JP7443162B2 - Inspection equipment and inspection method - Google Patents

Description

The present invention relates to an inspection device and an inspection method for inspecting the surface of an object.

Evaluating the appearance of an object is an important proposition, and in recent years, automation has been promoted in the inspection of the finish of automobiles and other industrial products. In particular, an apparatus is known that illuminates an object to be inspected with line pattern illumination (hereinafter referred to as line pattern illumination) that changes in brightness and darkness at a certain spatial period to inspect defects such as irregularities and scratches on the object. With technology that uses line pattern illumination to inspect defects, it is particularly difficult to detect defects such as irregularities on objects that are not completely mirror-like and have diffuse reflective properties, such as processed metal surfaces or resin molded surfaces. Therefore, the following technology has been disclosed in which the object to be inspected is illuminated with pattern light at a relatively gentle angle (large incident angle) and observed.

Japanese Unexamined Patent Publication No. 2011-89981 (Patent Document 1) discloses an inspection device for unevenness defects in a steel plate. In this inspection device, the evaluation area of the steel plate is illuminated at an angle of 3 to 10 degrees with line pattern illumination light from the illumination unit, and the lines reflected on the steel plate are captured using a camera placed in the specular reflection direction with respect to the illumination unit. Photograph pattern illumination light. At this time, defects in the unevenness of the steel plate are detected by moving a holding member to which both the illumination unit and the camera are fixed and capturing an image while changing the phase of the line pattern with respect to the surface of the steel plate. In addition, the positional relationship between the illumination unit and the steel plate is such that they do not overlap on the captured image, so that in the captured image, there are areas where the line pattern illumination light is reflected and areas where the line pattern illumination light is not reflected, with no change in light intensity. The defect evaluation area is determined by automatically recognizing the defective parts.

JP2011-89981A

When measuring a rough surface such as a metal, it is better to capture the image at a shallower angle (larger reflection angle) to capture the bright and dark patterned light stripes more clearly. Although Patent Document 1 generates and acquires an image that does not include an illumination portion on a captured image, it does not disclose an inspection when an illumination portion is included on a captured image. Further, in Patent Document 1, when the illumination section and the steel plate overlap on the captured image, it is difficult to automatically recognize a monotonous portion where the amount of light does not change.

Therefore, an object of the present invention is to easily inspect the evaluation region of an object when a captured image includes an illumination part.

An inspection apparatus according to one aspect of the present invention for solving the above problems is an inspection apparatus that inspects the surface of an object, and includes an illumination section that irradiates a bright and dark pattern onto the surface of the object, and an illumination section that illuminates the bright and dark pattern. an image capturing section that captures an image of the surface of the object, and a processing section that evaluates the surface of the object by processing the image captured by the image capturing section; The processing unit generates a plurality of images in which the surface of the object illuminated while shifting the phase of the light and dark pattern and the illumination unit are included within an angle of view, and the processing unit generates an image based on the plurality of images. The method is characterized in that the surface of the object and the area of the illumination section within the object are separated, and an image on the separated surface of the object is evaluated.

According to the present invention, it is possible to easily inspect the evaluation region of an object when a captured image includes an illumination section.

It is a schematic diagram showing an inspection device in a 1st embodiment. FIG. 2 is a schematic diagram showing the arrangement of an illumination unit and an evaluation surface of an object. It is a figure showing BRDF of an inspection object. It is a flow chart of an inspection method. FIG. 3 is a diagram showing an image acquired by a camera. 6 is a diagram showing a phase change in the y direction in the image of FIG. 5. FIG. It is a figure which shows the image acquired by the camera when the illumination part is tilted. 8 is a diagram showing a phase change in the y direction in the image of FIG. 7. FIG. It is a schematic diagram showing an inspection device in a 2nd embodiment. It is a schematic diagram showing an inspection device in a 3rd embodiment. It is a schematic diagram showing an inspection device in a 4th embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail based on the accompanying drawings.

<First embodiment>
FIG. 1 is a schematic diagram showing an inspection device 50 in the first embodiment. The inspection device 50 includes an illumination section 1, a camera 2 (imaging section), a control section 3, an image processing section 5, a display section 6, and a movable mechanism 7 in order to inspect an object 4 to be inspected.

