JP2008051664A - Method and apparatus for inspecting vane shape of impeller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vane shape inspection method and apparatus for impellers capable of inspecting in a short time, whether or not each of a large number of vanes provided on the periphery of an impeller has a shape intended in designing without an overlook. <P>SOLUTION: The method is provided with a process for imaging the front part of the impeller from the direction of the rotation axis of the impeller to be put on its installation place, a process for performing binarizing processing of a pick-up image relating to the front part imaged from the direction of the rotation axis of the impeller by the imaging process, and acquiring its binarized image, a process for detecting the positions of the tip-corresponding bright areas of all vane-corresponding bright areas provided around an impeller-corresponding bright area on the basis of the binarized image, a process for calculating the positional relation between each tip-corresponding bright area and a predetermined reference area in the impeller-corresponding bright area, concerning all the detected tip-corresponding bright areas, and a process for judging whether or not a vane shape of the impeller is conforming by comparing the calculated positional relation with a predetermined specified value. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

The present invention relates to an impeller blade shape inspection method and inspection apparatus, and more particularly to an impeller blade shape inspection method and inspection apparatus including a plurality of blades around a shaft member.

Conventionally, in the manufacturing process of an impeller having a plurality of blades (for example, a turbine impeller, a turbine wheel of an automobile turbo engine), if an operator accidentally drops the impeller or hits the impellers, the impeller May be deformed or damaged. In particular, the turbine wheel has a complicated shape and the tip of the blade is often thin. Such thin portions are easily deformed during the manufacturing process.
Therefore, it is inspected whether or not each of the plurality of blades in the impeller has a design-scheduled shape (whether there is no deformation or damage). When performing the inspection, the inspection is usually performed visually.

However, there are individual differences among the inspectors in the visual inspection, and some people may miss a slight deformation or scratch. On the other hand, some people make false detections, and there is a problem that inspection accuracy in visual inspection is low.
Furthermore, there is a problem that it takes time to inspect all of the plurality of blades in the impeller.

  Therefore, in the prior art, the surface shape of each of the surface side and the back surface side of the turbine blade (hereinafter also referred to as “blade”) is measured using a displacement meter, and the surface shape of the blade is synthesized by combining the shapes of the front and back surfaces. The technique of the measuring method is disclosed (for example, refer patent document 1).

JP 2002-174512 A

  However, when inspecting the shape of a large number of blades provided around the impeller using this technique, a single blade is scanned over the entire area, and the three-dimensional shape of the single blade is measured. The measurement has a problem that it takes a very long time for the inspection.

The purpose of the present application is to check the impeller blade shape of the impeller, which can inspect whether or not a large number of blades provided around the impeller have a design-scheduled shape without "missing". A blade shape inspection method and an inspection apparatus are to be provided.
Another object of the present invention is to provide an impeller blade shape inspection method and inspection apparatus that can inspect the tip of the impeller in a short time.
Other objects and advantages will be readily apparent from the drawings and the following description associated therewith.

  The impeller blade shape inspection method according to the present invention includes a step of imaging the front face of the impeller 2 from the direction of the rotational axis of the impeller 2 placed at the installation location 10 and the rotation of the impeller 2 imaged in the imaging step. A process of binarizing a captured image related to the front of the impeller 2 from the axial direction to obtain a binarized image 31, and a periphery of the impeller corresponding bright part 32 based on the binarized image 31. The step of detecting the positions of the tip-corresponding light portions 35, 35... 35 of all the blade-corresponding light portions 34, 34... 34, and all the detected tip-corresponding light portions 35, 35,. For 35, the step of calculating the positional relationship between each of the leading end corresponding light portions 35, 35... 35 and the predetermined reference portion in the impeller corresponding light portion 32, and the calculated positional relationship and a predetermined rule And a step of comparing the values to determine the quality of the blade shape of the impeller.

