JP2003172711A - Surface inspection of inspected object using image processing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for performing surface inspection of an inspected object without needing a skilled inspection operator. <P>SOLUTION: The technique obtains a color information image MG0 with an incident white light diagonally irradiated to the surface of a printed circuit board and obtains an infrared light image IRM with an infrared light irradiated almost perpendicularly to the surface of the printed circuit board PCB. An image MG3 which displays a synthesis color area containing a golden area and a brown area by means of area separation processing based on the color information image MG0. Using the image MG3 and the infrared light image IRM, the golden area GL is discriminated from the brown area. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a surface inspection technique for an inspection object using image processing.

[0002]

2. Description of the Related Art A printed circuit board for an electronic circuit is a printed circuit board on which a conductive pattern such as circuit wiring and a conductive pad and silk characters are printed on an insulating substrate. In the inspection of the printed circuit board, the quality of the printed state of the conductor pattern and silk characters is inspected. Conventionally, the visual inspection has been mainly performed.

[0003]

In recent years, a mounting method has been used in which an electronic component such as a semiconductor chip is directly mounted on a conductor pattern on a substrate. At this time, if a defect occurs in a conductor pad or a wiring pattern that is in direct contact with an electronic component, a product defect is likely to occur, so that there is an increasing demand for inspection of the conductor pattern. On the other hand, due to the recent demand for higher integration, the area and spacing of the conductor pad portions have become smaller, making visual inspection more difficult year by year. In addition to it,
The number of skilled inspection workers tends to decrease. For this reason,
There has been a demand for a technique capable of inspecting the surface of a printed circuit board without requiring a skilled inspection operator.
Such requests are not limited to surface inspection of printed circuit boards,
Generally, it was common to the surface inspection of the inspection object.

The present invention has been made in order to solve the above-mentioned conventional problems, and provides a technique capable of inspecting the surface of an inspection object without requiring a skilled inspection operator. The purpose is to

[0005]

Means for Solving the Problem and Its Function and Effect In order to achieve the above object, the present invention provides a method and an apparatus for inspecting the surface condition of an inspection object.
The inspection apparatus includes a first light source for irradiating the surface of the inspection object with a predetermined first light from an oblique direction, and the first light source for emitting light from a direction substantially perpendicular to the surface of the inspection object.
An illumination optical system including a second light source for irradiating a second light having a spectrum on a wavelength side longer than that of the light, and an image component relating to at least two color components included in the first light. The first image including the second image is acquired, and the second image is acquired.
An image capturing section for acquiring a second image of the light of the
And an area identification unit for identifying a specific color area having a specific color on the surface of the inspection object based on the image and the second image.

The reason for irradiating the first light from a direction oblique to the surface is that it is difficult to obtain color information of the surface when the light is irradiated from a direction substantially perpendicular to the surface. That is, if light is irradiated from a direction substantially perpendicular to the surface, specular reflection (specular reflection) may occur, and the color information of the surface may not be obtained. In the present invention, since the first light is emitted from the direction oblique to the surface, the first image showing the color of the surface can be obtained. Further, when the second light is applied from a direction substantially perpendicular to the surface, a second image that reflects the difference in the regular reflectance for each area of the surface can be obtained. In the present invention, since the specific color area is identified based on the first image and the second image thus obtained, the specific color area of the inspection object can be recognized by image processing. Using this, surface inspection can be performed easily.

The area identifying section has a function of identifying, from the first image, an integrated color area including the specific color area and another color area having a color close to the specific color area. , And a function of identifying the specific color area from the integrated color area by using the gradation value of the second image regarding the second light. At this time, the spectrum of the second light is set such that the contrast between the specific color area and the other color area is larger in the second image than in the first image. It is preferable.

According to this structure, it is possible to identify a specific color area from the integrated color areas by utilizing the contrast in the second image. Therefore, even when it is difficult to distinguish between the specific color area and another color area having a color close to the specific color area by the image processing using the first image, it is possible to easily distinguish between them.

When the inspection object is a printed circuit board for an electronic circuit and the specific color area is an area plated with gold, it is preferable that the second light is infrared light.

According to this structure, it is possible to easily identify the gold-plated area on the printed circuit board.

The image pickup section includes a first image pickup section for obtaining the first image and a second image pickup section for picking up the second image at the same time as the image pickup by the first image pickup section. , May be included. At this time, the first light may be white light and the second light may be infrared light.

According to this structure, since the inspection can be performed in a relatively short time, the inspection throughput is improved.

Alternatively, the image pickup section may execute the acquisition of the first image and the acquisition of the second image by using the same image pickup device at different times.

According to this structure, since only one image pickup element is required, the number of parts of the image pickup section can be reduced.

It is preferable that the second light is convergent light that converges into a relatively small light spot on the surface of the inspection object.

According to this structure, the amount of light per unit area on the surface can be increased. Further, even when the surface has irregularities, the proportion of light scattered by the irregularities can be reduced, so that it is possible to prevent the amount of light reaching the image sensor from being excessively reduced.

Alternatively, the image pickup unit includes an image forming optical system housed in a lens barrel and the first image forming optical system passing through the image forming optical system.
Of the first image pickup unit for receiving the first image by receiving the second light, and for receiving the second light that has passed through the imaging optical system at the same time as the image pickup by the first image pickup unit. And a second image pickup unit that picks up the second image, the first light is visible light, and the second light has a main wavelength range different from that of the visible light. Infrared light that does not overlap, the visible light and the infrared light reflected from the inspection object passes through at least a part of the lenses of the imaging optical system housed in the lens barrel in common. It is also possible to form an image on each of the first and second imaging units.

In this configuration, since the lenses in which visible light and infrared light are housed in the same lens barrel are commonly used, the number of parts can be reduced and the optical system can be simplified. Further, since visible light and infrared light do not overlap with each other in their main wavelength ranges, two images can be taken simultaneously without interfering with each other.

The infrared light is convergent light that converges into a small light spot on the surface of the inspection object, and the convergence angle of the infrared light is set to be larger than the convergence angle of the visible light. May be.

Since the infrared light irradiates the inspection object from a substantially vertical direction, if the inspection object has unevenness, the reflected light amount may significantly change. In such a case, by increasing the convergence angle of the infrared light, it is possible to reduce the influence of the unevenness of the inspection object on the reflected light amount.

The illumination optical system may include a plate type dichroic mirror for separating the light reflected by the surface into the first light and the second light.

