the board inspection device 200 of FIG. It is a schematic side view which shows the structure of, and FIG. 3 is a schematic plan view which shows the internal structure of the substrate inspection apparatus 200 of FIG. As shown in FIG. 1, the substrate inspection device 200 has a housing portion 210. The housing portion 210 includes a rectangular bottom surface portion 211 and four rectangular side surface portions 212 to 215. The side surface portions 212 and 214 are located at both ends of the bottom surface portion 211 in the longitudinal direction, and the side surface portions 213 and 215 are located at both ends of the bottom surface portion 211 in the lateral direction (width direction), respectively. The housing portion 210 has a substantially rectangular upper opening. The housing portion 210 may further include an upper surface portion that closes the upper opening.

Hereinafter, the lateral direction of the bottom surface portion 211 is referred to as a left-right direction, and the longitudinal direction of the bottom surface portion 211 is referred to as a front-rear direction. Further, in the left-right direction, the direction from the side surface portion 215 to the side surface portion 213 is defined as the right side, and the opposite direction is defined as the left side. Further, in the front-rear direction, the direction from the side surface portion 214 to the side surface portion 212 is defined as the front, and the opposite direction is defined as the rear. A slit-shaped opening 216 for transporting the substrate W between the outside and the inside of the housing portion 210 is formed in a portion extending from the side surface portion 212 to the front portion of the side surface portion 213.

A light projecting unit 220, a reflecting unit 230, an imaging unit 240, a substrate holding device 250, a moving unit 260, and a notch detection unit 270 are housed in the housing unit 210.

The light projecting unit 220 includes, for example, one or a plurality of light sources, and is attached to the inner surface of the side surface portions 213 and 215 of the housing portion 210 so as to extend in the left-right direction. The reflecting portion 230 includes, for example, a mirror, and is attached to the inner surface of the side surface portions 213 and 215 of the housing portion 210 so as to extend behind the light emitting portion 220 and in the left-right direction.

The image pickup unit 240 is mounted on the bottom surface portion 211 of the housing portion 210 at a position behind the reflection portion 230. The image pickup unit 240 includes an image pickup element in which a plurality of pixels are linearly arranged so as to extend in the left-right direction, and one or a plurality of condenser lenses. In this example, a color CCD (charge-coupled device) line sensor is used as the image pickup device. A color CMOS (complementary metal oxide semiconductor) line sensor may be used as the image pickup device.

The reflecting unit 230 has a reflecting surface facing diagonally downward and rearward, and is arranged in the field of view of the imaging unit 240. The reflection surface of the reflection unit 230 forms an image pickup region of the image pickup unit 240 below the light projecting unit 220 and the reflection unit 230. The imaging region of the imaging unit 240 extends linearly in the left-right direction.

As will be described later, the substrate W to be inspected is carried into the housing portion 210 from the opening 216, and the carried-in substrate W passes below the light projecting portion 220. The light projecting unit 220 emits light having a cross-sectional line extending in the left-right direction longer than the diameter of the substrate W diagonally downward and rearward. As shown in FIG. 2, a part of the light emitted diagonally downward and backward from the light projecting unit 220 is reflected diagonally upward and backward by one surface (upper surface) of the substrate W in the image pickup region of the image pickup unit 240, and the reflection unit 230. Is reflected backward by the image pickup unit 240 and receives light by the image pickup unit 240.

The substrate holding device 250 is, for example, a spin chuck and includes a driving device 251 and a rotation holding unit 252. The drive device 251 is, for example, an electric motor and has a rotating shaft 251a. The drive device 251 is provided with an encoder (not shown). The rotation holding portion 252 is attached to the tip of the rotation shaft 251a of the drive device 251 and is rotationally driven around the vertical shaft while holding the substrate W to be inspected.

As shown in FIG. 3, the moving portion 260 includes a plurality of (two in this example) guide members 261 and a moving holding portion 262. The plurality of guide members 261 are attached to the bottom surface portion 211 of the housing portion 210 so as to be aligned in the left-right direction and extend in the front-rear direction. The movement holding unit 262 is configured to be movable in the front-rear direction along the plurality of guide members 261 while holding the substrate holding device 250. The moving holding unit 262 moves in the front-rear direction while the board holding device 250 holds the board W, so that the board W passes below the light projecting unit 220.

The notch detection unit 270 is a reflection type photoelectric sensor including, for example, a light emitting element and a light receiving element, and is attached to the front upper portion of the inner surface of the side surface portion 215 of the housing portion 210. When the peripheral edge of the substrate W to be inspected is located below the notch detection unit 270, the notch detection unit 270 emits light downward and receives the reflected light from the substrate W. Here, when a notch is formed in the portion of the substrate W located below the notch detection unit 270, the amount of light received by the notch detection unit 270 is reduced. The notch detection unit 270 detects the presence or absence of a notch in the substrate W based on the amount of received light received from the substrate W rotated by the substrate holding device 250. A transmissive photoelectric sensor may be used as the notch detection unit 270.

As shown in FIG. 1, a control device 400 and a display unit 280 are provided outside the housing unit 210. The control device 400 controls a light projecting unit 220, an image pickup unit 240, a substrate holding device 250, a moving unit 260, a notch detection unit 270, and a display unit 280. The display unit 280 displays an automatic determination result of the presence or absence of a defect in the substrate W to be inspected and an image of the substrate W for visual inspection. Details of the control device 400 will be described later. Note that in FIGS. 2 and 3, the control device 400 and the display unit 280 are not shown.

In the above-mentioned substrate inspection apparatus 200, the entire surface of the substrate W to be inspected is imaged at the time of inspection. The operation at the time of this imaging will be described. In the initial state, as shown in FIG. 1, the board holding device 250 is located at the front portion in the housing portion 210. In this state, the substrate W to be imaged is carried into the housing portion 210 through the opening 216 and held by the substrate holding device 250.

Next, while the substrate W is rotated once by the substrate holding device 250, light is emitted to the peripheral edge portion of the substrate W by the notch detection unit 270, and the reflected light is received by the notch detection unit 270. As a result, the notch of the substrate W is detected, and the orientation of the substrate W is determined. After that, the substrate W is rotated by the substrate holding device 250 so that the substrate W faces a specific direction.

Next, the substrate W is moved backward by the moving unit 260 in a state where the light is emitted from the light projecting unit 220. At this time, as the substrate W passes below the light projecting portion 220, the entire surface of the substrate W is irradiated with light having a linear cross section extending in the left-right direction. As described above, the light reflected from the substrate W in the image pickup region of the image pickup unit 240 is further reflected by the reflection unit 230 and guided to the image pickup unit 240. The image sensor of the image pickup unit 240 receives light reflected from one surface of the substrate W at a predetermined sampling cycle, thereby sequentially imaging a plurality of portions on one surface of the substrate W in the front-rear direction. Each pixel constituting the image sensor outputs pixel data indicating a value corresponding to the amount of received light. As a result, image data representing the entire image on one surface of the substrate W is generated based on the plurality of pixel data output from the image pickup unit 240.

After that, the emission of light by the light projecting unit 220 is stopped, and the substrate W is moved from the position behind the reflecting unit 230 to the position in front of the light projecting unit 220 by the moving unit 260. Finally, the substrate W is carried out of the housing portion 210 through the opening 216.

(2) Image based on image data generated by imaging FIG. 4A is a diagram showing an example of an actual image representing one surface of the substrate W, and FIG. 4B is a diagram showing the substrate inspection device 200 of FIG. It is a figure which shows an example of the conventional image based on the image data generated by imaging the substrate W of FIG. 4A using. It is assumed that there are no defects on one surface of the substrate W shown in FIG. 4 (a). In FIGS. 4A and 4B, the notch is not shown.

In the image of FIG. 4A, the average color of the plurality of portions on one surface of the substrate W is uniform as a whole. Thereby, it is possible to clearly visually recognize the periodic pattern of the film formed on one surface of the substrate W. On the other hand, on the image of FIG. 4B, both ends of the substrate W and their vicinity are represented by a different color from the central portion including the center of the substrate W, as shown by hatching. The reason for this will be explained.

Each pixel provided in the image pickup element of the image pickup unit 240 is composed of an R pixel that receives red light, a G pixel that receives green light, and a B pixel that receives blue light. The refractive index of light in the film formed on the substrate W differs depending on the wavelength. Further, the transmittance and reflectance of light in the film on the substrate W may also differ depending on the wavelength. Therefore, if the incident angle of the light reflected by one surface of the substrate W and incident on the image pickup surface of the image pickup device of the image pickup unit 240 is different, the ratio of the light receiving amount of the R pixel, the G pixel, and the B pixel is different. As a result, the ratio of the R pixel value, the G pixel value, and the B pixel value of the image data differs depending on the incident angle of the light on the image pickup surface of the image pickup unit 240. The color of an image is determined by the combination of hue, lightness and saturation. As described above, if the ratio of the R pixel value, the G pixel value, and the B pixel value is different, the hue will be different. Even if the above ratios are the same, the brightness will be different if the R pixel value, the G pixel value, and the B pixel value are different. As a result, when the incident angle of the light on the image pickup surface of the image pickup unit 240 is different, the color of the image represented by the image data is different.