  Further, preferably, the place 10 where the impeller 2 is installed and the image pickup means 15 which picks up the impeller 2 placed at the installation place 10 are arranged in the direction of the rotation axis of the impeller 2. And an image pickup means 15 arranged so that the front of the impeller 2 can be imaged, and further, a picked-up image related to the front of the impeller 2 from the rotational axis direction of the impeller 2 imaged by the image pickup means 15 Binarization processing means 26 for binarizing, and all blade corresponding light portions 34, 34 provided around the impeller corresponding light portion 32 based on the binarized image 31 obtained by the binarization processing means 26.・ ・ ・ 34 tip-corresponding light portions 35, 35... 35, and tip-portion detecting means 27 for detecting the position of the tip portion, and all tip-corresponding light portions 35, 35, detected by the tip portion detecting means 27. .. About 35, it is determined in advance at each tip corresponding light part 35 and impeller corresponding light part 32. A calculation means 28 for calculating a positional relationship with a reference portion to be determined, and a determination for determining the quality of the impeller blade shape by comparing the positional relationship calculated by the calculation means 28 with a predetermined specified value. Any image processing means 25 having the means 29 may be used.

    It is also preferable to image the impeller 2 that is rotatably installed at the installation location 10, in which the entire area of the free end surface 7 of each blade 4 of the impeller 2 is imaged, and the imaging is performed in the imaging process. The captured images relating to the entire area of the free end face 7 in each of the blades 4, 4... 4 of the respective impellers 2 are binarized and binarized images relating to the entire area of the free end faces 7 of all the blades 4. 31 and a step of comparing one binarized image with another binarized image and determining whether the shape of the impeller is good or bad for all the binarized images 31 described above. I just need it.

    Also preferably, a place 10 where the impeller 2 that is rotatably supported is installed, and an imaging means 15 that images the impeller 2 placed at the installation place 10, wherein the arrangement of the imaging means 15 is the impeller The imaging means 15 arranged so as to be capable of imaging the entire area of the free end surface 7 of the two blades 4 and the imaging means 15 enable imaging of the free end surfaces 7 of all the blades 4 provided around the impeller 2. And a rotating means 50 for rotating the impeller 2 installed at the place 10 where the impeller 2 is installed, and based on each of the captured images 30 captured by the imaging means 15. , Binarization processing means 54 for binarizing the captured image relating to the entire area of the free end face 7 in each blade 4, and all the binarized images 31 obtained by the binarization processing means 54, one binary Binary image and other binary By comparing the image, as long as it includes an image processing unit 53 and a determination unit 55 that it is so as to determine the quality of the blade shape of the impeller.

  As described above, in the present invention, when inspecting the blade shape of the impeller, the tip portions of all the many blades provided around the impeller 2 are picked up, and all the tip portion correspondences are based on the picked-up images. For the parts 35, 35 ... 35, the positional relationship with a predetermined reference part is calculated, and the quality of the blade shape is judged, so that each blade provided around the impeller has a shape that is planned by design. There is an effect that it is possible to inspect whether or not it is within the allowable range.

Furthermore, the present invention is characterized in that the tip-corresponding bright portion relating to all the blades can be inspected according to the same criterion, and the quality does not vary.
This has an effect on quality that can provide an impeller having a good balance when the impeller rotates at high speed.

  Moreover, even if the present invention inspects all the many blades provided around the impeller, when performing the above inspection, the tip of all the many blades provided around the impeller 2 is imaged. Since only the bright portion corresponding to the tip of the blade is to be inspected, there is a feature that the amount of arithmetic processing is small and the time required for the inspection is shortened. This has the operational effect of being able to inspect very quickly even if there are many blades around the impeller.

  Embodiments of the present invention will be described below with reference to the drawings. 1 to 14, reference numeral 2 denotes an impeller to be inspected, for example, a known turbine wheel. In the impeller 2, 3 indicates a shaft member, and 4 indicates a plurality of blades provided around the shaft member 3.

  Next, the inspection apparatus 1 for inspecting the quality of the blade shape of the impeller 2 will be described (see FIG. 1). Reference numeral 11 denotes a base in the impeller blade shape inspection apparatus 1, 12 denotes a support member connected to the base 11, and 10 denotes an installation place for mounting the impeller provided at the tip of the support member 12. . Reference numeral 13 denotes a support means for supporting the imaging means 15 and the illumination means 20 described later.