Since the plate type dichroic mirror has a better light separation characteristic than a prism type dichroic mirror (so-called dichroic prism), the first and second lights can be separated more efficiently.

The image pickup section has a first image pickup element for obtaining the first image and a second image pickup element for picking up the second image. 1 and 2
The image pickup devices may be devices having the same characteristics.

According to this structure, there is an advantage that the attachment mechanism of the image pickup device and the adjustment method thereof can be made common.

The present invention can be implemented in various forms, for example, a method and apparatus for inspecting the surface of an object to be inspected, a method and apparatus for recognizing or extracting a region on the surface of the object, and those methods. Alternatively, it can be realized in the form of a computer program for realizing the functions of the apparatus, a recording medium recording the computer program, a data signal embodied in a carrier wave including the computer program, and the like.

[0026]

BEST MODE FOR CARRYING OUT THE INVENTION Next, embodiments of the present invention will be described in the following order based on examples. A. First Example: B. Second Embodiment: C.I. Third Example: D. Fourth Example: E. Fifth Example: F.I. Sixth Embodiment: G.I. Details of region separation processing: Modification:

A. First Embodiment: FIG. 1 is a block diagram showing the configuration of an inspection apparatus as a first embodiment of the present invention.
This inspection device is a device for inspecting a printed circuit board PCB for an electronic circuit, and includes a host processor 100,
Imaging unit 200, drive mechanism 300, and illumination optical system 400
And are equipped with.

The host processor 100 has a function of controlling the operation of the entire inspection apparatus, and also has an area dividing section 102, an area recognizing section 104, and a reference image creating section 10.
6 and the function of the inspection execution unit 108. These functions are realized by a computer program stored in a recording medium (not shown) of the host processor 100. The contents of these functions will be described later.

The image pickup section 200 includes two CCD elements 21.
0, 212, A / D converter 220, and image capturing unit 23
It has 0 and. The first CCD element 210 is used when irradiating the surface of the printed circuit board PCB with white light to capture a color multi-gradation image. The second CCD element 212 is used when the surface of the printed circuit board PCB is irradiated with infrared light to capture a multi-tone image of infrared light.

The drive mechanism 300 includes an XY stage 310 on which a printed circuit board PCB is placed, and an XY stage 310.
And a drive control unit 330 for controlling the drive device 320. The drive mechanism 300 is used to position the printed circuit board PCB at a desired position with respect to the illumination optical system 400.

The illumination optical system 400 includes a white light source 410,
An infrared light source 420, a half prism 430, a lens system 440, and a dichroic prism 450 are provided. In this embodiment, a halogen lamp with an infrared cut filter is used as the white light source 410, and an infrared LED that emits near infrared light is used as the infrared light source 420.

FIG. 2 is a graph showing spectra of white light and infrared light. The solid line is the spectrum of the halogen light emitted from the halogen lamp, and the alternate long and short dash line is the spectrum of the halogen light after passing through the infrared cut filter. The broken line is the spectrum of infrared light emitted from the infrared LED. As can be understood from this figure, the white light and the infrared light used in the present embodiment are adjusted so as to have spectra in which the main wavelength regions hardly overlap with each other. Here, the “main wavelength range” means a wavelength range having an intensity of 20% or more of the intensity peak value.

The white light source 410 irradiates the surface of the printed circuit board PCB with white light from an oblique direction. When the surface of the printed circuit board PCB is obliquely irradiated with white light, light having a wavelength corresponding to the color of each pixel on the surface is reflected in the normal direction. Therefore, if this reflected light is acquired by the CCD element 210, a color image including color information of the surface of the printed circuit board PCB can be obtained.

The reflected light reflected on the surface of the printed circuit board PCB in the normal direction is reflected by the half prism 430 and the lens system 4.
40, is reflected by the dichroic prism 450, and is incident on the first CCD element 210. This CCD
The element 210 is a color CCD and is capable of outputting output signals regarding three color components of RGB. C
The output signals for the three colors of the CD element 210 are supplied to the A / D converter 2
Each is converted into digital data at 20. These digital data are captured by the image capturing unit 230 and stored in the image capturing unit 230 as color image data.

The infrared light emitted from the infrared light source 420 is
PCB reflected by half prism 430
Is incident almost vertically on the surface of. Where "almost vertical"
Means a range of 90 ± 5 degrees. The infrared light is specularly reflected (specularly reflected) on the surface of the printed circuit board PCB, and enters the second CCD element 212 via the half prism 430, the lens system 440, and the dichroic prism 450. The output signal of the CCD element 212 is the A / D converter 2
It is converted into digital data at 20 and is captured by the image capturing section 230 as multi-tone image data regarding infrared light.

The second CCD element 212 is a monochrome CC
Although D may be used, the same element as the first CCD element 210 may be used. If the same elements are used as the two CCD elements 210 and 212, there is an advantage that the attachment mechanism and the adjustment method for them can be made common.

In the following, a color multi-tone image acquired using white light or visible light will be referred to as a "color information image", and a monochrome multi-tone image acquired using infrared light will be referred to as "red information image". It is called an "outside light image".

FIG. 3 is a flow chart showing the inspection processing procedure in the first embodiment. In step T1, a master substrate (also referred to as a “reference substrate”) is prepared, and a color information image and an infrared light image of the master substrate are simultaneously captured using the inspection device in FIG. Here, the "master board" means a standard printed circuit board PCB used to obtain various reference images used for surface inspection of the printed circuit board PCB. As the master substrate, for example,
There are almost no defects and conductor patterns and silk characters are printed almost as designed.

FIG. 4 is an explanatory view showing a color information image of the printed circuit board PCB. The surface of the printed circuit board PCB has a first green region G1 in which a resist is applied on the substrate base.
And the second green area G where the resist is applied on the copper wiring.
2, a gold area GL plated with gold, a brown area BR of the substrate base, and a white area WH in which white silk characters are printed on the substrate base. Since the substrate base which is the base of the first green area G1 is brown and the copper wiring which is the base of the second green area G2 is copper colored, the colors of these two areas G1 and G2 are slightly different. , Both are still green. Therefore, in the present embodiment, these two green regions G1 and G2 are collectively referred to as "green region GR". In the processing described below, the two green areas G1 and G2 are treated as one green area GR (resist area).