FIG. 5 is a schematic plan view showing a state in which the entire surface of the substrate W is imaged by the image pickup unit 240 in the substrate inspection device 200 of FIG. 1. As shown in FIG. 5, a first virtual surface VS1 extending in the front-rear direction through the center WC of the substrate W held by the substrate holding device 250 and perpendicular to one surface of the substrate W is defined. Further, a second virtual surface VS2 orthogonal to the front-rear direction is defined behind the light projecting unit 220 and the reflecting unit 230. Further, a virtual point VP is defined on the intersection of the first virtual surface VS1 and the second virtual surface VS2.

The image pickup surface 242 of the image pickup unit 240 is arranged on the second virtual surface VS2, and the center of the image pickup surface 242 is arranged on the virtual point VP. In this case, the incident angles γ of the light incident on the image pickup surface 242 of the image pickup unit 240 from one end WE1 and the other end WE2 of the substrate W in the left-right direction can be made equal. However, if the distance from the reflecting unit 230 to the image pickup surface 242 in the front-rear direction is short, the incident angle γ of the light incident on the image pickup surface 242 of the image pickup unit 240 from one end WE1 and the other end WE2 of the substrate W and the substrate W The difference from the incident angle 0 of the light incident on the imaging surface 242 of the imaging unit 240 from the center WC becomes large. In the substrate inspection device 200, it is difficult to increase the distance in the front-rear direction from the reflection unit 230 to the image pickup surface 242 in order to suppress the increase in size of the housing portion 210 in the front-rear direction. Therefore, the colors of the images of the one end WE1 and the other end WE2 of the substrate W are significantly different from the colors of the image in the center of the substrate W including the center WC of the substrate W.

As described above, when both ends of the substrate W and their vicinity are represented by a different color from the central portion of the substrate W on the image based on the image data, the visual inspection of the substrate W using the image shows on the image. It is difficult to judge the defect with. Therefore, in the substrate inspection apparatus 200, the image data generated by the imaging is corrected so that the average colors of the plurality of portions on one surface of the substrate W are equal to or substantially equal to each other, and the display for visual inspection is performed. Image data is generated.

(3) Method for generating image display data according to the first embodiment FIGS. 6 to 18 are diagrams for explaining a method for generating image display data according to the first embodiment. In the following description, the image data generated by imaging the substrate W in the substrate inspection device 200 is referred to as actual image data. The actual image data includes a value indicating the light receiving amount of the R pixel, a value indicating the light receiving amount of the G pixel, and a value indicating the light receiving amount of the B pixel as information for each pixel. Further, in the image of the substrate W based on the actual image data, one radial direction of the substrate W is referred to as an X direction, and the other radial direction of the substrate W orthogonal to the X direction is referred to as a Y direction. The X direction corresponds to the left-right direction of the board inspection device 200 of FIG. 1, and the Y direction corresponds to the front-back direction of the board inspection device 200 of FIG.

First, as shown in FIG. 6, in an image based on actual image data, a plurality of unit regions UA arranged in the X direction are set in the central portion of the substrate W in the Y direction. Each unit region UA includes a plurality of pixel PIs, and has a width of one pixel in the X direction and a length of 100 pixels in the Y direction. As a result, each unit area UA corresponds to each pixel position in the X direction. The length of the unit region UA in the Y direction may be larger than the width in the X direction and smaller than the diameter of the substrate W.

Subsequently, the average value of the R pixels, the average value of the G pixels, and the average value of the B pixels of the plurality of pixel PIs in each set unit area UA are calculated. The average value of R pixels, the average value of G pixels, and the average value of B pixels calculated for each unit area UA are the average pixel value of R pixels and the average pixel value of G pixels corresponding to the unit area UA, respectively. And the average pixel value of B pixel. The color obtained based on the ratio of the average pixel values of RGB and the magnitude of those values becomes the average color representing the colors of the plurality of pixel PIs in each unit region UA. The median value of the R pixel, the median value of the G pixel, and the median value of the B pixel of the plurality of pixel PIs in each unit area UA are the average pixel value of the R pixel, the average pixel value of the G pixel, and B, respectively. It may be the average pixel value of a pixel.

FIG. 7 is a graph showing an example of the average pixel value of RGB calculated from the actual image data. In FIG. 7, the vertical axis represents the pixel value, and the horizontal axis represents the pixel position in the X direction on the image of the substrate W. Further, one end (left end) of the horizontal axis represents the pixel position of one end (left end) in the X direction on the image of the substrate W, and the other end (right end) of the horizontal axis is on the image of the substrate W. Represents the pixel position of the other end (right end) in the X direction. Further, in FIG. 7, a line connecting the average pixel values of R pixels from one end to the other end in the X direction is represented by a alternate long and short dash line (hereinafter referred to as R average pixel value line RR) in the X direction. The line connecting the average pixel values of the G pixels from one end to the other end is represented by a dotted line (hereinafter referred to as G average pixel value line RG), and the B pixel is formed from one end to the other end in the X direction. The line connecting the average pixel values is represented by a solid line (hereinafter, referred to as B average pixel value line RB). According to the graph of FIG. 7, it can be seen that the ratio of the average pixel values of RGB in the central portion of the substrate W and the ratio of the average pixel values of RGB in both ends of the substrate W and its vicinity are different.

Next, in order to obtain the period in the X direction of the pattern of the film formed on one surface of the substrate W, each of the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB is obtained. The average pixel value data of k (k is a natural number of 2 or more) continuously arranged in the X direction in a part of the region including the center of the substrate W (hereinafter referred to as the central region) and the central region thereof. The correlation coefficient with the data of k average pixel values arranged continuously in the X direction at positions shifted by one pixel in the X direction is sequentially calculated. In this embodiment, k is set to about 200.

FIG. 8 is a graph showing an example of the correlation coefficient calculated from the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB of FIG. 7. In FIG. 8, the vertical axis represents the correlation coefficient, and the horizontal axis represents the pixel position in the X direction on the image of the substrate W. Further, in FIG. 8, the correlation coefficient corresponding to the R average pixel value line RR is represented by a one-dot chain line, the correlation coefficient corresponding to the G average pixel value line RG is represented by a dotted line, and the B average pixel value is represented by a dotted line. The correlation coefficient corresponding to the line RB is represented by a solid line.

Next, a phase showing remarkable periodicity in the direction of the pixel position from three types of correlation coefficients calculated for each of the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB. Select the number of relationships. In the example of FIG. 8, the correlation coefficient corresponding to the B average pixel value line RB is selected. In the waveform of the selected correlation coefficient, as shown by the white arrow in FIG. 8, the peak that exceeds the predetermined threshold value th0 and corresponds to the central region of the substrate W is identified as the central peak.

Next, with respect to the central peak, as shown by the thick solid arrow in FIG. 8, one of the two peaks exceeding the predetermined threshold value th0 and adjacent to the central peak is extracted. In addition, the distance (number of pixels) in the X direction between the central peak and the extracted peak is calculated. The distance calculated in this way is defined as the period in the X direction of the pattern of the film formed on one surface of the substrate W (hereinafter referred to as the pattern period).

Next, the average pixel value of RGB shown in FIG. 7 is smoothed by the moving median method. Specifically, for each unit area UA, the median movement is calculated with a width of one pattern cycle on one side (left) of the unit area UA and one pattern cycle on the other side (right side) of the unit area UA. do. In this way, the median movement in the X direction is calculated by adding "1" to the value twice the pattern period. As a result, the variation in the average pixel value due to the pattern of the film is appropriately reduced.

In FIG. 9, the first R reference curve R1 obtained by smoothing the R average pixel value line RR of FIG. 7 is represented by a alternate long and short dash line, and the G average pixel value line RG of FIG. 7 is smoothed. The first G reference curve G1 obtained by being smoothed is represented by a dotted line, and the first B reference curve B1 obtained by smoothing the B average pixel value line RB in FIG. 7 is represented by a solid line. Ru.

On one surface of the substrate W, a film may not be formed at or near the outer peripheral end portion. Alternatively, even when a film is formed at the outer peripheral edge and its vicinity, the pattern of the film formed at the outer peripheral edge and its vicinity has a periodicity different from that of the film formed at the center of the substrate W. Have. Therefore, as shown by the thick two-dot chain line in FIG. 9, the first R reference curve R1, the first G reference curve G1 and the first B reference curve B1 are smoothed in the X direction. It is greatly disturbed at both ends of.

Therefore, the average pixel value of one end to be calculated when it is assumed that one end of the substrate W on the image is not disturbed is estimated. Specifically, the ideal average pixel value for one pattern cycle including one end on the image of the substrate W is estimated from the average pixel value for another one pattern cycle adjacent to that portion.

FIG. 10A shows a first R reference curve R1, a first G reference curve G1 and a first B reference curve B1 for about two pattern cycles from one end in the X direction. Here, a pixel position separated by two pattern cycles from one end is defined as p ₀ , and a pixel position separated from the pixel position p ₀ toward one end by an arbitrary distance smaller than one pattern cycle is defined as _pn . Further, a pixel position separated by one pattern cycle from one end is defined as _p'0 , and a pixel position separated from the pixel position _p'0 toward one end by an arbitrary distance smaller than one pattern cycle is defined as _p'n . The distance between the pixel positions _{p'0 and p'n is equal to the distance between the pixel positions p 0 and p n} .