Reference numeral 15 denotes image pickup means for picking up the impeller 2 placed at the installation place 10. The image pickup means 15 is arranged such that the front of the impeller 2 can be picked up from the direction of the rotational axis 9 of the impeller 2. It is. In particular, a large number of blades 4 provided around the impeller 2 can be clearly imaged at the same scale.
For example, a well-known CCD camera is used as the imaging unit 15. As is well known, the CCD camera 15 is installed on the support means 13 via an elevating mechanism (not shown) and is installed so as to be movable up and down. The distance H1 between the CCD camera 15 and the front surface of the impeller 2 may be set in consideration of the imaging range of the CCD camera 15, and is set to H1: 150 mm, for example.
Next, a captured image 30 by the imaging unit 15 will be described (see FIG. 2). Reference numerals 2a, 3a, and 4a denote corresponding portions of the impeller 2, the shaft member 3, and the blade 4, respectively, 5 is a front end portion of the blade corresponding portion 4a, 6 is a front end portion of the front end portion 5, and 7 is a free end of the blade 4. Each end face is shown. Reference numeral 8 denotes the front face of the impeller 2, 9 denotes a rotation axis, and R denotes a mounting location for fixing the rotation shaft in the next stage process for a product that has passed inspection. 2A relates to a healthy product, and the captured image 30 shown in FIG. 2B relates to a defective product, and 5p denotes a deformed tip.

Reference numeral 20 shown in FIG. 1 denotes illumination means for irradiating the impeller 2 placed at the installation place 10.
As the illumination means 20, for example, a well-known annular LED illumination device may be used so as to avoid a line connecting the CCD camera 15 and the impeller 2. According to this LED illuminating device, a large number of blades 4 provided around the impeller 2 can be irradiated with the same definition so as to be imaged.
Further, in place of the annular LED lighting device, in consideration of image processing (binarization processing) after imaging, two or more pieces are applied so that the front of the impeller 2 is irradiated from two directions (preferably four directions). A lighting device may be installed. The installation positions of the illuminating means 20 and 20 are respectively set so that the light beam can be emitted when the angle θ1 is about 45 ° to 80 ° with respect to the front surface of the impeller 2 (see FIG. 1). Reference numeral 21 denotes the presence of light rays by the illumination means 20.

Next, 25 shown in the upper right of FIG. 1A indicates an image processing means connected to the image pickup means 15, and includes a binarization processing means 26, a tip end detection means 27, a calculation means 28, and a determination means 29, which will be described later. Prepare. The image processing means 25 performs the image processing (S20 to S50) shown in FIG.
The binarization processing unit 26 can binarize the captured image of the front surface of the impeller 2 from the direction of the rotational axis 9 of the impeller 2 captured by the imaging unit 15. In the binarization process (S20), depending on the state of the captured image 30, a known enhancement process, differentiation process, or the like may be selectively performed as necessary. Moreover, it is good also as what masks (removes) an unnecessary range. By doing so, it is possible to perform binarization processing that is more stable (reproducibility).
The binarized image 31 binarized by the binarization processing means 26 will be described (see FIG. 3).
Reference numeral 32 denotes a bright portion corresponding to the front side surface of the impeller 2a (hereinafter also referred to as “impeller corresponding bright portion”), and reference numeral 34 denotes a bright portion corresponding to the front side surface of the blade 4a (hereinafter referred to as “blade corresponding bright portion”). Part "). Reference numeral 33 denotes a bright portion corresponding to the shaft member 3a, and reference numeral 35 denotes a bright portion corresponding to the distal end portion 5 (hereinafter also referred to as “front end corresponding bright portion”). Reference numeral 36 denotes a bright portion corresponding to the tip 6 (hereinafter also referred to as “tip-corresponding bright portion”). C shows the dark part corresponding to the rotating shaft attachment part R. FIG. 35p appearing in FIG. 3 (B) indicates a bright portion corresponding to the deformed tip portion 5p appearing in FIG. 2 (B).

The tip end detection means 27 is based on the binarized image 31 obtained by the binarization processing means 26, and the tip ends of all the blade corresponding bright portions 34, 34... 34 provided around the impeller corresponding bright portion 32. The position of the corresponding bright part 35 is detected. Specifically, the tip portion detecting means 27 detects the position of the tip portion corresponding bright portion 35 as follows.
The tip-corresponding bright part 35 is scanned in the circumferential direction around the axis 39 of the impeller corresponding bright part 32 of the binarized image 31. Since the dark part 41 and the bright part 35 can be detected repeatedly, the position from the dark part 41 to the bright part 35 is detected as the position of the bright part 35 corresponding to the tip part.
In addition, when detecting the position of the bright part 35 corresponding to the tip part, the example in which the position from the dark part 41 to the bright part 35 is detected as the position of the bright part 35 corresponding to the tip part has been described. Alternatively, the position where the dark portion 41 is changed to the tip portion corresponding bright portion 35 may be detected.