In step T2 of FIG. 3, the area dividing unit 10
2 (FIG. 1) performs the area separation of the color information image. FIG. 5 is an explanatory diagram showing the processing contents of steps T2 to T6 of FIG. As shown here, the color information image MG
When the region separation of 0 is performed, the white region image MG1 representing the white region WH, the green region image MG2 representing the resist region GR (G1 + G2), the gold region GL and the brown region BR.
A gold / brown region image MG3 representing the integrated region (integrated color region) is obtained. These three color region images MG1 to MG3 are binary images or multivalued images. If the color region images MG1 to MG3 are binary images, the following processing becomes easier. The details of this area separation processing will be described later.

The reason why the gold area GL (gold plated area) and the brown area BR (base area) are not separated in the area separation process and are extracted as an integrated area is that the colors of these two areas GL and BR are extracted. Because they are close to each other. That is, there are considerable shades in the gold region GL,
Depending on how the light hits, there are areas that appear brown. Therefore, in the area separation of the color information image MG0, these two areas may not be well separated. Therefore, in the area separation using the color information image MG0, an integrated color area including gold and brown areas is extracted, and as described later,
The infrared region is used to separate the gold region GL and the brown region BR.

In step T3 of FIG. 3, the reference image creating unit 106 creates a first reference image RG1 for silk character inspection from the white area image MG1. The first reference image RG1 is obtained by thickening the white area in the white area image MG1 by a predetermined width. The reason for performing such thickening processing is to allow the silk character to have an allowable error in its position and size. In step T4, similarly to step T3, the reference image creating unit 106 thickens the green area image MG2 to perform the second resist area inspection.
The reference image RG2 is created.

In step T5, the area recognition unit 104 acquires the gold / brown area image MG3 as a mask image, and in step T6, the infrared light image IRM is masked using this mask image MG3. As a result of this mask processing, the gold area image MG4 representing the gold area GL is recognized. Specifically, an area corresponding to the golden / brown area (GL + BR) of the mask image MG3 is extracted from the infrared light image IRM, and an area having a brightness higher than a predetermined threshold value is extracted from the extracted image. It is recognized as a gold area GL.

The reason why the gold area GL can be distinguished from the brown area BR by this mask processing is as follows. FIG. 6 shows a gold area GL (gold plated area) and a brown area BR.
It is a graph which shows the spectral reflectance characteristic of (base region).
In general, the contrast of two areas is proportional to the ratio of their reflectivities. Therefore, the gold area GL and the brown area BR
The contrast is relatively small in the visible light region and relatively large in the infrared light region. The visible light region of FIG. 6 corresponds to the wavelength region of white light emitted from the white light source 410 as shown in FIG. 2, and the infrared light region of FIG. 6 emits from the infrared light source 420. It corresponds to the wavelength range of infrared light that is generated. In the infrared light image IRM (FIG. 5), since the contrast between the gold area GL and the brown area BR is large, the color information image M
Only the gold area GL can be identified from the gold / brown areas extracted from G0.

The reference image creation unit 106 (FIG. 1) creates the third reference image RG3 for gold plating area inspection by thickening the image MG4 (FIG. 5) representing the recognized gold area GL. (Fig. 5). In this way, the preparation of the first reference image RG1 for silk character inspection, the second reference image RG2 for resist area inspection, and the third reference image RG3 for gold plating area inspection is completed.

In step T7 of FIG. 3, the substrate to be inspected is inspected using these three reference images RG1 to RG3. Specifically, X of the inspection apparatus shown in FIG.
The printed circuit board PCB to be inspected is placed on the Y stage 310, and the color information image MG0 and the infrared light image IRM are acquired. Then, the inspection execution unit 108 (FIG. 1) executes predetermined image processing using these images MG0 and IRM and the three reference images RG1 to RG3 relating to the master substrate. In this inspection, for example, it is determined whether or not the silk character area, the resist area, and the gold pad area are within their respective allowable ranges.

As inspection items, defective shapes of silk characters, defective shapes of resist, defective shapes of gold-plated portions, unevenness, and the like can be adopted. If the gold-plated portion has irregularities, a large difference occurs in the gradation value between the concave portion and the convex portion in the infrared light image IRM. Therefore, if the gold color region GL of the inspection target image is extracted using the infrared light image IRM, it is also possible to detect abnormal irregularities of the gold-plated portion.

As described above, in the first embodiment, the gold area GL is recognized from the gold / brown areas extracted by the area separation of the color information image MG0 by utilizing the contrast of the infrared light image IRM. Since the color information image M
Gold area GL and brown area B, which are difficult to identify in G0
It is possible to distinguish R and R easily and accurately. Further, in the first embodiment, the color information image M is formed by using the dichroic prism 450 and the two CCD elements 210 and 212.
Since G0 and the infrared light image IRM are imaged at the same time, the inspection time can be shortened and the inspection throughput can be improved.

B. Second Embodiment: FIG. 7 is a block diagram showing the configuration of the inspection apparatus of the first embodiment. This inspection apparatus has a configuration in which the dichroic prism 450 of the apparatus of the first embodiment shown in FIG. 1 is replaced with a dichroic mirror 460 and a correction glass flat plate 470 is added to the image side of the dichroic mirror 460. . The other points are the same as in the first embodiment.

The reason why the dichroic prism 450 is replaced with the dichroic mirror 460 is that the dichroic mirror 460 has a better separation characteristic for white light and infrared light. FIG. 8 shows the dichroic prism 4
It is a graph which shows the transmittance | permeability characteristic of 50 (cube type dichroic mirror) and the dichroic mirror 460 (plate type dichroic mirror). Each of the two types of dichroic mirrors in FIG. 7 is designed to reflect light having a wavelength of less than 700 nm and transmit light having a wavelength of 700 nm or more. Since the plate type has better light separation characteristics depending on the wavelength than the cube type,
It is preferable to use a plate type dichroic mirror.

However, the plate type dichroic mirror tends to cause astigmatism. The correction glass plate 470 is for correcting this astigmatism.
The correction glass flat plate 470 has a flat plate shape similar to that of the dichroic mirror 460, and its orientation with respect to the dichroic mirror 460 is 9 with respect to the vertical axis.
It is installed so that it faces in a direction different by 0 degrees. The correction glass flat plate 470 looks like a block in FIG. 7 because the tilted flat plate is viewed from the front.