In this case, for the first R reference curve R1, the pixel value at the pixel position p ₀ is set to R ₀ , the pixel value at the pixel position _pn is set to R _n , and the pixel value at the pixel position p ' ₀ is set to R ' ₀ . In this case, the ideal pixel value _R'n of the pixel position _p'n can be estimated based on the following equation (1).

_{R'n = (R n -} R ₀ ) + _R'0 ... (1)
Further, regarding the first G reference curve G1, the pixel value of the pixel position p ₀ is set to G ₀ , the pixel value of the pixel position _pn is set to G _n , and the pixel value of the pixel position p ' ₀ is set to G ' ₀ . In addition, the ideal pixel value _G'n of the pixel position _p'n can be estimated based on the following equation (2).

_{G'n = (G n -} G ₀ ) + _G'0 ... (2)
Further, regarding the first B reference curve B1, the pixel value of the pixel position p ₀ is set to B ₀ , the pixel value of the pixel position _pn is set to B _n , and the pixel value of the pixel position p ' ₀ is set to B ' ₀ . In addition, the ideal pixel value _B'n of the pixel position _p'n can be estimated based on the following equation (3).

_{B'n = (B n -} B ₀ ) + _B'0 ... (3)
Using the above equations (1), (2), and (3), the average pixel value for one pattern cycle including one end in the X direction is estimated, and the estimation result is for one pattern cycle located at one end. It is determined as the average pixel value of RGB in a plurality of unit areas UA. As a result, as shown in FIG. 10B, the disorder at one end of the first R reference curve R1, the first G reference curve G1 and the first B reference curve B1 is removed.

Further, the ideal average pixel value for one pattern cycle including the other end on the image of the substrate W is also estimated from the average pixel value for another one pattern cycle adjacent to that portion. FIG. 10C shows a first R reference curve R1, a first G reference curve G1 and a first B reference curve B1 for about two pattern cycles from the other end in the X direction. In this example, a pixel position separated by two pattern cycles from the other end is defined as p ₀ , and a pixel position separated from the pixel position p ₀ toward the other end by an arbitrary distance smaller than one pattern cycle is defined as _pn . Further, the pixel position separated by one pattern cycle from the other end is _p'0 , and the pixel position separated from the pixel position _p'0 toward the other end by an arbitrary distance smaller than one pattern cycle is defined as _p'n . .. The distance between the pixel positions _{p'0 and p'n is equal to the distance between the pixel positions p 0 and p n} .

In this case, the average pixel value for one pattern cycle including the other end in the X direction is estimated using the above equations (1), (2), and (3), and the estimation result is located at the other end. It is determined as the average pixel value of RGB of a plurality of unit areas UA for one pattern cycle. As a result, as shown in FIG. 10D, the turbulence at the other end of the first R reference curve R1, the first G reference curve G1 and the first B reference curve B1 is removed. FIG. 11 shows the entire first R reference curve R1, the first G reference curve G1 and the first B reference curve B1 with both ends modified based on the estimation result.

Next, the average pixel value represented by the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB in FIG. 7 is also disturbed at one end on the image of the substrate W. Estimate the average pixel value at both ends to be calculated when it is assumed that there is no such value, and correct the average pixel value at both ends based on the estimation result. Here, the following processing is performed in order to more accurately determine the range of pixel positions where the pattern of the film is disturbed (hereinafter referred to as an unnecessary range).

First, for each unit region UA, the average pixel value before smoothing represented by the R average pixel value line RR in FIG. 7 and the average after smoothing represented by the first R reference curve R1 in FIG. 11 The absolute value of the difference from the target pixel value is calculated as the deviation evaluation value of the R pixel. Further, for each unit region UA, the average pixel value before smoothing represented by the G average pixel value line RG in FIG. 7 and the average after smoothing represented by the first G reference curve G1 in FIG. 11 The absolute value of the difference from the target pixel value is calculated as the deviation evaluation value of the G pixel. Further, for each unit region UA, the average pixel value before smoothing represented by the B average pixel value line RB in FIG. 7 and the average after smoothing represented by the first B reference curve B1 in FIG. 11 The absolute value of the difference from the target pixel value is calculated as the deviation evaluation value of the B pixel. FIG. 12 shows the deviation evaluation value of the B pixel. The deviation evaluation values of the R pixel and the G pixel are calculated so as to substantially overlap with the deviation evaluation value of the B pixel, for example.

Next, with respect to the average pixel value before smoothing represented by the R average pixel value line RR in FIG. 7, the average pixel value of the two left and right unit region UAs adjacent to the unit region UA for each unit region UA. The absolute difference value of is calculated as the difference evaluation value of R pixels. Further, regarding the average pixel value before smoothing represented by the G average pixel value line RG in FIG. 7, the average pixel value of the two left and right unit region UAs adjacent to the unit region UA for each unit region UA. The difference absolute value is calculated as the difference evaluation value of G pixels. Further, regarding the average pixel value before smoothing represented by the B average pixel value line RB in FIG. 7, the average pixel value of the two left and right unit region UAs adjacent to the unit region UA for each unit region UA. The difference absolute value is calculated as the difference evaluation value of the B pixel. FIG. 13 shows the difference evaluation value of the B pixel. The difference evaluation value of the R pixel and the G pixel is calculated so as to substantially overlap with the difference evaluation value of the B pixel, for example.

Then, based on the RGB deviation evaluation value and the difference evaluation value, an unnecessary range including one end in the X direction on the image of the substrate W is determined. FIG. 14A shows RGB deviation evaluation values for about two pattern cycles from one end in the X direction. Further, FIG. 14B shows RGB difference evaluation values for about two pattern cycles from one end in the X direction. In FIGS. 14 (a) and 14 (b) and FIGS. 14 (c) and 14 (d) described later, the evaluation value corresponding to the R pixel is indicated by the alternate long and short dash line, and the evaluation value corresponding to the G pixel is indicated by the dotted line. The evaluation value corresponding to the B pixel is shown by a solid line.

Here, it is assumed that the threshold value th1 is predetermined for the deviation evaluation value as shown in FIG. 14A. Further, as for the difference evaluation value, it is assumed that the threshold value th2 is set in advance as shown in FIG. 14 (b). In this state, all the deviation evaluation values of RGB are equal to or less than the threshold value th1 and all the difference evaluation values of RGB are in each unit region UA (for each pixel) from one end to the other end in the X direction. Determines whether or not is equal to or less than the threshold value th2. Therefore, the range from the pixel position to one end of the unit region UA when all the deviation evaluation values of RGB are initially equal to or less than the threshold value th1 and all the differential evaluation values of RGB are equal to or less than the threshold value th2. Is determined to be the unnecessary range UN1.

Further, based on the RGB deviation evaluation value and the difference evaluation value, an unnecessary range including the other end portion in the X direction on the image of the substrate W is determined. FIG. 14C shows RGB deviation evaluation values for about two pattern cycles from the other end in the X direction. Further, FIG. 14 (d) shows RGB difference evaluation values for about two pattern cycles from the other end in the X direction.

Regarding the other end, basically the same as the above example of one end, all the deviation evaluation values of RGB are calculated for each unit region UA (for each pixel) from the other end in the X direction toward one end. It is determined whether or not the threshold value is th1 or less and all the RGB difference evaluation values are the threshold value th2 or less. Therefore, first, when all the deviation evaluation values of RGB are equal to or less than the threshold value th1 and all the difference evaluation values of RGB are equal to or less than the threshold value th2, from the pixel position of the unit region UA to the other end. The range is determined to be the unnecessary range UN2.

After that, as shown by hatching in FIG. 15 (a), FIGS. 14 (a) and 14 (b) of the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB in FIG. 7 are shown. ) Is corrected by the average pixel value represented by the portion located in UN1. Specifically, as in the example of FIG. 10A, the ideal RGB to be calculated in the unnecessary range UN1 based on the average pixel value of RGB for one pattern cycle adjacent to the unnecessary range UN1. Estimate the average pixel value. Further, the estimation result is determined as the average pixel value of RGB of the plurality of unit regions UA located in the unnecessary range UN1. As a result, as shown in FIG. 15B, the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB are corrected.

Further, as shown by hatching in FIG. 15 (c), of the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB in FIG. 7, FIGS. 14 (c) and 14 (d). ) Is corrected by the average pixel value represented by the portion located in UN2. Specifically, as in the example of FIG. 10C, the ideal RGB to be calculated in the unnecessary range UN2 based on the average pixel value of RGB for one pattern cycle adjacent to the unnecessary range UN2. Estimate the average pixel value. Further, the estimation result is determined as the average pixel value of RGB of the plurality of unit regions UA located in the unnecessary range UN2. As a result, as shown in FIG. 15D, the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB are corrected. FIG. 16 shows the entire R average pixel value line RR, G average pixel value line RG, and B average pixel value line RB whose both ends are modified based on the estimation result.

Next, the figure shows the average pixel value of RGB represented by each of the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB of FIG. The smoothing process by the moving median method is performed in the same manner as in the example of 9. As a result, the variation in the average pixel value due to the pattern of the film is appropriately reduced. Here, the smoothing process by the moving average method may be performed instead of the moving median method.

In FIG. 17, the second R reference curve R2 obtained by smoothing the R average pixel value line RR of FIG. 16 is represented by a alternate long and short dash line, and the G average pixel value line RG of FIG. 16 is smoothed. The second G reference curve G2 obtained by being smoothed is represented by a dotted line, and the second B reference curve B2 obtained by smoothing the B average pixel value line RB in FIG. 16 is represented by a solid line. Ru.