The illumination unit 1 has a highly diffused light distribution characteristic, and irradiates the object 4 with bright and dark line-shaped pattern illumination light by surface illumination consisting of a bright part and a dark part. The illumination unit 1 will be explained using FIG. 2. The illumination part 1 is a surface illumination in which bright parts (light transmitting parts) 1A and dark parts (non-transmitting parts, light shielding parts) 1B are spatially arranged alternately on a plane at a pitch P. For example, in uniformly bright surface illumination obtained by placing a diffuser plate on top of the LEDs lined inside, a non-transparent part can be silk-printed with black ink or a plate material such as black metal can be placed. It can be configured by

The camera 2 is composed of a two-dimensional area sensor such as a CCD or CMOS, and a lens, and is capable of acquiring an image of the evaluation surface (inspection surface). The camera 2 photographs the reflected light of the illumination light irradiated onto the evaluation surfaces 41 and 42 by the illumination unit 1.

The control section 3 controls each section of the inspection device 50. The control section 3 is constituted by a board having, for example, a CPU and a memory, and controls the illumination section 1, the camera 2, and the movable mechanism 7 in synchronization.

The image processing unit 5 has a function of processing images, and may be, for example, a general-purpose PC or a machine dedicated to image processing, and includes a processing device such as a CPU, GPU, or DSP. The image captured by the camera 2 is transferred to the image processing section 5 via a cable and the control section 3. The image processing unit 5 performs image processing on the image generated by the camera 2, performs various numerical processing to determine defects on the object surface, and can determine the presence or absence of defects and pass/fail. Specifically, the image processing unit 5 can quantify the size and depth of defects and scratches on the evaluation surfaces 41 and 42, how conspicuous they are, and make a pass/fail determination.

The display unit 6 is, for example, a display, and can display the acquired image, the result of numerical processing by the image processing unit 5, the pass/fail result of a defect determined using the result of numerical processing, and the like.

The surface of the object 4 to be inspected includes, for example, a metal processed surface, a resin molded surface, or a light diffusing surface such as a printed surface. In this embodiment, the evaluation surfaces 41 and 42 are metal processed surfaces processed and manufactured using a milling cutter, and will be described using, as an example, a light diffusing surface in which a score is formed by machining on the surface during processing. As shown in FIG. 3(a), when the incident light 100 faces directly, that is, when the incident angle θ is a small value of 45° or less, the metal processed surface has a BRDF characteristic 105B (BRDF = bidirectional reflectance distribution) close to diffusion. function). On the other hand, as shown in FIG. 3(b), when the incident light 100 on a metal processed surface is at a gentle angle such that it follows the reflective surface, that is, when the incident angle θ is between 80° and 90°, most of the light is reflected in the specular direction. It has a sharp BRDF characteristic of 105C that reflects light.

Therefore, the illumination is made such that the evaluation surfaces 41 and 42 and the direction of reflected light of the illumination light on the evaluation surfaces 41 and 42 photographed by the camera 2 (optical axis of the camera 2) are at an angle of 10 degrees or less. A unit 1, an object 4, and a camera 2 are arranged. With such arrangement, a mirror image of the pattern illumination light from the illumination section 1 is projected onto the evaluation surfaces 41 and 42, and can be imaged by the camera 2.

It is preferable that the focal position and aperture of the lens of the camera 2 are adjusted so that the entire evaluation surface 41 and the evaluation surface 42 are in focus and can be observed clearly. In this embodiment, assuming that the length of the evaluation surfaces 41 and 42 is 20 mm and the angle of incidence θ is 85 degrees, in order to focus on the entire surface of the evaluation surfaces 41 and 42, the depth of focus is 20 mm x sin (85 degrees )=19.9mm is required. On the other hand, a defect with a width of 1 mm on the evaluation surfaces 41 and 42 becomes 1 mm×cos(85°)=0.087 mm when the defect is viewed from the direction of the camera 2. In order to identify this defect, it is preferable that the blur tolerance of the optical system is smaller than this. In other words, in this application where only about 50 μm of blur is allowed over 19.9 mm in the focal direction of the lens of the camera 2, and defects are observed using macro photography, it is necessary to reduce the lens aperture by a considerable number. Therefore, in order to balance inspection time and imaging time, we created a Scheimpflug configuration in which the lens optical axis and sensor surface are tilted so that the focal plane is tilted to the optical axis, and the lens aperture is opened as much as possible to reach the sensor. It is also effective to secure sufficient light. However, it does not matter if the image is out of focus, depending on the size of the flaw to be determined.