C. Third Embodiment: FIG. 9 is an explanatory diagram showing the configuration of an infrared light source 420 in the third embodiment. The infrared light source 420 includes a near infrared LED 422 and a convex lens 424. In the configuration of FIG. 9A, the light emitted from the infrared light source 420 is substantially parallel light, and the surface of the printed circuit board PCB is illuminated by this parallel light. On the other hand, FIG.
In the configuration of (b), the light emitted from the infrared light source 420 is convergent light and is converged into a small light spot on the surface of the printed circuit board PCB. The configuration other than the infrared light source 420 can be the same as the configuration of the first or second embodiment described above.

As shown in FIG. 9B, the printed circuit board PCB
By irradiating convergent light that converges to a relatively small light spot on the surface of, it is possible to increase the amount of light per unit area. Further, when the surface of the printed circuit board PCB has irregularities, the amount of reflected light that reaches the CCD element 212 can be increased. The reason for this is that when the surface of the printed circuit board PCB is illuminated with parallel light, the light is largely scattered by the unevenness on the surface, and the CCD element 21
This is because the amount of light reaching 2 decreases.

In the case of convergent light as shown in FIG. 9B, the outermost rays of the bundle of rays are considered and the angle 2θ formed by these rays (hereinafter referred to as “convergence angle 2θ”) is defined. You can In addition, the numerical aperture NA of this ray bundle is NA = si
It can be defined as nθ. In the case of parallel light as shown in FIG. 9A, the convergence angle 2θ of the ray bundle is zero. As can be understood from this description, the convergence angle 2θ of the ray bundle is
It is also possible to define the non-convergence angle.

By increasing the convergence angle 2θ of the light (eg, infrared light) radiated almost perpendicularly to the printed circuit board PCB, as described above, the influence of the unevenness of the surface of the printed circuit board PCB on the detected light amount can be reduced. You can On the other hand, with respect to light emitted from an oblique direction (for example, visible light or white light), even if the convergence angle 2θ is small, the unevenness of the surface of the printed circuit board PCB has a small effect on the detected light amount. Therefore, in general, it is preferable to make the convergence angle 2θ of the light flux emitted from the vertical direction larger than the convergence angle 2θ of the light flux emitted from the oblique direction to the printed circuit board PCB. Such characteristics are applicable to the above-described first and second embodiments, and also applicable to other embodiments described later.

D. Fourth Embodiment: FIG. 10 is a flowchart showing the inspection processing procedure in the fourth embodiment. As the configuration of the device, any of the above-mentioned first to third embodiments can be adopted.

In the processing procedure of FIG. 10, the first step T1 of the processing procedure of the first embodiment shown in FIG.
In addition, step T12 is inserted between step T5 and step T6. Step T11
Then, only the white light is emitted without irradiating the infrared light to obtain the color image (color information image) of the master substrate. The processing of steps T2 to T5 using this color image is the same as in the first embodiment (FIG. 3).

Next, in step T12, only infrared light is irradiated without irradiating white light to acquire an infrared light image of the master substrate. The subsequent processing is the first embodiment (FIG. 3).
Is the same as.

As described above, when the color information image MG0 and the infrared light image IRM are imaged at different times, the main wavelength regions of the white light and the infrared light may be set to be slightly overlapped with each other. Therefore, as the dichroic prism 450 (FIG. 1) and the dichroic mirror (FIG. 7), those having somewhat inferior wavelength separation performance may be used. It is also possible to omit the dichroic prism 450 and the CCD element 210. This has the advantage that the device configuration can be simplified. On the other hand, in the first and second embodiments, since the color information image MG0 and the infrared light image IRM are acquired at the same time, there is an advantage that the inspection time can be further shortened as compared with the fourth embodiment. .

E. Fifth Embodiment: FIG. 11 is a block diagram showing the arrangement of an inspection apparatus according to the fifth embodiment. In this inspection apparatus, the illumination optical system is separated into a first optical system 402 and a second optical system 404. The first optical system 402 is used when acquiring the color information image MG0. The second optical system 404 is used when acquiring the infrared light image IRM. The first optical system 402 has a white light source 410 and a lens system 440. White light emitted from the white light source 410 illuminates the surface of the printed circuit board PCB from an oblique direction, and reflected light in the normal direction is guided to the first CCD element 210 using the lens system 440. Second optical system 4
04 is an infrared light source 420, a half prism 430,
And a lens system 442. The infrared light emitted from the infrared light source 420 is reflected by the half prism 430 to illuminate the surface of the printed circuit board PCB substantially vertically, and the reflected light is transmitted through the half prism 430 and the lens system 442 to the second CCD. It is guided to the element 212.

At the time of image pickup, first, the printed circuit board PCB is positioned below the first optical system 402, and the first CCD
The color information image MG0 is acquired using the element 210. Then, using the XY stage 310, the printed circuit board PCB
Is positioned below the second optical system 404, and the second CC
The infrared light image IRM is acquired using the D element 212.
The processing procedure may be the same as that described with reference to FIG.

However, in the fifth embodiment, when the infrared light image of one printed circuit board PCB is acquired, the color information image of the next printed circuit board PCB can be acquired at the same time. Therefore, it is possible to obtain a throughput that is substantially the same as that of the first and second embodiments.

The inspection apparatus of the fifth embodiment does not require the dichroic prism 450 (FIG. 1) or the dichroic mirror (FIG. 7), and the number of parts is smaller than those of the first and second embodiments. There are advantages. Further, in the fifth embodiment, similarly to the fourth embodiment, the spectra of white light and infrared light may be set so that the main wavelength regions may be slightly overlapped.

F. Sixth Embodiment: FIG. 12 is a block diagram showing the arrangement of an inspection apparatus according to the sixth embodiment. The optical system 406 of this inspection apparatus has a lens system 440 (also called “imaging optical system”) commonly used for both infrared light and visible light, and this lens system 440 is one mirror. It is stored in the cylinder 441. The visible light emitted from the light source 410 is
The surface of the printed circuit board PCB is illuminated from an oblique direction, and the reflected light in the normal direction passes through the lens system 440 and the first C
An image is formed on the CD element 210. On the other hand, the infrared light emitted from the infrared light source 420 is reflected by the half mirror 480 (or half prism) to illuminate the surface of the printed circuit board PCB almost vertically, and the reflected light is reflected by the half mirror 480.
And passes through the lens system 440 to form an image on the second CCD element 212. However, visible light and infrared light are
Illuminate different locations on the CB. Note that, as described in the first embodiment, if the same elements are used as the two CCD elements 210 and 212, there is an advantage that the attachment mechanism and the adjusting method for them can be made common.