Here, as shown by a thick solid line in FIG. 18, a band-shaped region BA extending in the Y direction on the image of the substrate W from each unit region UA corresponding to each pixel position in the X direction is defined.

Subsequently, a correction process for adjusting all the average pixel values represented by the second R reference curve R2 in the X direction to a predetermined reference value is applied to the values of all R pixels in the actual image data. In this example, among the average pixel values of the R pixels represented by the second R reference curve R2, the average pixel value of the unit region UA located at the center of the substrate W is set as the reference value Rv. Further, the difference between the reference value Rv and the average pixel value represented by the second R reference curve R2 is calculated for each unit region UA. Further, the difference corresponding to the unit region UA is added to the value of the R pixel of each pixel in the band-shaped region BA extending from each unit region UA in the Y direction. Regarding this correction process, the value of the R pixel of any pixel on the image of the substrate W is expressed as R _{(x, y)} using the plane coordinate system defined by the axes in the X direction and the Y direction, and the value of the pixel is expressed as R (x, y). When the average pixel value on the second R reference curve R2 corresponding to the pixel position in the X direction is expressed as R _x , the corrected R pixel value R' _{(x, y)} is expressed by the following equation (4). Can be represented by.

R' _{(x, y)} = R _{(x, y)} -R _x + Rv ... (4)
Further, the same correction processing as the above-mentioned correction processing for the R pixel value is performed for all the G pixel values of the actual image data. In this example, among the average pixel values of the G pixels represented by the second G reference curve G2, the average pixel value of the unit region UA located at the center of the substrate W is set as the reference value Gv. Regarding this correction process, the value of the G pixel of any pixel on the image of the substrate W is expressed as G _{(x, y)} , and is on the second G reference curve G2 corresponding to the pixel position of the pixel in the X direction. When the average pixel value is expressed as G _x , the corrected G pixel value G' _{(x, y)} can be expressed by the following equation (5).

G' _{(x, y)} = G _{(x, y)} -G _x + Gv ... (5)
Further, the values of all the B pixels of the actual image data are also corrected in the same manner as the correction processing for the values of the R pixels described above. In this example, among the average pixel values of the B pixels represented by the second B reference curve B2, the average pixel value of the unit region UA located at the center of the substrate W is set as the reference value Bv. Regarding this correction process, the value of the B pixel of any pixel on the image of the substrate W is expressed as B _{(x, y)} , and is on the second B reference curve B2 corresponding to the pixel position of the pixel in the X direction. When the average pixel value is expressed as B _x , the corrected B pixel value B' _{(x, y)} can be expressed by the following equation (6).

B' _{(x, y)} = B _{(x, y)} -B _x + Bv ... (6)
Display image data is generated by correcting all the pixels of the actual image data using the above equations (4), (5), and (6). As a result, in the image represented by the display image data, the average colors of the plurality of strip-shaped regions BA on one surface of the substrate W in the left-right direction are equal to or substantially equal to each other.

(4) Control system of the substrate inspection device 200 FIG. 19 is a block diagram showing a control system of the substrate inspection device 200 according to the first embodiment. The control device 400 is composed of a CPU (central processing unit), RAM (random access memory), and ROM (read-only memory), and as shown in FIG. 19, the control unit 401, the image generation unit 402, and the determination unit 403, It has an image determination storage unit 404 and a display image data generation unit 405. In the control device 400, each of the above functional units is realized by the CPU executing a computer program stored in a ROM or another storage medium. It should be noted that some or all of the functional components of the control device 400 may be realized by hardware such as an electronic circuit.

Regarding the board holding device 250 and the notch detection unit 270, the control unit 401 acquires an output signal from the encoder of the driving device 251 (FIG. 2) of the board holding device 250 and acquires the rotation angle of the driving device 251 (rotation angle of the board W). Is detected, and the notch detection result by the notch detection unit 270 is acquired. The control unit 401 determines the orientation of the substrate W based on the rotation angle of the drive device 251 when the notch is detected, and controls the operation of the substrate holding device 250 based on the determination result.

With respect to the light projecting unit 220, the moving unit 260, and the image pickup unit 240, the control unit 401 performs the light projecting unit 220, the moving unit 260, and the image pickup unit so that the entire one surface of the substrate W held by the substrate holding device 250 is imaged. The unit 240 is controlled.

The image generation unit 402 generates image data representing the entire image on one surface of the substrate W based on a plurality of pixel data output from the image pickup unit 240. The determination unit 403 automatically determines the presence or absence of defects in the surface state on one surface of the substrate W based on the image data generated by the image generation unit 402. The image determination storage unit 404 stores the automatic determination result of the presence / absence of a defect by the determination unit 403.

The display image data generation unit 405 uses the image data generated by the image generation unit 402 as the above-mentioned actual image data, and generates display image data representing an image to be displayed on the display unit 280 based on the actual image data. do.

More specifically, the display image data generation unit 405 includes a smoothing unit 411 and a correction unit 412. The smoothing unit 411 calculates the average pixel value of the plurality of pixels constituting the unit region UA for each of the plurality of unit regions UA (FIG. 6) arranged in the X direction on the image based on the actual image data. Further, the smoothing unit 411 performs a smoothing process on the average pixel value of the plurality of unit regions UA.

The smoothing unit 411 sets the ideal average pixel value to be calculated for each of the plurality of unit region UAs located at both ends of the substrate W as the average calculated for the other plurality of unit region UAs. It may be estimated by the pixel value, and the average pixel value of the plurality of unit regions UA located at both ends of the substrate W may be determined based on the estimation result.

The correction unit 412 calculates the difference between the smoothed average pixel value of the unit region UA located at the center of the substrate W and the smoothed average pixel value of each unit region UA. Further, the correction unit 412 corrects the actual image data so as to add the difference corresponding to the unit region UA to the value of each pixel in the band-shaped region BA (FIG. 18) extending from each unit region UA in the Y direction. .. As a result, display image data is generated.

The image determination storage unit 404 further stores the display image data generated by the display image data generation unit 405. An image based on the automatic determination result stored in the image determination storage unit 404 and the display image data is displayed on the display unit 280.

(5) Defect determination processing As described above, in the substrate inspection device 200, the control device 400 of FIG. 1 automatically determines the presence or absence of defects in the surface state on one surface of the substrate W, and an image showing one surface of the substrate W. Is displayed to the user for visual inspection. These series of processes are called defect determination processes. FIG. 20 is a flowchart of the defect determination process according to the first embodiment. Here, the substrate W to be inspected is referred to as an inspection substrate W.

Before the start of the defect determination process, an inspection is performed with high accuracy in advance, and a substrate determined to have no defects by the inspection is prepared as a sample substrate. The control unit 401 and the image generation unit 402 of FIG. 19 first image a sample substrate without defects to generate image data representing one surface of the sample substrate (step S11).

Next, the control unit 401 and the image generation unit 402 generate image data representing one surface of the inspection board W by imaging the inspection board W (step S12). Further, the determination unit 403 in FIG. 19 automatically determines whether or not there is a defect in the surface state of the inspection substrate W based on the image data of the sample substrate and the image data of the inspection substrate W (step S13). The image determination storage unit 404 of FIG. 19 stores the automatic determination result (step S14).

Subsequently, the smoothing unit 411 of FIG. 19 uses the image data generated in the process of step S12 as the actual image data, and has a plurality of unit regions UA (FIG. 6) arranged in the X direction on the image based on the actual image data. For each, the average pixel value of the plurality of pixels constituting the unit region UA is calculated (step S15). Further, the smoothing unit 411 performs a smoothing process on the average pixel value of the plurality of unit regions UA (step S16).

Next, the correction unit 412 in FIG. 19 calculates the difference between the smoothed average pixel value of the unit region UA located at the center of the inspection substrate W and the smoothed average pixel value of each unit region UA. (Step S17). Further, the correction unit 412 adds the difference corresponding to the unit area UA to the value of each pixel in the band-shaped area BA (FIG. 18) extending from each unit area UA (step S18). As a result, display image data is generated. The image determination storage unit 404 stores display image data (step S19).

After that, the image determination storage unit 404 causes the display unit 280 of FIG. 19 to display an image based on the automatic determination result and the display image data (step S20). In this state, the user can visually inspect one side of the inspection board W by visually recognizing the image of the inspection board W displayed on the display unit 280.

In the above defect determination process, the processes of steps S13 and S14 may be performed after the processes of steps S15 to S19, or may be performed in parallel with the processes of steps S15 to S19.

The inspection board W determined to be defective in the process of step S13, or the inspection board W determined to be defective by the user's visual inspection during the process of step S20, is subject to detailed inspection or regeneration processing. ..

(6) Effect of First Embodiment (a) In the substrate inspection apparatus 200 according to the first embodiment, an actual image showing an image of one surface of the substrate W by being imaged on one surface of the substrate W. Data is generated. For each of the plurality of unit area UAs set on the image based on the actual image data, the average value of the plurality of pixels constituting the unit area UA is calculated as the average pixel value, and the average of the plurality of unit area UAs is calculated. Smoothing processing is performed on the target pixel value.