The illumination part 1 is moved by the movable mechanism 7 in a direction (X direction in the figure, periodic direction of pitch P) orthogonal to the direction in which the lines of the bright part 1A and dark part 1B of the illumination part 1 extend (longitudinal direction). It is configured as possible. The movable mechanism 7 is connected to the control section 3. The control unit 3 controls the movable mechanism 7 to move the illumination unit 1 by ΔXi (i=1, 2, . . . N, where N is an integer of 3 or more) and causes the camera 2 to take N images. . Here, ΔX1, ΔX2, . However, the configuration is not limited to this, and for example, the movable mechanism 7 may be manually operated to move the object 4, and then the object 4 may be photographed by the camera 2 using a manual trigger.

Next, a method for detecting (inspecting) defects will be described. FIG. 4 shows a flowchart of a method for inspecting defects on the surface of an object using the inspection apparatus of this embodiment.

First, the control unit 3 controls the movable mechanism 7 to move the illumination unit 1 and move the relative position of the illumination unit 1 and the object 4 by ΔX1 with respect to the reference position (S11). Next, at the position of ΔX1, the control unit 3 controls the illumination unit 1 to emit light, and the camera 2 takes a picture to generate the first image I1 (x, y) (S12). Here, x and y represent the position of a pixel on the image. Note that ΔX1 may be zero and the first image may be placed at the reference position. Next, it is determined whether i is equal to the set N (S13). If i is not N, i is incremented by 1 (S14) and the process returns to S11.

Further, the control unit 3 controls the movable mechanism 7 to move the illumination unit 1 and change the relative position of the illumination unit 1 and the object 4 by ΔX2 with respect to the reference position (S11). Next, at the position of ΔX2, the control unit 3 controls the illumination unit 1 to emit light, and the camera 2 takes a picture to generate a second image I2 (x, y) (S12). Steps S11 to S14 are repeated N times until i becomes N, and the camera 2 generates a total of N images, which the image processing unit 5 acquires. When i becomes N, the process advances to S15.

Next, the image processing unit 5 uses the information regarding the intensity change of the frequency component whose phase shifts by 4πΔXi/P radians when the relative position between the illumination unit 1 and the object 4 changes by ΔXi to convert the N images into , generates a first composite image (S15). An example of a composite image is an amplitude image of a frequency component whose phase shifts by 4πΔX1/P radians (if the inspection target is a flat surface, a frequency component corresponding to a striped pattern with a period of P/2 that occurs on the image). It is. When the relative positions of the illumination unit 1 and the object 4 are shifted in steps of P/N width, ΔXi (i=1, 2, . . . N) is expressed by the following equation 1.

ΔXi=(P/N)×(i-1)...(1)
Equation 1 includes the case where ΔX1 is zero, and when the first image has changed from the reference position, the following Equation 2 is obtained.

ΔXi=(P/N)×i...(2)
At this time, the amplitude image A(x, y) can be calculated using Equation 3 below.

これが、Ｎ枚の画像を処理して得られる、物体４の表面（評価面）の情報を含む処理画像であって、位相が４πΔＸｉ／Ｐラジアンでシフトする周波数成分の強度変化に関する情報を用いて生成された処理画像である。

This is a processed image containing information on the surface (evaluation surface) of the object 4, obtained by processing N images, and using information on intensity changes of frequency components whose phase shifts by 4πΔXi/P radians. This is the generated processed image.

When the position of the illumination section 1 is moved, a bright point (light transmission section) corresponding to a light emitting point of the illumination section 1 and a dark point (non-transmission section) corresponding to a dark section of the illumination section 1 move. Therefore, the illumination light from the illumination unit 1 reflected by the evaluation surface of the inspection object changes the brightness and darkness of the intensity at one point on the pixel of the camera 2.

In this embodiment, the angles θ41 and θ42 at which the evaluation surfaces 41 and 42 of the object 4 are viewed from the camera 2 side are set to 5° or less. Therefore, even if the evaluation surfaces 41 and 42 are rough surfaces such as processed metal surfaces, they have gloss, and the bright and dark images of the illumination are clearly reflected on the evaluation surfaces 41 and 42, resulting in a high S/N ratio. An amplitude corresponding to the difference between brightness and darkness occurs. Another example of the composite image is a phase image of frequency components whose phase shifts by 4πΔXi/P radians. The phase image θ(x, y) can be calculated using Equation 4 below.