The optical system 406 further includes a light shielding plate 490 provided near the surface of the printed circuit board PCB. The light blocking plate 490 is for preventing visible light and infrared light from interfering with each other. Here, “the visible light and the infrared light do not interfere with each other” means that they do not enter the optical path of the other party as effective light. The light shielding plate 490 may be omitted if other measures are taken to prevent visible light and infrared light from interfering with each other. For example, when the visible light spot and the infrared light spot on the printed circuit board PCB are sufficiently distant from each other and the respective light fluxes are sufficiently narrowed, the light shielding plate 490 may be omitted. .

The processing procedure is the same as in the first embodiment shown in FIG.
The procedure of can be adopted. That is, at the time of image capturing, infrared light and visible light are simultaneously emitted, and two CCDs are used.
The infrared light image IRM and the color information image MG0 are simultaneously acquired using the elements 210 and 212. Also in the sixth embodiment, since the infrared light and the visible light are set so that the main wavelength ranges do not overlap each other as illustrated in FIG. 6,
It is possible to acquire two images at the same time.

In the inspection apparatus of the sixth embodiment, since the infrared light and the visible light commonly use the same lens system 440 housed in the same lens barrel 441, the fifth embodiment (Fig. 1
Compared with 1), there is an advantage that the number of parts is small. However, it is not necessary to commonly use all the lenses of the imaging optical system in the same lens barrel 441, and at least some of the lenses may be commonly used. However, the lens system 440
If all lenses of (imaging optics) are used in common,
It has the advantage of a simpler structure.

By the way, as described in the third embodiment (FIG. 9), the convergence angle 2θ of infrared light that irradiates the printed circuit board PCB substantially vertically is the convergence angle of visible light that obliquely irradiates the printed circuit board PCB. It is preferably set to be larger than 2θ. Also in this case, the optical system 406 can be configured so that infrared light and visible light commonly use all the lenses of the lens system 440. For example, by adjusting the distance between the two CCD elements 210, 212 from the lens system 440, the two CCD elements 210,
It is possible for both lights to be well imaged at both 212. Specifically, CCD for infrared light
The distance between the element 210 and the lens system 404 may be set to be larger than the distance between the CCD element 212 for visible light and the lens system 404.

G. Details of Area Separation Processing: FIG. 13 is an explanatory diagram showing the configuration of the host processor 100. The host processor 100 has an external storage device 50 that stores various data and computer programs.

The area dividing unit 102 includes a representative color setting unit 110.
The pre-processing unit 120, the composite distance calculation unit 130, the color region dividing unit 140, and the post-processing unit 150 have the functions. The function of each of these units is that the computer program stored in the external storage device 50 is stored in the host processor 10.
It is realized by executing 0. As will be understood from the description below, the composite distance calculation unit 130
It also has a function as an angle index value calculation unit and a distance index value calculation unit.

FIG. 14 is a flow chart showing the procedure of area division. In step S1, the area dividing unit 102
Acquires the color image (FIG. 4) of the printed circuit board PCB from the image capturing section 230. In addition, when performing the process after step S2 with respect to the image imaged beforehand, image data is read from the external storage device 50 in step S1.

In step S2, the user sets a plurality of representative colors using a pointing device such as a mouse while observing the color image displayed on the display unit of the host processor 100. At this time, the representative color setting unit 110
Displays a predetermined dialog box for the representative color setting process on the display unit of the host processor 100 to allow the user to set the representative color.

FIG. 15 is an explanatory diagram showing how representative colors are set. The user has four types of areas GR (G1 + G
2), GL, BR, WH names (for example, "resist area", "gold plated area", etc.) are entered in the dialog box on the screen, and sample points (stars) for obtaining representative colors of each area Is shown on the color image. At least one sample point is designated in each area. When a plurality of sample points are designated in the same area, the average color of those sample points is adopted as the representative color of the area.

The user further specifies whether each area should be merged with another area. In the example of FIG. 15, it is specified that the green region GR alone constitutes the first divided region DR1. Further, it is specified that the gold area GL and the brown area BR are merged to form the second divided area DR2, and the white area WH is independently formed to form the third divided area DR3. The representative color setting unit 110 has four areas GR, G.
The RGB color components of the representative colors of L, BR, and WH are acquired from the image data of the color image and registered. In general, n (n is an integer of 2 or more) representative colors are registered.

In step S3 (FIG. 14), the preprocessing unit 120 (FIG. 13) executes smoothing processing (blurring processing) on the color image to be processed. In the smoothing process, various smoothing filters such as a median filter, a Gaussian filter, and a moving average can be used. By performing this smoothing process, it is possible to remove the peculiar pixels existing in the image data, and thus it is possible to obtain image data with less dust (noise component). The pretreatment can be omitted.

In step S4, the composite distance calculation unit 130
Calculates a composite distance index value with a plurality of representative colors for the color of each pixel in the color image (referred to as “individual color”), and classifies each individual color into a representative color cluster. FIG.
3 is a flowchart showing a detailed procedure of step S4. In step S11, n (n is an integer of 2 or more) representative color vectors representing representative colors and individual color vectors representing individual colors of each pixel in the color image are standardized. The standardization of the representative color vector is performed according to the following equations (1a) to (1d).

[0077]

[Equation 1]

Here, Rref (i) is the i-th (i = 1 to n)
Is the R component of the representative color, and Gref (i) is the G component, Bre
f (i) is its B component. Also, Rvref (i), Gvref
(i) and Bvref (i) are standardized RGB components. (1
In the expression a), three components Rref (i), Gref (i), Bref (i)
The value Lref (i) used for standardization is obtained by the arithmetic sum of the above. In (1b) to (1c), each component is standardized using this standardized value Lref (i).

FIG. 17 is an explanatory diagram showing a color standardizing method according to the equations (1a) to (1d). Here, for convenience of illustration, a point representing a representative color (white circle) and a point representing an individual color (black circle) in a two-dimensional space composed of two color components of an R component and a B component are drawn. Above (1a)
The expression (1d) means that the representative color vector is standardized on the plane PL defined by R + G + B = 1. However, when the representative color is completely black (when Lref (i) = 0), the normalized components Rvref (i), Gvref (i), Bvref
The values of (i) are set to 1/3, respectively. This is (1
This is to prevent the right side of the expressions (b) to (1d) from becoming infinite.

The individual color vector of each pixel is also standardized in accordance with the following equations (2a) to (2d) like the representative color.