The difference between the average pixel value after smoothing of the unit region UA located at the center of the substrate W and the average pixel value after smoothing of each unit region UA is calculated. The difference corresponding to the unit area UA is added to the value of each pixel in the band-shaped area BA extending parallel to each unit area UA in the Y direction. As a result, in the image based on the display image data, the average value of the plurality of pixels in each band-shaped region BA is equal to or substantially equal to the average pixel value of the unit region UA located at the center of the substrate W. Will be equal. Therefore, in the image based on the display image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate W in the X direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate W in a visual inspection using an image based on display image data.

(B) In the present embodiment, the average pixel value calculated for each unit region UA is smoothed by the moving median method. Further, the actual image data is corrected based on the average pixel value after smoothing, and the display image data is generated. In this case, the smoothing process by the moving median method appropriately reduces the variation in the local average pixel value.

(C) In the present embodiment, the average pixel value calculation process, the average pixel value smoothing process, and the actual image data are obtained for each type of R pixel, G pixel, and B pixel constituting one pixel. Correction processing is performed. As a result, it is suppressed that the average hues of the plurality of portions of the surface of the substrate W in the X direction are visually recognized as different on the image based on the display image data.

(D) In the present embodiment, the ideal average pixel values of the plurality of unit regions UA located at both ends of the substrate W on the image based on the actual image data are located at the portions adjacent to the both ends. It is estimated based on the average pixel value of a plurality of unit regions UA. The estimation result is determined as a plurality of average pixel values corresponding to each of the plurality of unit regions UA located at each end. As a result, display image data is generated based on the average pixel value determined for each unit area UA. As a result, in the image based on the display image data, it is suppressed that the average color of both ends of the substrate W is visually recognized as being significantly different from the average color of the portions adjacent to both ends.

(E) In the substrate inspection device 200 according to the present embodiment, the entire image of one surface of the substrate W is represented by the relative movement of the light projecting unit 220, the reflection unit 230, the image pickup unit 240, and the substrate holding device 250. Image data is generated. Therefore, the increase in size of the image pickup unit 240 is suppressed.

[2] Second Embodiment The substrate inspection apparatus according to the second embodiment has basically the same configuration as the substrate inspection apparatus 200 of FIG. 1 according to the first embodiment. In the substrate inspection device 200 according to the present embodiment, the method for generating image display data is different from the method for generating image display data according to the first embodiment.

(1) Method for generating image display data according to the second embodiment FIGS. 21 to 24 are diagrams for explaining a method for generating image display data according to the second embodiment. First, it is represented by the R average pixel value line RR, the G average pixel value line RG, and the B average pixel value line RB of FIG. 16 in the same procedure as the method for generating image display data according to the first embodiment. The average pixel value of RGB to be obtained is calculated.

Further, by performing smoothing processing on the calculated average pixel value of RGB by the moving median method, the second R reference curve R2, the second G reference curve G2, and the second B reference curve B2 in FIG. 17 are performed. The average pixel value of RGB after smoothing represented by is calculated.

Next, for each unit region UA, the average pixel value before smoothing represented by the R average pixel value line RR in FIG. 16 and the average pixel value after smoothing represented by the second R reference curve R2 in FIG. 17 The difference from the average pixel value is calculated as the deviation evaluation value of the R pixel. Further, for each unit region UA, the average pixel value before smoothing represented by the G average pixel value line RG in FIG. 16 and the average after smoothing represented by the second G reference curve G2 in FIG. 17 The difference from the target pixel value is calculated as the deviation evaluation value of the G pixel. Further, for each unit region UA, the average pixel value before smoothing represented by the B average pixel value line RB in FIG. 16 and the average after smoothing represented by the second B reference curve B2 in FIG. 17 The difference from the target pixel value is calculated as the deviation evaluation value of the B pixel. FIG. 21 shows the calculated deviation evaluation values of the R pixel, the G pixel, and the B pixel as a alternate long and short dash line, a dotted line, and a solid line, respectively.

Next, smoothing processing is performed for each of the RGB deviation evaluation values shown in FIG. 21 by the moving maximum method. Specifically, for each unit area UA, the maximum movement value is calculated with a width of one pattern cycle on one side (left) of the unit area UA and one pattern cycle on the other side (right side) of the unit area UA. do. In this way, the maximum value of movement in the X direction is calculated by adding "1" to the value twice the pattern period. As a result, the variation in the maximum movement value due to the pattern of the film is appropriately reduced. The maximum movement value calculated by this smoothing process is called the maximum deviation value.

Further, each of the RGB deviation evaluation values shown in FIG. 21 is smoothed by the movement minimum method. Specifically, for each unit area UA, the minimum movement value is calculated with a width of one pattern cycle on one side (left) of the unit area UA and one pattern cycle on the other side (right side) of the unit area UA. do. In this way, the minimum movement value in the X direction is calculated by adding "1" to the value twice the pattern period. As a result, the variation in the minimum movement value due to the pattern of the film is appropriately reduced. The minimum movement value calculated by this smoothing process is called the minimum deviation value.

In FIG. 22, the maximum deviation value and the minimum deviation value corresponding to the R pixel are represented by the alternate long and short dash lines RDmax and RDmin, respectively. Further, the maximum deviation value and the minimum deviation value corresponding to the G pixel are represented by the dotted lines GDmax and GDmin, respectively. Further, the maximum deviation value and the minimum deviation value corresponding to the B pixel are represented by solid lines BDmax and BDmin, respectively.

Next, each of the RGB maximum deviation value and the minimum deviation value shown in FIG. 22 is smoothed by the moving average method. Specifically, for each unit area UA, the moving average value is calculated with a width of one pattern cycle on one side (left) of the unit area UA and one pattern cycle on the other side (right side) of the unit area UA. do. In this way, the moving average value in the X direction is calculated by adding "1" to the value twice the pattern period.

In FIG. 23, the maximum deviation value and the minimum deviation value after smoothing corresponding to the R pixel are represented by the alternate long and short dash lines RDmax and RDmin, respectively. Further, the maximum deviation value and the minimum deviation value after smoothing corresponding to the G pixel are represented by the dotted lines GDmax and GDmin, respectively. Further, the maximum deviation value and the minimum deviation value after smoothing corresponding to the B pixel are represented by solid lines BDmax and BDmin, respectively.

Subsequently, for each unit region UA, the sum of the average pixel value represented by the second R reference curve R2 in FIG. 17 and the maximum deviation value represented by the alternate long and short dash line RDmax in FIG. 23 corresponds to the R pixel. Calculated as the maximum difference. Further, for each unit region UA, the sum of the average pixel value represented by the second R reference curve R2 in FIG. 17 and the minimum deviation value represented by the alternate long and short dash line RDmin in FIG. 23 corresponds to the R pixel. Calculated as the minimum difference. Further, for the G pixel and the B pixel, the difference maximum value and the difference minimum value are calculated for each unit region UA as in the example of the R pixel.

In FIG. 24, the maximum difference value and the minimum difference value corresponding to the R pixel are represented by the alternate long and short dash lines R3max and R3min, respectively. Further, the maximum difference value and the minimum difference value corresponding to the G pixel are represented by dotted lines G3max and G3min, respectively. Further, the maximum difference value and the minimum difference value corresponding to the B pixel are represented by solid lines B3max and B3min, respectively.

Subsequently, for each of the RGB pixels, the range from the difference minimum value calculated for the predetermined unit region UA to the difference maximum value is determined as a reference range. In this example, the range from the minimum difference value to the maximum difference value corresponding to the unit region UA located at the center of the substrate W is determined as a reference range. Further, the range from the difference minimum value to the difference maximum value corresponding to each unit area UA is corrected so as to match the reference range, and the value of each pixel in the band-shaped area BA extending from each unit area UA in the Y direction is adjusted. Correct to fit the corrected range.

Regarding this correction process, the value of the R pixel of any pixel on the image of the substrate W is expressed as R _{(x, y)} using the plane coordinate system defined by the axes in the X direction and the Y direction, and the value of the pixel is expressed as R (x, y). The maximum and minimum differences on the alternate long and short dash lines R3max and R3min in FIG. 24 corresponding to the pixel positions in the X direction are expressed as maxR _x and minR _x , and the R pixels corresponding to the unit region UA located at the center of the substrate W. When the maximum difference value and the minimum difference value are expressed as maxRv and minRv, respectively, the corrected R pixel value R' _{(x, y)} can be expressed by the following equation (7).

R' _{(x, y)} = {(R _{(x, y)} -minR _x ) ÷ (maxR _x -minR _x )} × (maxRv-minRv) + minRv ... (7)
Further, the value of the G pixel of any pixel on the image of the substrate W is expressed as G _{(x, y)} , and the maximum difference value on the dotted lines G3max and G3min in FIG. 24 corresponding to the pixel position of the pixel in the X direction. And, when the difference minimum value is expressed as maxG _x and minG _x , and the difference maximum value and the difference minimum value of the G pixel corresponding to the unit region UA located at the center of the substrate W are expressed as maxGv and minGv, respectively, after correction. The value G' _{(x, y)} of the G pixel of can be expressed by the following equation (8).