一般に、位相は‐π～πの値で算出されるため、それ以上に位相が変化する場合は、位相画像では非連続な位相の飛びが発生する。このため、必要に応じて位相接続（位相アンラップ）が必要である。位相画像では、表面性状として評価面４１と評価面４２の表面の傾きを評価することができ、可視化することもできる。

Generally, the phase is calculated as a value between -π and π, so if the phase changes more than that, discontinuous phase jumps will occur in the phase image. Therefore, phase unwrapping (phase unwrapping) is required as necessary. In the phase image, the inclination of the surfaces of the evaluation surface 41 and the evaluation surface 42 can be evaluated as a surface property, and can also be visualized.

Various algorithms have been proposed for phase unwrapping (phase unwrapping), but errors may occur if image noise is large. As a means to avoid phase coupling, a phase difference corresponding to a phase differential may be calculated. The phase differences Δθx(x, y) and Δθy(x, y) can be calculated using Equation 5 below.

さらなる合成画像の例は平均画像である。平均画像Ｉａｖｅ（ｘ，ｙ）は、以下の式６により算出できる。

A further example of a composite image is an average image. The average image Iave (x, y) can be calculated using Equation 6 below.

平均画像では、表面性状として反射率の分布を評価できる。したがって、平均画像では、汚れや錆びなど、正常な部分と反射率に違いがある欠陥の情報を得ることができる。可視化することもできる。このように、振幅画像、位相画像または位相差画像、平均画像で、光学的に評価可能な表面性状が異なる結果、可視化される欠陥も異なるため、これらの画像を組み合わせることで、多様な表面性状を評価して、多様な欠陥を可視化することができる。

In the average image, the distribution of reflectance can be evaluated as a surface quality. Therefore, the average image can provide information on defects such as dirt and rust that have a different reflectance from normal parts. It can also be visualized. In this way, the surface textures that can be optically evaluated are different in the amplitude image, phase image, phase contrast image, and average image, and as a result, the defects that are visualized are also different. By combining these images, it is possible to visualize various surface textures. can be evaluated to visualize various defects.

In this embodiment, in order to measure a surface treated as a diffusion surface, the image acquired by the camera 2 includes not only the evaluation surfaces 41 and 42 of the object 4 but also the illumination section 1. A lighting section 1, a camera 2, and an object 4 are arranged. In the image obtained by moving the illumination part 1 described above, the bright part 1A and the dark part 1B of the illumination part 1 are shown moving by P/N width in N images, so the illumination part 1 is An amplitude image, a phase image or a phase difference image, and an average image can also be obtained for each part. Therefore, during evaluation (inspection), it is necessary to clearly separate the illumination unit 1 and the evaluation surfaces 41 and 42 of the object 4. The separation method will be explained below.

The angle of incidence of the illumination light on the evaluation surfaces 41 and 42 of the object 4 is from 80° to 90°. For example, if the material of the object 4 is an aluminum alloy, the reflectance due to Fresnel reflection of the material is approximately in the range of 90% to 95%. Furthermore, since the object 4 is a machined surface, it has a somewhat diffuse reflective property. Considering the spread of BRDF, it can be said that the reflectance of the object to be inspected in the specular reflection direction is 85% to 90% or less.

Utilizing the difference in the amount of light received by the camera 2 due to the reflectance in the regular reflection direction, and using the maximum brightness value of the same pixel in the N images, the area of the illumination part 1 in the image and the object 4 are determined. (the area of the illumination light reflected by the object 4). For example, the threshold value is set to a brightness value of 95, assuming that the maximum brightness value in N images of a pixel at a location known as the illumination unit 1, for example, the top center of the image, is 100 due to layout. On the other hand, the luminance value of the pixel in the portion of the illumination light reflected by the object decreases depending on the reflectance of the object 4, and becomes 90 or less at the maximum. Therefore, the image processing unit 5 can determine an area where the brightness value (pixel value) of the image is less than or equal to the threshold value as the evaluation area of the object 4, and use the area for defect detection. In this way, the area of the illumination unit 1 and the area of the object 4 to be evaluated can be separated in the image using the pixel brightness value and the threshold value.

Furthermore, when the amplitude image A(x, y) is used, evaluation is made based on the difference between the maximum brightness value and the minimum brightness value in the N acquired images, that is, the brightness and contrast of the bright part 1A and the dark part 1B of the illumination part 1. It turns out. In the region of the illumination light reflected by the object 4, the brightness value of the image decreases due to the reflectance and reflection properties of the evaluation surface of the object 4. In other words, as in the example of the maximum brightness value, by setting a threshold value for the brightness value of each pixel of the amplitude image A (x, y) in consideration of the decrease due to reflectance and reflection properties, the illumination part 1 The area and the evaluation area of the object 4 can be separated.