[0081]

[Equation 2]

Here, j is a number for identifying each pixel in the color image.

In FIG. 17B, the individual colors appear to be dispersed around the plane PL even after the normalization, but this is because the state of the three-dimensional space is observed two-dimensionally. Yes, in practice, all standardized individual colors also exist on the plane PL.

In step S12 of FIG. 16, the angle index value V (i, j) between the n representative color vectors and the individual color vector of each pixel is calculated according to the following equation (3a) or equation (3b). To be done.

[0085]

[Equation 3]

The first term in the parentheses on the right side of the equation (3a) is i
This is the absolute value of the difference between the normalized R component Rvref (i) of the th representative color and the normalized R component Rv (j) of the individual color of the jth pixel. The second and third terms are the corresponding G and B component values. Further, k1 is a predetermined coefficient that is not zero. Therefore, the right side of the expression (3a) has a strong correlation with the distance between the standardized representative color and the standardized individual color on the plane PL. The expression (3b) uses the square of the difference instead of the absolute value of the difference, and is an expression that gives the distance itself between the representative color after standardization and the individual color after standardization. is there. By the way, in general, the smaller the distance between the representative color and the individual color on the plane PL, the smaller the angle between the representative color vector and the individual color vector. Value V given by equation (3a) or (3b)
(I, j) is a value determined according to the distance between the representative color and the individual color on the plane PL, and has a strong correlation with the angle between the representative color vector and the individual color vector. Therefore,
In the present embodiment, the value V (i, j) given by the expression (3a) or the expression (3b) is used as an angle index value that substantially represents the angle between the representative color vector and the individual color vector. There is.

As can be understood from the expressions (3a) and (3b), the angle index value V (i, j) is a value that substantially represents the angle between the representative color vector and the individual color vector in the color space. However, other values given by equations other than the equations (3a) and (3b) can be used.

When the coefficient k1 is 1, the angle index value V (i, j) takes a value in the range of 0-2. This angle index value V (i, j) is the individual color vector of each pixel and n
It is calculated for all combinations with the individual representative color vectors.

In step S13, the distance index value D (i, i, n) between the n representative color vectors and the individual color vector of each pixel is calculated.
j) is calculated according to the following equation (4a) or equation (4b).

[0090]

[Equation 4]

The first term in the parentheses on the right side of the equation (4a) is i
It is the absolute value of the difference between the R component Rref (i) of the th representative color before normalization and the R component R (j) of the individual color of the jth pixel before normalization. The second and third terms are the corresponding G and B component values. Further, k2 is a predetermined coefficient that is not zero. Expression (4b) uses the square root of the sum of squares of the difference instead of the absolute value of the difference. (4a), (4
In the equation (b), unlike the above equations (3a) and (3b), the unstandardized values Rref (i) and R (j) are used. Therefore, the right side of the expressions (4a) and (4b) gives a value according to the distance between the non-standardized representative color and the individual color. Therefore, in the present embodiment, the value D (i, j) given by the equation (4a) or the equation (4b) is used as a distance index value that substantially represents the distance between the representative color and the individual color. There is.

As can be understood from the expressions (4a) and (4b), the distance index value D (i, j) should be a value that substantially represents the distance between the representative color and the individual color in the color space. However, it is also possible to use other values given by equations other than the equations (4a) and (4b).

When each color component is 8-bit data and the coefficient k2 is 1, the distance index value D (i,
j) takes a value in the range of 0 to 765. This distance index value D
(I, j) is also calculated for all combinations of the individual color vector of each pixel and the n representative color vectors.

In step S14, the compound distance index value C (i, j) for the i-th representative color and the individual color of the j-th pixel is calculated according to the following expression (5a) or (5b).

[0095]

[Equation 5]

In the equation (5a), the angle index value V (i, j)
And the sum of the distance index value D (i, j) is the composite distance index value C
It is adopted as (i, j). Further, in the equation (5b), the angle index value V (i, j) and the distance index value D (i, j)
The product of is adopted as the composite distance index value C (i, j). Therefore, the compound distance index value C (i, j) given by the equation (5a) or the equation (5b) has a small angle between the individual color vector of the jth pixel and the ith representative color vector, and The smaller the distance between the individual color and the representative color in the space, the smaller the value.

In this way, when the composite distance index value C (i, j) for a plurality of representative colors is calculated for the color of each pixel, the individual color of each pixel is changed to the composite distance index value C (i in step S15. , J) is classified into the cluster of the representative color having the smallest value. Here, the “cluster” means a group of colors associated with one representative color. Since n composite distance index values C (i, j) corresponding to n representative colors are obtained for each pixel, these n composite distance index values C (i, j) are obtained. The individual color of the pixel is classified into the representative color cluster that gives the minimum value in j).

FIG. 18 is classified into four types of representative color clusters.
The distribution of the individual colors that have been displayed is shown. I explained in Figure 15.
As described above, in step S2 (FIG. 14), the green region GR,
Of the gold area GL, the brown area BR, and the white area WH
Four representative colors corresponding to four types of color areas are set
It was Therefore, in FIG. 18, the individual color of each pixel is
Representative color cluster CL corresponding to these four representative colors_GR，
CL_GL, CL_BR, CL _WHIt is classified into. Where gold
Color cluster CL_GLThe darker of the individual colors (3
A color close to the origin O of the dimensional color space) is a brown cluster CL_BR
The distance to the relatively dark color inside is small. However, this example
Then, the composite distance index value C described above when classifying individual colors
Since (i, j) is used, the distance between the individual color and the representative color
Are close and the angle between them is small
Individual color of each pixel is classified into a representative color cluster
There is. Therefore, improper clustering can occur
Less likely to allow better clustering
Is.

Thus, the individual color of each pixel is the representative color class.
If it is classified as one of the
In addition, the color area dividing unit 140 determines that the representative color to which each pixel belongs.
The image area is divided according to the cluster. For example, each class
A pixel belonging to a star has a unique number different from other clusters.
No. (representative color number)
Split. Specifically, for example, each CL in FIG._GR，
CL_GL, CL_BR, CL _WHThe pixel values 0, 1, 2, 3
Assigned respectively. In the following, step S5
Area that is assigned the same representative color number in
This is called a "color area".

In step S6, the color area dividing unit 140
However, the representative color areas are merged as necessary. In the present embodiment, as described in FIG. 15, it is specified in step S2 that the gold area GL and the brown area BR are merged. Therefore, in step S5, these areas G
L and BR are merged as the second divided region DR2.