  The computing means 28 calculates the positional relationship between each of the front end corresponding light portions 35 and the predetermined reference portion in the impeller corresponding light portion 32 for all the front end corresponding light portions 35 detected by the front end portion detecting means 27. I have to do it. Specifically, the process (S120) shown in FIG. 4 is performed. In S120, the “distance between adjacent tip-corresponding bright portions 35” (hereinafter also referred to as “distance between each other”) L is calculated. For example, in the case of the adjacent front end corresponding bright portions 35a and 35b (see FIG. 4B), the edge 38a and the front end corresponding bright portion 35b are set with the edge 38a of the front end corresponding bright portion 35a as a predetermined reference portion. The distance L1 between the edges 38b is calculated. Similarly, the distances L2 to L11 are calculated for all the tip-corresponding bright portions 35a to 35k.

The determination unit 29 compares the positional relationship calculated by the calculation unit 28 with a predetermined specified value to determine whether the impeller blade shape is good or bad. Specifically, the processing (S130 to S140) shown in FIG. 4 is performed.
In S130, for the distances L1 to L11 calculated in S120, the difference D between the different distances L is calculated. For example, the difference D1 between the distances L1 and L2 is calculated. Similarly, the difference between the remaining distances L1 to L11 and the difference D2 to D11 is calculated.
Next, in S140, each of the differences D (D1 to D11) is compared with a predetermined specified value to determine whether or not it is equal to or less than the specified value. If the difference D is less than the specified value, it is determined that the blade shape of the impeller is good, and if the difference D is greater than the specified value, it is determined that the blade shape of the impeller is defective (with deformation). To do. Thus, a defective product can be detected only by the value of the difference D, regardless of the value of a healthy product.
In the above case, only the processing between the tip-corresponding bright portions 35 is sufficient, and the processing can be performed very easily in a short time.
In addition, it can provide an impeller with a favorable balance at the time of high-speed rotation of an impeller by test | inspecting that there is no irregularity about the front-end | tip of all the blades.
In the above inspection, the example in which the determination in S140 is determined by comparing each of the differences D (D1 to D11) with a predetermined specified value, but one difference D, for example, the difference D1 is defined as the specified value, You may make it determine by comparing with other difference D2-D11.
Furthermore, although the difference D of the distance L between mutually different distances L1 to L11 is calculated, the example in which each difference D (D1 to D11) is compared with a predetermined specified value has been described. For all the distances L between each other, the difference D between the different distances L is calculated (S130), and the distance L between each other and a predetermined specified value (for example, a design-scheduled value) ) And may be determined.

An inspection method for determining the quality of the blade shape of the impeller using the above-described configuration will be described with reference to the flowcharts shown in FIGS. 1 (B) and 4 (A).
First, the impeller 2 is installed at the installation place 10.
Next, the impeller 2 placed at the installation place 10 is irradiated by the illumination means 20, and the front face of the impeller 2 is imaged by the imaging means 15 from the direction of the rotational axis 9 of the impeller 2 (S10).
As described above, the captured image 30 as shown in FIG. 2A is generated. Data of the captured image 30 is sent to the image processing means 25.
Next, the binarization processing unit 26 of the image processing unit 25 receives the data of the captured image 30 and performs binarization processing of the captured image on the front surface of the impeller 2 from the rotational axis direction of the impeller 2. Perform (S20). By this image processing, a binarized image 31 as shown in FIG.

  Next, the tip portion detection means 27 receives the data of the binarized image 31 and performs processing for detecting the tip portion position (S30). Specifically, as described with reference to FIG. 4 described above, the process of step S110 is performed. By this process, the position of the leading edge corresponding bright part 35 is detected.

Next, in the calculation means 28, the detected position data of all the tip-corresponding bright parts 35 are received, and the positions of the respective reference-proposed bright parts 35 and the predetermined reference part in the impeller-corresponding bright part 32 are determined. Processing for calculating the relationship is performed (S40). Specifically, as described with reference to FIG. 4 described above, the process of step S120 is performed. By this processing, the distances L1 to L11 are calculated.
Subsequently, the determination means 29 performs a process of comparing the calculated positional relationship with a predetermined specified value to determine the quality of the impeller shape (S50). Specifically, as described with reference to FIG. 4 described above, the processes of steps S130 to S140 are performed.
As a result, the difference D is determined to be equal to or less than the specified value, and it is determined that “the impeller blade shape is good”.