G' _{(x, y)} = {(G _{(x, y)} -minG _x ) ÷ (maxG _x -minG _x )} × (maxGv-minGv) + minGv ... (8)
Further, the value of the B pixel of any pixel on the image of the substrate W is expressed as B _{(x, y)} , and the maximum difference value on the solid lines B3max and B3min in FIG. 24 corresponding to the pixel position in the X direction of the pixel. And, when the difference minimum value is expressed as maxB _x and minB _x , and the difference maximum value and the difference minimum value of the B pixel corresponding to the unit region UA located at the center of the substrate W are expressed as maxBv and minBv, respectively, after correction. The value B' _{(x, y)} of the B pixel of can be expressed by the following equation (9).

B' _{(x, y)} = {(B _{(x, y)} -minB _x ) ÷ (maxB _x -minB _x )} × (maxBv-minBv) + minBv ... (9)
Display image data is generated by correcting all the pixels of the actual image data using the above equations (7), (8), and (9). As a result, in the image represented by the display image data, the average colors of the plurality of strip-shaped regions BA on one surface of the substrate W in the left-right direction are equal to or substantially equal to each other.

(2) Control system of the board inspection device 200 The control system of the board inspection device 200 according to the second embodiment has the same configuration as the example of FIG. 19 except for the functional configuration of the display image data generation unit 405. Have. FIG. 25 is a block diagram showing a functional configuration of the display image data generation unit 405 according to the second embodiment.

As shown in FIG. 25, the display image data generation unit 405 according to the second embodiment has a smoothing unit 421, a deviation calculation unit 422, a deviation maximum value calculation unit 423, a deviation minimum value calculation unit 424, and a difference maximum value calculation. It includes a unit 425, a difference minimum value calculation unit 426, a reference range determination unit 427, and a correction unit 428.

Similar to the smoothing unit 411 of FIG. 19 according to the first embodiment, the smoothing unit 421 has a plurality of unit regions UA (FIG. 6) arranged in the X direction on the image based on the actual image data. Calculate the average pixel value of the pixels of. Further, the smoothing unit 421 performs smoothing processing by the moving median method on the average pixel values of the plurality of unit regions UA.

Similar to the smoothing unit 411, the smoothing unit 421 estimates ideal average pixel values of a plurality of unit regions UA located at both ends of the substrate W, and based on the estimation result, both ends of the substrate W. The average pixel value of a plurality of unit regions UA located in the unit may be determined.

The deviation calculation unit 422 calculates the difference between the average pixel value before smoothing and the average pixel value after smoothing of each unit region UA as a deviation evaluation value. The deviation maximum value calculation unit 423 calculates the deviation maximum value for each unit region UA by performing smoothing processing by the moving maximum method for the deviation evaluation value of each unit region UA. The deviation minimum value calculation unit 424 calculates the deviation minimum value for each unit region UA by performing smoothing processing by the movement minimum method for the deviation evaluation value of each unit region UA.

The maximum deviation value calculation unit 423 may perform smoothing processing by the moving average method on the maximum deviation value calculated for the plurality of unit regions UA. Further, the deviation minimum value calculation unit 424 may perform smoothing processing by the moving average method on the deviation minimum values calculated for the plurality of unit regions UA.

The difference maximum value calculation unit 425 calculates the sum of the average pixel value and the deviation maximum value after smoothing by the moving median method for each unit region UA as the difference maximum value. The difference minimum value calculation unit 426 calculates the sum of the average pixel value and the deviation minimum value after smoothing by the moving median method for each unit region UA as the difference minimum value. The reference range determination unit 427 determines the range from the minimum difference value to the maximum difference value calculated for the unit region UA located at the center of the substrate W as the reference range.

The correction unit 428 corrects the range from the difference minimum value to the difference maximum value corresponding to each unit area UA so as to match the reference range, and the band-shaped area BA extending in the Y direction from each unit area UA (FIG. 18). The value of each pixel in is corrected so as to fit in the corrected range. As a result, display image data is generated.

(3) Defect determination process In the defect determination process according to the second embodiment, the process for generating display image data is different from the defect determination process according to the first embodiment. FIG. 26 is a flowchart of the defect determination process according to the second embodiment. In the following description, a series of processes from steps S11 to S14 in the defect determination process according to the first embodiment shown in FIG. 20 will be referred to as an automatic determination process. Further, here, the substrate W to be inspected is referred to as an inspection substrate W.

As shown in FIG. 26, when the defect determination process is started, the control unit 401, the image generation unit 402, the determination unit 403, and the image determination storage unit 404 of FIG. 19 perform the automatic determination process (step S20). Next, the smoothing unit 421 of FIG. 25 uses the image data of the inspection board W generated by the automatic determination process as the actual image data, and a plurality of unit regions UA (FIG. 2) arranged in the X direction on the image based on the actual image data. For each of 6), the average pixel value of the plurality of pixels constituting the unit region UA is calculated (step S21). Further, the smoothing unit 421 performs a smoothing process on the average pixel value of the plurality of unit regions UA (step S22).

Next, the deviation calculation unit 422 of FIG. 25 calculates the deviation evaluation value for each unit region UA (step S23). Further, the maximum deviation value calculation unit 423 in FIG. 25 calculates the maximum deviation value for each unit region UA by performing smoothing processing on the calculated deviation evaluation value by the moving maximum method (step S24). Further, the deviation minimum value calculation unit 424 in FIG. 25 calculates the deviation minimum value for each unit region UA by performing smoothing processing on the calculated deviation evaluation value by the movement minimum method (step S25).

After that, the difference maximum value calculation unit 425 of FIG. 25 calculates the difference maximum value by adding the smoothed average pixel value and the deviation maximum value for each unit region UA (step S26). Further, the difference minimum value calculation unit 426 of FIG. 25 calculates the difference minimum value by adding the smoothed average pixel value and the deviation minimum value for each unit region UA (step S27).

Next, the reference range determination unit 427 of FIG. 25 determines a range from the calculated difference minimum value to the difference maximum value for the unit region UA located at the center of the inspection board W as the reference range (step S28). Finally, the correction unit 428 of FIG. 25 corrects the range from the difference minimum value to the difference maximum value of each unit area UA so as to match the reference range, and the band-shaped area BA extending from each unit area UA in the Y direction. The value of each pixel in FIG. 18 is corrected so as to fit in the corrected range (step S29). As a result, display image data is generated. The image determination storage unit 404 of FIG. 19 stores display image data (step S30).

After that, the image determination storage unit 404 causes the display unit 280 of FIG. 19 to display an image based on the automatic determination result and the display image data (step S31). In this state, the user can visually inspect one side of the inspection board W by visually recognizing the image of the inspection board W displayed on the display unit 280.

In the defect determination process according to the present embodiment, of the automatic determination processes in step S20, the processes of steps S13 and S14 in FIG. 20 may be performed after the processes of steps S21 to S30, or steps S21 to S30. It may be performed in parallel with the processing of.

(4) Effect of the Second Embodiment (a) In the substrate inspection apparatus 200 according to the second embodiment, a plurality of units are similar to the procedure for generating display image data in the first embodiment. The average pixel value shown in FIG. 16 is calculated for each of the region UAs. Further, the average pixel value of the plurality of unit regions UA is smoothed by the moving median method, so that the average pixel value after smoothing shown in FIG. 17 is calculated. At this time, by performing the smoothing process by the moving median method, the variation in the local average pixel value is appropriately reduced.

After that, the difference between the average pixel value before smoothing and the average pixel value after smoothing of each unit region UA is calculated as a deviation evaluation value. The deviation evaluation values of the plurality of unit areas UA are smoothed by the moving maximum method, and the plurality of deviation evaluation values after smoothing are calculated as the plurality of deviation maximum values. Further, the deviation evaluation values of the plurality of unit regions UA are smoothed by the movement minimum method, and the plurality of deviation evaluation values after smoothing are calculated as the plurality of deviation minimum values.

For each unit region UA, the maximum difference value is calculated by adding the average pixel value and the maximum deviation value after smoothing by the moving median method. Further, for each unit region UA, the difference minimum value is calculated by adding the average pixel value and the deviation minimum value after smoothing by the moving median method.

The range from the difference minimum value to the difference maximum value corresponding to the unit region UA located at the center of the substrate W is determined as the reference range. Each unit area The range from the difference minimum value to the difference maximum value corresponding to the unit area UA is corrected so as to match the reference range, and each pixel in each band-shaped area BA extending in the Y direction from each unit area UA is corrected. The value is corrected to fit the corrected range. As a result, in the image based on the display image data, the average value of the plurality of pixels in each band-shaped region BA is equal to or substantially equal to the average pixel value of the unit region UA located at the center of the substrate W. Will be equal. Further, the value of each pixel in each band-shaped region BA is represented so as to conform to the reference range. Therefore, in the image based on the display image data, it is suppressed that the average color of the plurality of portions on one surface of the substrate W in the X direction is visually recognized as different. As a result, it becomes possible to easily and accurately determine the presence or absence of defects in the surface state of the substrate W in a visual inspection using an image based on display image data.

(B) In the present embodiment, the average pixel value calculation process, the average pixel value smoothing process, and the deviation evaluation value are used for each of the types of R pixel, G pixel, and B pixel constituting one pixel. Calculation processing, calculation processing of the maximum difference value, calculation processing of the minimum difference value, calculation processing of the maximum difference value, calculation processing of the minimum difference value, determination processing of the reference range, and correction processing of the actual image data are performed. As a result, it is suppressed that the average hues of the plurality of portions of the surface of the substrate W in the X direction are visually recognized as different on the image based on the display image data.