FIG. 19 is an explanatory diagram showing how the divided areas after merging are displayed on the display unit of the host processor 100. The first to third divided regions DR1 to DR3 are
Different colors or patterns are displayed according to the representative color numbers. Note that the correspondence between each representative color number and the color (display color) that fills each divided area during display is preset by the user. Alternatively, the color area dividing section 140 may automatically determine the relationship between each representative color number and the display color. As you can see from this example,
In this embodiment, first, the color image is classified into a plurality of representative color areas, and then, some representative color areas are merged as necessary. By using such a process, there is an advantage that a plurality of areas having different colors can be classified into the same type of divided areas according to the user's request.

When the image area of the color image is classified into a plurality of divided areas in this way, the post-processing unit 150 performs post-processing in step S7 of FIG. This post-processing is thinning processing (contraction processing) and thickening processing (expansion processing)
This is a noise removal process including and. In this noise removal processing, for pixels in a specific divided area to be processed, thinning processing with a predetermined pixel width is performed, and then thickening processing with the same pixel width is performed. Further, the thinning process and the thickening process can be similarly performed on the other divided areas. By performing such post-processing, a pinhole-shaped fine area (noise) is removed.

As described above, in the above embodiment, the distance index value that substantially represents the distance between the individual color of each pixel and the representative color and the angle between the individual color vector and the representative color vector are substantially equal to each other. Composite distance index value C based on the angle index value
(I, j) is calculated, and the individual color of each pixel is classified into the representative color area so that the value C (i, j) becomes the smallest. Therefore, it is possible to classify the colors of pixels, which are originally the same color but have greatly different brightness, into the same representative color region. As a result, it is possible to perform more appropriate area division as compared with the related art.

The above-described area separation processing is merely an example, and area separation processing other than this can be adopted.

H. Modifications: The present invention is not limited to the above-described embodiments and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are also possible. is there.

H1. Modification 1: In the above-described various embodiments, the color image including the three color components of RGB is acquired as the color information image, but the color information image only needs to include at least two color components. For example, RG
Two of the B components may be included, or two other color components may be included.

H2. Modified Example 2: In the various embodiments described above, white light and infrared light were used as the illumination light, but instead of white light and infrared light, two arbitrary lights having different main wavelength ranges are used. This may be adopted so that two types of images relating to those two lights are respectively acquired. The present invention can be applied to identify a region having a specific color on the surface of an inspection object based on such two types of images.

However, as the two types of images, a color information image used for performing area separation (area extraction) by color and a contrast in a specific area recognized by the area separation are used. It is preferable to acquire a contrast image (also referred to as a “multi-tone image”) used for further separating a specific region into a plurality of regions. At this time, it is preferable that the light used for acquiring the contrast image has a spectrum on the longer wavelength side than the light used for acquiring the color information image. The reason for this is that, as illustrated in FIG. 6, light on the long wavelength side tends to have a higher contrast due to the difference in members.

H3. Modification 3: In each of the various embodiments described above, the inspection is performed using the printed circuit board PCB as the inspection target, but the present invention can be applied to the surface inspection of any inspection target other than the printed circuit board. However, when the present invention is applied to the inspection of an inspection object in which a member having a metallic luster such as a printed circuit board exists on the surface, it has a great effect in recognizing a plurality of regions on the surface. In particular, the effect is remarkable for the inspection of the inspection object whose surface is plated with gold.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an inspection apparatus as a first embodiment of the present invention.

FIG. 2 is a graph showing spectra of white light and infrared light.

FIG. 3 is a flowchart showing an inspection processing procedure in the first embodiment.

FIG. 4 is an explanatory diagram showing a color information image of a printed circuit board PCB.

FIG. 5 is an explanatory diagram showing the processing contents of steps T2 to T6.

[FIG. 6] Gold area GL (gold plated area) and brown area BR
The graph which shows the spectral reflectance characteristic of (base area | region).

FIG. 7 is a block diagram showing the configuration of an inspection device according to a second embodiment.

FIG. 8 is a graph showing transmittance characteristics of a cube type dichroic mirror and a plate type dichroic mirror.

FIG. 9 is an explanatory diagram showing a configuration of an infrared light source 420 according to a third embodiment.

FIG. 10 is a flowchart showing an inspection processing procedure in the fourth embodiment.

FIG. 11 is a block diagram showing the configuration of an inspection device according to a fifth embodiment.

FIG. 12 is a block diagram showing the configuration of an inspection device according to a sixth embodiment.

FIG. 13 is an explanatory diagram showing a configuration of a host processor 100.

FIG. 14 is a flowchart showing a procedure of area division in the embodiment.

FIG. 15 is an explanatory diagram showing how representative colors are set.

16 is a flowchart showing a detailed procedure of step S4 of FIG.

FIG. 17 is an explanatory diagram showing a color standardizing method.

FIG. 18 is an explanatory diagram showing distribution of individual colors classified into four types of representative color clusters.

FIG. 19 is an explanatory diagram showing a plurality of divided areas.

[Explanation of symbols]

50 ... External storage device 100 ... Host processor 102 ... Area division unit 104 ... Area recognition unit 106 ... Reference image creation unit 108 ... Inspection execution unit 110 ... Representative color setting section 120 ... Pretreatment unit 130 ... Compound distance calculation unit 140 ... Color area dividing unit 150 ... Post-processing unit 200 ... Imaging unit 210, 212 ... CCD element 220 ... A / D converter 230 ... Image capturing unit 300 ... Drive mechanism 310 ... XY stage 320 ... Drive device 330 ... Drive control unit 400 ... Illumination optical system 402 ... First optical system 404 ... Second optical system 410 ... White light source 420 ... Infrared light source 422 ... Near infrared LED 424 ... Convex lens 430 ... Half prism 440 ... Lens system 442 ... Lens system 450 ... Dichroic prism 460 ... Dichroic mirror 470 ... Glass plate for correction 480 ... Half mirror 490 ... Shading plate