Next, the case of an impeller as shown in FIG. 2B, that is, an impeller in which the tip portion 5p of some of the blades 4a among the plurality of blades is deformed will be described. The processing of S10 to S40 is performed as described above. Subsequently, in S50, the process of S130 is performed. By this processing, the distances L1 to L11 are calculated. Since the tip-corresponding bright portion 35b is deformed, the distance L1 between the distances L1 to L11 is shorter than the distances L3 to L11 between the other distances as can be understood from FIG. The distance L2 between them is longer than the distances L3 to L11 between the other.
Subsequently, the process of step S140 is performed.
As a result, as can be understood from FIG. 4C, the differences D1, D2, and D11 calculated based on the distances L1, L2 having a length different from the distances L3 to L11 between each other are specified values. Further, it is judged as “different”, it is judged that “the impeller blade shape is defective (with deformation)”, and this impeller is processed as a defective product.

Next, FIG. 5 showing an example different from the processing related to the image processing means in FIG. 4 in the processing in the calculation means and the processing in the determination means will be described.
The configuration according to FIG. 5 is the same as described above, that is, as described with reference to FIGS. 1, 2, and 3, the impeller 2 is installed at the installation location 10 and the processes of S <b> 10 to S <b> 30 are performed. Next, an example in which the processing of S40 and S50, that is, the processing of S220 to S280 described later is performed will be shown.
In FIG. 5, parts that are considered to have the same or equivalent configuration in function, property, characteristics, or the like as those in FIGS. 3 and 4 are denoted by the same reference numerals as those in FIGS. 3 and 4. Description is omitted. (Also, in the following figures, the same reference numerals are used for the same concept, and the same reference numerals as those in FIGS.
1A calculates the coordinates of the center of gravity position B and the coordinates of the minimum value M based on the binarized image 31, and the coordinates of the minimum value M and the coordinates of the center of gravity position B are calculated. A straight line S1 passing through the front end portion, and for the front end corresponding bright portion 35 detected by the front end portion detecting means 27, the midpoint coordinates of the front end in each front end corresponding bright portion 35 are calculated, and the midpoint coordinates and the above described A straight line S2 passing through the coordinates of the center of gravity position is calculated, and an angle θ2 formed by the straight line S1 passing through the coordinates of the minimum value M and the coordinates of the center of gravity position B, and the straight line S2 passing through the coordinates of the center point coordinates and the center of gravity position B is formed. Can be calculated.

Specifically, the calculation means 28 is configured to perform the processing (S220 to S270) shown in FIG. In S220, first, a coordinate system is set, and the coordinates of the center of gravity position B (hereinafter also referred to as “center of gravity coordinates”) are calculated based on the dark part C corresponding to the rotating shaft attachment part R. Next, in S230, the minimum value M existing in the recess 32a located between the tip corresponding bright parts 35 of all the blade corresponding bright parts 34, 34... (Also called “minimal coordinates”). The minimum value M may be extracted as follows. The binarized image 31 is scanned for light and dark in the circumferential direction around the center of gravity B. In this scanning process, a minimum value is always set at every angle α (an angle obtained by dividing 360 ° by the number of blades, for example, the angle α is about 32.8 ° when the number of blades in the illustrated impeller is 11). Since there is one M, it is preferable to extract based on the angle α.
In S240, a straight line S1 passing through the center-of-gravity coordinates and the minimum coordinates calculated in S220 and S230, respectively, is calculated. In S250, the coordinates of the midpoint (hereinafter referred to as “midpoint coordinates”) 36a of the tip corresponding bright portion 36 in each tip corresponding bright portion 35 detected by the tip detection means 27 in S210 are calculated.
In S260, a straight line S2 is calculated that passes through the center-of-gravity coordinates calculated in S220 and S250 and the coordinates of the leading edge corresponding bright part 35, respectively.
In S270, an angle θ2 formed by a straight line S1 passing through the minimum coordinates and the center of gravity coordinates, and a straight line S2 passing through the midpoint coordinates 36a and the coordinates of the center of gravity is calculated. The combination of the straight lines S1 and S2 for calculating the angle θ2 is the straight line S2 located on the counterclockwise direction side of the straight line S1, and is the closest S2 (see FIG. 5B). Of course, the angle θ2 may be calculated in combination with the straight line S2 positioned on the clockwise side of the predetermined straight line S1 and positioned closest to the straight line S1.