In addition, for the unevaluable portion where the illumination light of the illumination unit 1 reflected by the object 4 is not reflected, the amplitude image A (x, y ) is less than a predetermined threshold value (for example, 10) can be separated as an unevaluable portion.

Next, a case of determining based on phase will be explained. As shown in FIG. 2, when determining based on the phase, the line direction of the illumination unit 1 and the normal direction of the evaluation surfaces 41 and 42 of the object 4 are arranged so as not to be parallel when viewed from the direction of the camera 2. This is effective. Specifically, when viewed from the direction of the camera 2, the normal direction 41F of the evaluation surface 41 and the normal direction 42F of the evaluation surface 42 are inclined with respect to the line direction 1F of the bright section 1A and the dark section 1B of the illumination section 1. Suppose that it is located as follows. Then, in the image acquired by the camera 2, as shown in FIG. 5, the directions 41FF and 42FF of the bright and dark line pattern light of the illumination part 1 reflected on the evaluation surfaces 41 and 42 are the line directions of the illumination part 1 reflected as they are. The direction is slanted from 1F. Here, if the direction of the line direction 1F of the illumination unit 1 is arranged in the y direction (lateral direction) of the screen in the image of FIG. It can be seen that these are the boundaries between the evaluation surfaces 41 and 42 and the reflected portion of the illumination unit 1.

When comparing the phase θ(x, y) in the y direction for each pixel arranged in the x direction in the image, as shown in Figure 6, there is a phase jump where the phase changes beyond the S/N value of the image. occurs. The pixel area with phase jump is the boundary of the evaluation area of object 4, the pixel area above it where the phase does not change is the illumination reflection area, and the pixels located below the boundary are the area where the object to be evaluated is reflected. It can be separated from the evaluation section. Further, depending on the clarity of the image, it is possible that the boundary portion spans a plurality of pixels, but in that case, the plurality of pixels may be used as the boundary portion and separated from the portion that does not require evaluation.

Similarly, it is also possible to discriminate based on phase difference. When determining based on phase difference, the line direction 1F of the bright portion 1A and dark portion 1B of the illumination unit 1 does not need to be parallel to the y direction, which is the pixel arrangement direction. As shown in FIG. 7, the line direction 1F may be tilted with respect to the x direction or the y direction, which is the pixel arrangement direction. In this case, the phase difference Δθx(x, y) in the y direction for each pixel arranged in the x direction in the image is such that a phase jump occurs at the boundary between the illumination reflection part and the evaluation part, and then , resulting in a waveform in which the slope of the stripes changes. Therefore, the illumination reflection part and the evaluation part can be separated by the phase jump and the phase difference changing part.

As described above, the ability to separate the illumination reflection part and the evaluation part by the phase or phase difference is an effect that occurs precisely because the light and dark line direction of the illumination and the normal direction of the evaluation surface are tilted when viewed from the camera direction. This effect only needs to be able to distinguish the boundary between the reflected lighting and the object, so a part of the evaluation surface, for example, the evaluation surface inside the boundary, can be used even if the line direction of the illumination part and the normal direction of the evaluation surface are the same. good. In other words, when viewed from the camera 2 side, the direction in which the lines of the bright and dark pattern of the illumination unit 1 extend and the direction of the normal to at least a part of the surface of the object 4 need only be different.

Furthermore, in the present embodiment, the illumination reflection portion and the evaluation portion are distinguished from each other using the boundary portion as a boundary. However, depending on the shape of the object 4, there may be chamfers or corners near the boundary of the evaluation surface, or if the object 4 itself is not a simple plane such as a curved surface, the object 4 may be placed so that it is observed almost parallel to the evaluation surface. It is possible that this will happen. In that case, the accuracy of defect evaluation and determination deteriorates. In that case, after determining the boundary line, an area starting from a pixel that is a predetermined pixel distance from the boundary line is set as an evaluation area. Alternatively, setting an area from pixels corresponding to a location a predetermined distance away from the boundary line as an evaluation area is also an effective means for increasing evaluation determination accuracy.