[3] Third Embodiment The substrate processing apparatus according to the third embodiment includes the substrate inspection device 200 according to the first or second embodiment. FIG. 27 is a schematic block diagram showing the overall configuration of the substrate processing apparatus according to the third embodiment. As shown in FIG. 27, the substrate processing apparatus 100 is provided adjacent to the exposure apparatus 500 and includes the substrate inspection apparatus 200 according to the first or second embodiment, as well as the control apparatus 110 and the transfer apparatus 120. It includes a coating processing unit 130, a developing processing unit 140, and a heat treatment unit 150.

The control device 110 includes, for example, a CPU and a memory, or a microcomputer, and controls the operations of the transfer device 120, the coating processing unit 130, the development processing unit 140, and the heat treatment unit 150. Further, the control device 110 gives a command for inspecting the surface state of one surface of the substrate W to the control device 400 (FIG. 1) of the substrate inspection device 200.

The transport device 120 transports the substrate W between the coating processing unit 130, the development processing unit 140, the heat treatment unit 150, the substrate inspection device 200, and the exposure device 500.

The coating processing unit 130 includes a plurality of processing units PU. The processing unit PU is provided with a processing liquid nozzle 132 that supplies a processing liquid for forming a resist film on the substrate W rotated by the spin chuck 131. Each processing unit PU forms a resist film on one surface of the untreated substrate W (coating treatment). The substrate W after the coating process on which the resist film is formed is exposed by the exposure apparatus 500.

The development processing unit 140 performs the development processing of the substrate W by supplying the developer to the substrate W after the exposure processing by the exposure apparatus 500. The heat treatment unit 150 heat-treats the substrate W before and after the coating process by the coating process unit 130, the development process by the development process unit 140, and the exposure process by the exposure apparatus 500.

The substrate inspection apparatus 200 inspects the substrate W (defect determination processing) after the resist film is formed by the coating processing unit 130. For example, the substrate inspection device 200 inspects the substrate W after the coating process by the coating process unit 130 and before the exposure process by the exposure device 500. At this time, the user can visually inspect the substrate W by visually recognizing the image displayed on the display unit 280 based on the display image data.

The transport device 120 transports the substrate W determined to have no defects to the exposure device 500. On the other hand, the transfer device 120 does not transfer the substrate W determined to be defective to the exposure device 500. As a result, it is possible to prevent the exposure process from being performed on the substrate W on which the defect is present.

Further, the substrate inspection device 200 may inspect the substrate W after the coating process by the coating process unit 130, the exposure process by the exposure device 500, and the development process by the development processing unit 140. Alternatively, the substrate inspection device 200 may inspect the substrate W after the coating treatment by the coating processing unit 130, after the exposure processing by the exposure device 500, and before the development processing by the development processing unit 140. Even in these cases, the user can visually inspect the substrate W by visually recognizing the image based on the display image data generated at the time of inspecting the substrate W.

In the substrate processing apparatus 100 described above, the coating processing unit 130 may be provided with a processing unit for forming an antireflection film on the substrate W. In this case, the heat treatment unit 150 may perform an adhesion strengthening treatment for improving the adhesion between the substrate W and the antireflection film. Further, the coating processing unit 130 may be provided with a processing unit for forming a resist cover film for protecting the resist film formed on the substrate W.

When the antireflection film and the resist cover film are formed on one surface of the substrate W, the substrate W may be inspected by the substrate inspection device 200 after the formation of each film.

In the substrate processing apparatus 100 according to the present embodiment, the surface state on one surface of the substrate W on which a film such as a resist film, an antireflection film, or a resist cover film is formed is inspected by the substrate inspection apparatus 200 of FIG. .. This makes it possible to easily and accurately determine the presence or absence of defects in the film formed on one surface of the substrate W in a visual inspection using an image showing one surface of the substrate W. As a result, the occurrence of processing defects in the substrate W of the substrate W is reduced.

[4] Other Embodiments (a) In the above embodiment, the width of each unit region UA set on the image based on the actual image data in the X direction is set to one pixel, but the present invention has the present invention. Not limited to this. The width of each unit region UA in the X direction may have a width of several pixels of about 2 pixels or 3 pixels.

(B) In the first embodiment, in the correction process for generating display image data, the average pixel value of the unit region UA located at the center of the substrate W among the plurality of unit region UAs is used as a reference value. Although used, the invention is not limited to this. The average pixel value of the unit region UA located at a predetermined pixel position in the X direction other than the center of the substrate W may be used as a reference value.

(C) In the second embodiment, in the correction process for generating display image data, from the difference minimum value to the difference maximum value of the unit region UA located at the center of the substrate W among the plurality of unit region UAs. Is determined as a reference range, but the present invention is not limited thereto. A range from the minimum difference value to the maximum difference value of the unit region UA located at a predetermined pixel position in the X direction other than the center of the substrate W may be determined as a reference range.

(D) In the above embodiment, since the color image sensor is used in the image pickup unit 240, each pixel data constituting the actual image data includes the R pixel value, the G pixel value, and the B pixel value. , The present invention is not limited to this. A monochromatic image sensor may be used for the image pickup unit 240. Even in this case, in the image based on the display image data, it is suppressed that the average brightness of the plurality of portions on one surface of the substrate W in the X direction is different from each other.

(E) In the above embodiment, the average pixel value is calculated for each of the plurality of unit regions UA each time the defect determination process is performed on one substrate W, but the present invention is not limited to this. When performing defect determination processing on a plurality of substrates W having a common surface structure, for example, an average of a plurality of unit regions UA from actual image data obtained by imaging a sample substrate or a first inspection substrate W. The pixel value may be calculated and the calculation result may be stored in the memory of the control device 400. In this case, when performing the defect determination processing of the second and subsequent boards W, the average pixel value of the plurality of unit areas UA stored in the memory can be used. Therefore, it is not necessary to calculate the average pixel value of the plurality of unit areas UA each time the defect determination process is performed. As a result, it is possible to simplify the processing at the time of generating the display image data for the second and subsequent images.

Alternatively, the memory of the control device 400 may store predetermined design data for the average pixel value of the plurality of unit areas UA. Even in this case, the processing at the time of generating the display image data can be simplified.

(F) In the method for generating image display data according to the first embodiment, for display based on the average pixel value shown in FIG. 16 and the average pixel value after smoothing shown in FIG. Correction processing is performed to obtain image data, but the present invention is not limited to this. Correction processing for obtaining display image data may be performed based on the average pixel value shown in FIG. 7 and the average pixel value after smoothing shown in FIG.

(G) In the method for generating image display data according to the second embodiment, for display based on the average pixel value shown in FIG. 16 and the average pixel value after smoothing shown in FIG. A deviation evaluation value for obtaining image data is calculated, but the present invention is not limited to this. A deviation evaluation value for obtaining display image data may be calculated based on the average pixel value shown in FIG. 7 and the average pixel value after smoothing shown in FIG.

(H) In the substrate inspection device 200 according to the above embodiment, the image pickup unit 240 has an image pickup region extending in one direction, but the present invention is not limited thereto. The image pickup unit 240 may have a planar imaging region. That is, a color CCD area sensor or CMOS area sensor in which a plurality of pixels are arranged in a matrix may be used as the image pickup element of the image pickup unit 240.

In this case, for example, it is necessary to relatively move the image pickup unit 240 and the substrate holding device 250 in the front-rear direction by providing the image pickup unit 240 so that the entire surface of the substrate W is covered by the image pickup region of the image pickup unit 240. Is gone. Thereby, the time required for the inspection of the substrate W can be shortened.

(I) In the above embodiment, the substrate inspection device 200 is provided with a reflection unit 230, but the present invention is not limited thereto. When the image pickup unit 240 is configured to directly receive the light from the substrate W, the reflection unit 230 may not be provided.

(J) In the above embodiment, the moving unit 260 is configured to move the substrate holding device 250 in the front-rear direction with respect to the light emitting unit 220, the reflecting unit 230, and the imaging unit 240. Not limited to. The moving unit 260 moves the light projecting unit 220, the reflecting unit 230, and the imaging unit 240 in the front-rear direction with respect to the substrate holding device 250 so that the imaging region of the imaging unit 240 passes through the entire surface of the substrate W. It may be configured in.

[5] Correspondence between each component of the claim and each element of the embodiment The example of correspondence between each component of the claim and each element of the embodiment will be described below, but the present invention is described below. Not limited to the example.

In the above embodiment, the substrate holding device 250 and the control unit 401 are examples of the holding unit, the image pickup unit 240 and the image generation unit 402 are examples of the image pickup unit, and the moving unit 260 and the control unit 401 are examples of the moving unit. This is an example.

Further, in the first embodiment, the process of smoothing the average pixel value of FIG. 16 by the moving median method is an example of the smoothing process, and in the second embodiment, the average pixel value of FIG. 16 is moved. The process of smoothing by the median method is an example of the first smoothing process, and in the second embodiment, the average pixel value shown in FIG. 16 and the average pixel value after smoothing shown in FIG. 17 are used. The deviation evaluation value calculated based on is an example of deviation.