Next, the determination means 29 in FIG. 1A compares the angle θ2 calculated by the calculation means 28 with a predetermined specified value to determine the quality of the impeller shape. Specifically, the process of S280 is performed.
In S280, the angle θ2 is compared with a predetermined specified value, for example, an angle planned in design, and it is determined whether the angle is within the specified value range. When the angle θ2 is within the predetermined value range, it is determined that “the impeller blade shape is good”, and when the angle θ2 is not within the predetermined value range, “the impeller blade shape is defective (deformation). Yes) ”. For example, in the example shown in FIG. 5C, the angle θ2 is larger than the angle θ2 (θ2 indicated by a two-point difference line) in the case of a blade without deformation. As a result, in S280, the angle θ2 is determined to be out of the predetermined range, and it is determined that “the impeller blade shape is defective (with deformation)”.
In the above case, the minimum value M can be easily calculated, and the constituent member is a heavy (thick) portion, and thus has a feature that it is not easily deformed. This makes it suitable as a reference part.
In addition, although the example in which the midpoint coordinates of the tip corresponding bright portion 36 are calculated and the straight line S2 is calculated in S250 and S260 has been described, the coordinates of the edge 38 of the tip corresponding bright portion 36 (endpoint coordinates) are used instead of the midpoint coordinates. ) 38a may be calculated to calculate the straight line S2 (S241 to S251 shown in FIG. 5D).

Next, FIG. 6 will be described which shows an example different from the processing related to the image processing means in FIG. 4 in terms of the processing in the calculation means and the processing in the determination means.
The configuration according to FIG. 6 is the same as described above, that is, as described with reference to FIGS. 1, 2, and 3, the impeller 2 is installed at the installation location 10 and the processes of S <b> 10 to S <b> 30 are performed. Next, an example in which the processing of S40 and S50, that is, the processing of S320 to S350 described later is performed will be shown.
Based on the binarized image 31, the calculation means 28 exists in the concave portion located between the coordinates of the center of gravity position B and the tip-corresponding bright portions 35 of all the blade-corresponding bright portions 34, 34. The coordinates of the minimum value M are calculated, and a straight line S1 passing through the coordinates of the respective minimum values M and the coordinates of the barycentric position B is calculated.
Specifically, the calculation means 28 performs the processing (S320 to S340) shown in FIG. The processing of S320 to S340 may be performed in the same manner as S220 to S240 in FIG.

Next, the determining means 29 is positioned on both sides of the symmetry axis S1 with respect to each of the leading edge corresponding bright parts 35 detected by the leading edge detecting means 27, with the straight line S1 calculated by the calculating means 28 as the symmetry axis. The pattern matching processing of the two leading edge corresponding bright portions 35, 35 is performed, respectively, to determine whether the shape of the impeller is good or bad (S350).
As a result of the processing of S350, two tip-corresponding bright portions 35 and 35 (for example, tip-corresponding bright portions 35a and 35b shown in FIG. 6B) located on both sides of all the symmetry axes S1 are matched (completely coincident). (Or match more than a certain percentage), it is judged that there is a correlation and it is judged that the blade shape of the impeller is good. If it does not match, the blade shape of the impeller is defective (with deformation). Is determined. For example, in the example shown in FIG. 6C, the tip-corresponding bright part 35b and the tip-corresponding bright part 35c do not match and it is determined that there is no correlation. It is determined that there is deformation).