Further, in this embodiment, when the distance of the illumination unit 1 is increased from the camera 2 and the object 4, the pitch of the illumination light of the bright and dark pattern reflected on the evaluation surfaces 41 and 42 becomes smaller, and the sensitivity of evaluation to defects increases. Conversely, when the distance of the illumination unit 1 is brought closer to the camera 2 and the object 4, the pitch of the bright and dark illumination reflected on the evaluation surfaces 41 and 42 increases, and the sensitivity of evaluation to defects decreases. However, as the pitch of the projected striped pattern changes, the sensitivity of cracks during processing, which are unrelated to evaluation, changes and becomes visible as noise components. Therefore, it is also effective to adjust the position of the illumination unit 1 so that each sensitivity becomes a sensitivity suitable for evaluation and determination, depending on the size and depth of the defect. In this case, the illumination position may be changed and adjusted depending on the type of defect to be evaluated, such as a minute scratch or a large concave defect.

Furthermore, a method has been described in which the determination (separation) of evaluation regions is performed through arithmetic processing by the image processing unit 5. However, it is also possible to simply display data and graphs of changes in each value involved in determining the evaluation area on the display unit 6. For example, the display data may include data on the maximum luminance value of each pixel of N images or images thereof, amplitude image A(x,y), phase image θ(x,y), phase difference image Δθx(x, y) and Δθy(x, y). Alternatively, data and graphs of changes in each value in the x direction of each pixel arranged in the y direction may be cited. In this case, the evaluation area may be determined by the user using the GUI based on the data displayed on the display section 6 or the data output from the image processing section 5.

<Second embodiment>
FIG. 9 is a schematic diagram showing an inspection device according to the second embodiment. In this embodiment, the object 4 has the shape of a gear (comb teeth), and the evaluation surfaces 41 and 42 to be inspected are two surfaces of different teeth facing in different directions. In the inspection of such a complex-shaped object, this configuration in which the evaluation surfaces 41 and 42 and the line pattern illumination are placed within the same field of view of the camera 2 is suitable, and the evaluation area and the area of the line pattern illumination that is the background are can also be easily separated by the method described in the first embodiment.

<Third embodiment>
FIG. 10 is a schematic diagram showing an inspection device according to the third embodiment. Similar to the first embodiment, the evaluation surfaces 41 and 42 of the object 4 are illuminated by the illumination section 1 consisting of a bright section and a dark section, and the camera 2 is placed in the specular reflection direction of the evaluation surfaces 41 and 42 with respect to the illumination section 1, The position of the illumination unit 1 is moved by a predetermined pitch amount by the driving unit 7.

Line pattern illumination is reflected on the evaluation surfaces 41 and 42, and a plurality of images whose phases are shifted are acquired, and the image processing unit 5 performs the arithmetic processing described in the first embodiment to form the evaluation surface. 41 and 42 defects were detected.

The present embodiment differs from the first embodiment in that, in addition to this, a transflective illumination unit 101 is disposed between the camera 2 and the object 4, and is movable by a drive unit 701. Details of the translucent line pattern illumination are disclosed in Japanese Patent Application Laid-Open No. 2019-2762. Translucent line pattern lighting is a special type of lighting that looks like a semi-transparent screen when viewed from the camera 2 side, and line pattern lighting when viewed from the object 4 side, and has a wide normal angle range. It is effective for simultaneous evaluation of both aspects.

When the illumination part 1 is driven, the translucent illumination part 101 is simultaneously driven by the driving part 701 according to a command from the control part 3, and the driving amount is preferably the same pitch amount as that of the driving of the illumination part 1.

According to this configuration, the illumination light of the translucent illumination unit 101 also hits the surface 43 of the object 4, which was not illuminated with the illumination light of the illumination unit 1 and was regarded as an unevaluable surface in the first embodiment, so that defects can be evaluated. It becomes like this. However, since the reflected light on the surface 43 is illumination light from an angle close to directly opposite, the surface 43 must be a high-gloss surface that reflects the line pattern illumination of the illumination 101 even from this angle. However, it is now possible to inspect multiple surfaces with various orientations at the same time, which has the effect of speeding up the inspection of parts for flaws and defects. For the surface 43 as well, calculation of the evaluation value for defect evaluation and determination of the evaluation area can be performed using the same method.