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI Theme Coat (reference) H05K 3/00 H05K 3/00 Q (72) Inventor Atsushi Imamura 4 Horikawa Toujinokami, Kamigyo-ku, Kyoto Chome Tenjin Kitamachi No. 1 No. 1 Dainippon Screen Manufacturing Co., Ltd. (72) Inventor Hiroshi Sano No. 4 Chome Tenjin Kitamachi No. 1 Dainippon Screen Manufacturing Co. Ltd. (72) ) Inventor Haruo Uemura 4-chome, Horikawa-dori Teranouchi, Kamigyo-ku, Kyoto City 1 Tenjin Kitamachi No. 1 Dai Nippon Screen Mfg. Co., Ltd. Tenjin Kitamachi No. 1 Dai Nippon Screen Manufacturing Co., Ltd. (72) Inventor Masahiko Kokubo 4-chome Tenjin, Horikawa-dori Teranouchi, Kamigyo-ku, Kyoto 1 in Japan 1 Dainippon Screen Manufacturing Co., Ltd. (72) Inventor Masahiro Horie 4-chome Tenjin Kitamachi, Kamio-ku, Kyoto City Tenjin Kitamachi 1 Dai Nippon Screen Manufacturing Co., Ltd. F term (reference) ) 2G020 AA08 DA05 DA22 DA23 DA31 DA34 DA35 DA52 2G051 AA65 AB11 AC21 BA01 BA06 BA08 BA20 BB01 BB07 CA03 CB01 CB05 CC12 EA11 EA17 5B047 AA12 AB04 BA01 BC07 BC01 FA01 FA01 DB01 FA01 CA02 DB02 CA06 CA02 CA06 CA02 CA06 CA02 DB02 CA06 CA02 CA06 CA02 CA06 CA02 DB02 CA06 CA02 CA06 DB02 CA02 CA02 HA07 JA03 JA09

Claims (18)

[Claims] 1. A method for inspecting the surface condition of an inspection object, comprising: (a) irradiating the surface of the inspection object with a predetermined first light from an oblique direction to obtain the first light. A first image component containing at least two color components contained in
And (b) irradiating the second light having a spectrum on the longer wavelength side than the first light from a direction substantially perpendicular to the surface of the inspection object to obtain the first image. Two
Acquiring a second image relating to the light of (c), and (c) a specific color region having a specific color on the surface of the inspection object based on the first image and the second image. And a step of identifying the surface of the object to be inspected. 2. The inspection method according to claim 1, wherein the step (c) includes, from the first image, the specific color region and another color having a color close to the specific color region. A step of identifying an integrated color area including an area, and a step of identifying the specific color area from the integrated color area by using a gradation value of the second image, The second light spectrum is set so that the contrast between the specific color region and the other color region is larger in the second image than in the first image. 3. The inspection method according to claim 1, wherein the inspection object is a printed circuit board for an electronic circuit, and the specific color area is a gold-plated area, and The light is infrared light, the inspection method. 4. The inspection method according to claim 1, wherein the image acquisition in the step (a) and the image acquisition in the step (b) are performed simultaneously, and the first light is white light. And the second light is infrared light. 5. The inspection method according to claim 1, wherein the image acquisition in the step (a) and the image acquisition in the step (b) are performed at different times. 6. The inspection method according to claim 1, wherein the second light is convergent light that converges into a relatively small light spot on the surface of the inspection object. Method. 7. The inspection method according to claim 1, wherein the image acquisition in the step (a) and the image acquisition in the step (b) are simultaneously performed, and the first light is visible light. The second light is infrared light whose main wavelength regions do not overlap with the visible light, and the visible light and the infrared light reflected from the inspection target are the same lens barrel. An inspection method in which at least some of the lenses in the image are commonly passed to form an image. 8. The inspection method according to claim 7, wherein the infrared light is convergent light that converges into a small light spot on the surface of the inspection object, and the convergence angle of the infrared light is the visible light. Inspection method that is set larger than the convergence angle of light. 9. An apparatus for inspecting a surface state of an inspection object, comprising: a first predetermined direction obliquely to the surface of the inspection object.
A first light source for irradiating the first light and a second light having a spectrum having a longer wavelength side than the first light from a direction substantially perpendicular to the surface of the inspection object. An illumination optical system including a second light source, a first image including an image component relating to at least two color components included in the first light, and a second image relating to the second light are acquired. And an area identification section that identifies a specific color area having a specific color on the surface of the inspection object based on the first image and the second image. A surface inspection device for an inspection object, which is characterized in that 10. The inspection apparatus according to claim 9, wherein the area identification unit includes, from the first image, the specific color area and another color area having a color close to the specific color area. And a function of identifying the integrated color area including, and a function of identifying the specific color area from the integrated color area by using the gradation value of the second image, The spectrum of the second light is set so that the contrast between the specific color area and the other color area is larger in the second image than in the first image. . 11. The inspection apparatus according to claim 9, wherein the inspection object is a printed circuit board for an electronic circuit, the specific color area is a gold-plated area, and The inspection device is infrared light. 12. The inspection apparatus according to claim 9, wherein the imaging unit includes a first imaging unit that acquires the first image, and an image captured by the first imaging unit. A second image capturing unit that captures the second image at the same time as, the first light is visible light, and the second light is infrared light. apparatus. 13. The inspection apparatus according to claim 9, wherein the imaging unit performs the same acquisition of the first image and acquisition of the second image at different times. An inspection apparatus that executes using an image sensor. 14. The inspection apparatus according to claim 9, wherein the second light is convergent light that converges into a relatively small light spot on the surface of the inspection object. apparatus. 15. The inspection apparatus according to claim 9, wherein the imaging unit includes an imaging optical system housed in a lens barrel and the imaging optical system passing through the imaging optical system. A first image pickup unit that receives the first light to obtain the first image, and receives the second light that has passed through the imaging optical system at the same time as the image pickup by the first image pickup unit. And a second image pickup unit that picks up the second image, the first light is visible light, and the second light has a main wavelength range different from that of the visible light. Infrared lights that do not overlap with each other, the visible light and the infrared light reflected from the inspection object are commonly passed through at least some lenses of the imaging optical system housed in the lens barrel. And an image is formed on each of the first and second imaging units. 16. The inspection apparatus according to claim 12, wherein the infrared light is convergent light that converges into a small light spot on the surface of the inspection object, and the convergence angle of the infrared light is the visible light. Inspection device that is set larger than the convergence angle of light. 17. The inspection apparatus according to claim 9, wherein the illumination optical system converts the light reflected by the surface into the first light and the second light. An inspection apparatus having a plate type dichroic mirror for separating. 18. The inspection apparatus according to claim 9, wherein the imaging unit includes a first imaging element for acquiring the first image and the second image. A second image sensor for capturing an image, wherein the first and second image sensors have the same characteristics.