Further, in the second embodiment, the process of smoothing the deviation evaluation value of FIG. 21 by the moving maximum method is an example of the second smoothing process, and in the second embodiment, the deviation evaluation value of FIG. 21 is used. The process of smoothing by the movement minimum method is an example of the third smoothing process.

As each component of the claim, various other components having the structure or function described in the claim can also be used.

The present invention can be effectively used for inspecting the surface of various substrates.

100 ... Board processing device, 110, 400 ... Control device, 200 ... Board inspection device, 240 ... Imaging unit, 242 ... Imaging surface, 250 ... Board holding device, 260 ... Moving unit, 280 ... Display unit, 401 ... Control unit, 402 ... Image generation unit, 403 ... Judgment unit, 404 ... Image judgment storage unit, 405 ... Display image data generation unit, 411, 421 ... Smoothing unit, 421,428 ... Correction unit, 422 ... Deviation calculation unit, 423 ... Maximum deviation value calculation unit, 424 ... Minimum deviation value calculation unit, 425 ... Maximum difference value calculation unit, 426 ... Minimum difference value calculation unit, 427 ... Reference range determination unit, B1 ... First B reference curve, B2 ... Second B reference curve, BA ... strip region, G1 ... first G reference curve, G2 ... second G reference curve, PI ... pixel, R1 ... first R reference curve, R2 ... second R reference curve, RB ... B average pixel value line, RG ... G average pixel value line, RR ... R average pixel value line, UA ... unit area, UN1, UN2 ... unnecessary range, VP ... virtual point, VS1 ... first virtual Surface, VS2 ... second virtual surface, W ... substrate, WC ... center, WE1 ... one end, WE2 ... other end

Claims (13)

The holding part that holds the board and
An imaging unit that captures an image of one surface of the substrate held by the holding unit and generates real image data representing the image of the one surface of the substrate.
For each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the plurality of units are calculated. A smoothing unit that performs smoothing processing on the average pixel value of the unit area of
The difference between the average pixel value of the predetermined unit area after the smoothing process and the average pixel value of each unit area after the smoothing process is calculated, and each unit area is calculated. A substrate inspection apparatus comprising a correction unit for adding a difference corresponding to the unit region to the value of each pixel in a band-shaped region extending parallel to the second radial direction orthogonal to the first radial direction. The substrate inspection apparatus according to claim 1, wherein the smoothing process is a smoothing process by a mobile median method. A film having a periodic pattern in the first diameter direction is formed on the one surface of the substrate.
The substrate inspection apparatus according to claim 2, wherein the width used in the smoothing process by the moving median method is larger than the period of the pattern in the first diametrical direction. Each pixel constituting each of the plurality of unit regions includes an R pixel, a B pixel, and a G pixel.
The smoothing unit performs the smoothing process for each type of pixel, and then performs the smoothing process.
The substrate according to any one of claims 1 to 3, wherein the correction unit performs the calculation processing of the plurality of differences for each pixel type and the addition processing of the plurality of differences for each pixel type. Inspection device. The holding part that holds the board and
An imaging unit that captures an image of one surface of the substrate held by the holding unit and generates real image data representing the image of the one surface of the substrate.
For each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the plurality of units are calculated. A smoothing unit that performs the first smoothing process by the moving median method for the average pixel value in the unit area of
A deviation calculation unit that calculates the difference between the average pixel value before the first smoothing process and the average pixel value after the first smoothing process in each unit region as a deviation.
A deviation maximum value calculation unit that calculates a plurality of deviations after the second smoothing process as a plurality of deviation maximum values by performing a second smoothing process by the moving maximum method for the deviations of the plurality of unit regions.
A deviation minimum value calculation unit that calculates a plurality of deviations after the third smoothing process as a plurality of deviation minimum values by performing a third smoothing process by the movement minimum method for the deviations of the plurality of unit regions.
A difference maximum value calculation unit that calculates the sum of the average pixel value after the first smoothing process of each unit area and the deviation maximum value as the difference maximum value.
A difference minimum value calculation unit that calculates the sum of the average pixel value after the first smoothing process of each unit area and the deviation minimum value as the difference minimum value.
A reference range determination unit that determines the range from the minimum difference value to the maximum difference value corresponding to a predetermined unit area as the reference range.
The range from the minimum difference value to the maximum difference value corresponding to each unit region is corrected so as to match the reference range, and the range extends from each unit region parallel to the second diameter direction orthogonal to the first diameter direction. A substrate inspection device including a correction unit that corrects the value of each pixel in each band-shaped region so as to match the corrected range. A film having a periodic pattern in the first diameter direction is formed on the one surface of the substrate.
The width used in the first smoothing process by the moving median method, the second smoothing process by the moving maximum method, and the third smoothing process by the moving minimum method is the pattern of the pattern in the first diametrical direction. The substrate inspection apparatus according to claim 5, which is larger than the cycle. Each pixel constituting each of the plurality of unit regions includes an R pixel, a B pixel, and a G pixel.
The smoothing unit performs the first smoothing process for each type of pixel, and then performs the first smoothing process.
The deviation calculation unit performs the calculation processing of the plurality of deviations for each pixel type, and performs the calculation processing.
The deviation maximum value calculation unit performs the calculation processing of the plurality of difference maximum values for each pixel type, and then performs the calculation processing.
The deviation minimum value calculation unit performs the calculation processing of the plurality of difference minimum values for each pixel type, and then performs the calculation processing.
The difference maximum value calculation unit performs the calculation processing of the plurality of difference maximum values for each pixel type, and performs the calculation processing.
The difference minimum value calculation unit performs the calculation processing of the plurality of difference minimum values for each pixel type, and performs the calculation processing.
The reference range determination unit performs the reference range determination process for each pixel type, and then performs the reference range determination process.
The correction unit performs correction processing for matching the range from the difference minimum value to the difference maximum value corresponding to each unit area to the reference range for each pixel type, and sets the value of each pixel in each band-shaped area to the corrected range. The substrate inspection apparatus according to claim 5 or 6, which performs a conforming correction process. The smoothed portion is an average to be calculated for each of the plurality of unit regions located at one end in the first radial direction of the substrate on the image based on the actual image data, which is a part of the plurality of unit regions. The target pixel value is estimated based on a plurality of average pixel values calculated for a plurality of unit regions located in a portion adjacent to the one end portion, and a plurality of unit regions located in the one end portion based on the estimation result. The substrate inspection apparatus according to any one of claims 1 to 7, wherein a plurality of average pixel values corresponding to each of the above are determined. The imaging unit has a linear imaging region extending parallel to the first radial direction on the one surface of the substrate held by the holding unit.
The board inspection device is
The claim further comprises a moving portion that relatively moves the imaging portion and the holding portion so that the imaging region passes through the one surface of the substrate held by the holding portion in the second diameter direction. The substrate inspection apparatus according to any one of 1 to 8. A coating processing unit that forms a film on one surface of the substrate by supplying the treatment liquid to one surface of the substrate.
The substrate inspection apparatus according to any one of claims 1 to 9, which inspects a substrate on which a film is formed by the coating processing unit.
A substrate processing apparatus including a transfer device for transporting a substrate between the coating processing unit and the substrate inspection device. A step of capturing an image of one surface of a substrate held by a holding unit and generating real image data representing an image of the one surface of the substrate.
For each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the plurality of units are calculated. The step of smoothing the average pixel value of the unit area of
The difference between the average pixel value of the predetermined unit area after the smoothing process and the average pixel value of each unit area after the smoothing process is calculated, and each unit area is calculated. A substrate inspection method comprising a step of adding a difference corresponding to the unit region to the value of each pixel in a band-shaped region extending parallel to the second diametrical direction orthogonal to the first diametrical direction. A step of capturing an image of one surface of a substrate held by a holding unit and generating real image data representing an image of the one surface of the substrate.
For each of the plurality of unit regions arranged in the first radial direction of the substrate on the image based on the actual image data, the average value of the plurality of pixels constituting the unit region is calculated as the average pixel value, and the plurality of units are calculated. The step of performing the first smoothing process by the moving median method for the average pixel value of the unit area of
A step of calculating the difference between the average pixel value before the first smoothing process and the average pixel value after the first smoothing process of each unit region as a deviation, and
A step of calculating the plurality of deviations after the second smoothing process as a plurality of maximum deviation values by performing a second smoothing process by the moving maximum method for the deviations of the plurality of unit regions.
A step of calculating the plurality of deviations after the third smoothing process as a plurality of deviation minimum values by performing a third smoothing process by the movement minimum method for the deviations of the plurality of unit regions.
A step of calculating the sum of the average pixel value after the first smoothing process of each unit region and the maximum deviation value as the maximum difference value, and
A step of calculating the sum of the average pixel value after the first smoothing process of each unit region and the minimum deviation value as the minimum difference value, and
The step of calculating the range from the difference minimum value to the difference maximum value corresponding to the predetermined unit area as the reference range, and
The range from the minimum difference value to the maximum difference value corresponding to each unit region is corrected so as to match the reference range, and the range extends from each unit region parallel to the second diameter direction orthogonal to the first diameter direction. A substrate inspection method comprising a step of correcting the value of each pixel in each strip region to fit the corrected range. The step of forming a film on one surface of the substrate by supplying the treatment liquid to one surface of the substrate,
A substrate processing method comprising the step of inspecting a substrate on which the film is formed on one surface by using the substrate inspection method according to claim 11 or 12.

Images