<Fourth embodiment>
FIG. 11 is a schematic diagram showing an inspection device according to the fourth embodiment. Unlike the second embodiment, the object 4 is sandwiched between translucent illumination units 101 and 102, and is placed in a position where the evaluation surfaces 42, 43, and 45 of the object 4 can be observed from the camera 2. 41, 44, and 46 are placed at a position where they can be observed. The translucent illumination units 101 and 102 are moved by P/N by drive mechanisms 701 and 702, respectively, to obtain N images each, for a total of 2N images. The images obtained by the camera 2 and the camera 201 are subjected to the arithmetic processing described in the first embodiment by the image processing unit 5, respectively, to simultaneously eliminate defects on the evaluation surfaces 41, 42, 43, 44, 45, and 46. Can be detected. In this embodiment, two translucent line pattern illuminations are used, and the measurement is performed by sandwiching the measurement from two directions. However, the line pattern illumination is not limited to this, depending on the size and shape of the part, it may be three or more, or it may be a combination of translucent light pattern illumination or normal line pattern illumination. Demonstrates the effect of

In the inspection apparatus of the above embodiment, the front surface of the illumination is The measurement sample placed in the area is illuminated. In addition, by photographing the reflected light from the object with a camera that can acquire a two-dimensional image placed in the direction of specular reflection on the measurement sample while shifting the spatial phase of the bright and dark areas of the illumination, it is possible to A plurality of images including both the illumination and the object within the angle of view are acquired. Then, using the brightness information in each pixel of the obtained image and the evaluation value calculated from the brightness information, the illumination that is the background and the object area are separated, and defects in the object are detected. This makes it possible to inspect scratches and defects on rough surfaces such as metal surfaces and objects with complex shapes. Specifically, the evaluation area is calculated using the maximum brightness value of the same pixel in each of the N images obtained, and the amplitude image obtained by taking the difference between the maximum and minimum brightness values of the same pixel in each of the N images. Able to make decisions and evaluate defects. Alternatively, the evaluation area can be determined by the phase when the change in each pixel of each of the N images is simulated as a sine wave, or by the phase difference that digitizes the phase change of each pixel in the pixel arrangement direction. Defects can be evaluated.

Although preferred embodiments and examples of the present invention have been described above, the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the gist thereof. For example, the present invention can be applied to an external appearance inspection device that uses a time-correlated image sensor as the camera 2 and moves the illumination unit 1 while exposing the camera to take pictures.

Claims (8)

In an inspection device that inspects the surface of an object,
an illumination unit that irradiates the surface of the object with a bright and dark pattern;
an imaging unit that images the surface of the object illuminated with the bright and dark pattern by the illumination unit;
a processing unit that evaluates the surface of the object by processing the image captured by the imaging unit,
The imaging unit generates a plurality of images in which the surface of the object illuminated by the illumination unit while shifting the phase of the brightness and darkness pattern and the illumination unit are included in an angle of view,
The inspection device is characterized in that the processing section separates the surface of the object and the region of the illumination section in the image based on the plurality of images, and evaluates the separated image on the surface of the object. 2. The object according to claim 1, wherein the direction in which the line of the light and dark pattern of the illumination section extends and the direction of a normal to at least a part of the surface of the object are different from each other when viewed from the side of the imaging section. Inspection equipment. 2. The object according to claim 1, wherein in the image, the direction in which the lines of the brightness pattern of the illumination section extend and the direction in which the lines of the brightness and darkness pattern reflected on at least a part of the surface of the object extend are different. Inspection equipment as described. The inspection device according to any one of claims 1 to 3, wherein the processing unit determines an evaluation area on the surface of the object by calculation based on pixel values of the plurality of images. The inspection device according to claim 4, wherein the processing unit outputs information regarding the determined evaluation area. The inspection device according to claim 5, further comprising a display section that displays information regarding the evaluation area. 7. The inspection apparatus according to claim 1, wherein the processing unit detects a defect on the surface of the object based on an evaluation value of an image on the surface of the separated object. In an inspection method for inspecting the surface of an object,
an illumination step of irradiating the surface of the object with a bright and dark pattern using an illumination unit;
an imaging step of imaging the surface of the object illuminated with the bright and dark pattern;
an evaluation step of evaluating the surface of the object by processing the captured image,
In the imaging step, generating a plurality of images in which the surface of the object illuminated while shifting the phase of the brightness and darkness pattern and the illumination section are included within an angle of view,
An inspection method characterized in that, in the evaluation step, the surface of the object and the area of the illumination section in the image are separated based on the plurality of images, and the image on the separated surface of the object is evaluated.

Images