Next, FIG. 6D illustrating an example different from the processing related to the image processing unit in FIG. 6A in terms of the processing in the calculation unit and the processing in the determination unit will be described.
The computing means 28 calculates coordinates for the tip corresponding bright part 35 detected by the tip detection means 27 and, based on the binarized image 31, coordinates of the center of gravity position and all blade corresponding bright parts 34. , 34... 34, the coordinates of the minimum value M existing in the recesses located between the bright portions 35 corresponding to the tip portions are calculated, and a straight line passing through the coordinates of the respective minimum values M and the coordinates of the gravity center position B. S1 is calculated (S320 to S340, S370).
Next, the determining means 29 uses the straight line S1 passing through the coordinates of the minimum value M calculated by the calculating means 28 and the coordinates of the center of gravity position B as the symmetry axis, and the corresponding tip ends are located on both sides of the symmetry axis S1. The coordinates of the part 35 are compared to determine whether or not the shape of the impeller is good (S380). As a result of the processing of S380, if the coordinates of the two leading edge corresponding bright portions 35, 35 located on both sides of all the symmetry axes S1 are coincident (completely coincident or coincident at a predetermined ratio or more), the “blade shape of the impeller” Is determined to be “good”, and if they do not match, it is determined that “the impeller blade shape is defective (with deformation)”.
In addition, regarding the straight line S1 passing through the coordinates of the minimum value M and the coordinates of the barycentric position B according to the description with reference to FIGS. 5 (A) to (D) and FIGS. Alternatively, the straight line S1 passing through the coordinates of the other position and the gravity center position B may be calculated. For example, as shown in FIG. 6E, for a shape in which the minimum value M is difficult to calculate, the straight line S1 may be calculated as follows.
First, the center of gravity position B is detected, and an arc 43 is created around the center of gravity position B. The arc 43 only needs to pass through a thick (thick) portion that does not easily deform in the blade corresponding light portion 34. For example, the arc 43 passes through a portion near the base portion 34b of the blade corresponding light portion 34 as shown in the drawing. It is good to do so.
Next, the midpoint (point where distance W1 = W2 in the arc portion 43a) p is calculated in the arc portion 43a located between the tip-corresponding bright portions 35 of all the blade corresponding bright portions 34, 34. (See FIG. 6F). A straight line S1 passing through the calculated middle point p and the gravity center position B is calculated.

Next, FIGS. 5E and 5F showing an example different from the processing related to the image processing unit in FIG. 4 in the processing in the calculation unit and the processing in the determination unit will be described.
FIGS. 5E and 5F provide a method for determining the quality of the impeller blade shape based on the area of the tip-corresponding light portion 35.
First, the computing means 28 detects the center of gravity position B (S820) and creates an arc 43 passing through the center of gravity position B (S830). Next, the midpoint of the arc portion 43b of each of the tip corresponding bright portions 35, 35... 35 in all the blade corresponding bright portions 34, 34... 34 (the point where the distance d1 = d2 in the arc portion 43b). q is calculated (S840). Then, a straight line S3 passing through the midpoint q and the gravity center position B is calculated (S850). Next, in each of the leading end corresponding bright portions 35, 35... 35, two areas surrounded by the straight line S3, the arc portion 43b, and the contour of the leading end corresponding bright portion 35 are calculated as area A1 and area A2. S860).
Next, in the determination means 29, the area A1 and the area A2 calculated by the calculation means 28 are compared to determine whether or not the shape of the impeller is good (S870).
As described above, since the method shown in FIGS. 5E and 5F is based on the midpoint q and the gravity center position B in the thick part that is difficult to deform, the straight line S3 is close to a healthy product. Therefore, a defective product can be identified with high accuracy.

  As described above, the arithmetic processing and the determination processing according to FIGS. 4, 5, and 6 may be performed independently to inspect the shape of the impeller blades, but may be combined and inspected as necessary. It may be. For example, you may perform a test | inspection combining all the arithmetic processing and determination processing which concern on FIG.4, FIG.5, FIG.6. Further, the inspection may be performed by combining the arithmetic processing and the determination processing according to FIGS. 4 and 6 or the combination of the arithmetic processing and the determination processing according to FIGS. By performing the inspection in combination as described above, the detection accuracy of the defective product is improved.

Next, FIG. 7 showing an example different from the impeller in FIGS. 1 to 4 in terms of the position to be inspected will be described. FIG. 7 provides a method for inspecting the surface opposite to the inspection target position of the impeller in FIGS. 1 to 4 (the lower surface of the impeller 2 in FIG. 1) as the inspection target. 7A is a diagram for explaining a captured image 30 related to a healthy product, FIG. 7B is a diagram for explaining a captured image 30 related to a defective product, and FIG. 7C is an image of FIG. A diagram for explaining a binarized image 31 that has been binarized, (D) a diagram for explaining a binarized image 31 that has been binarized from the image of (B), and (E) an image process. Schematic for explaining the processing in the means, (F) shows a schematic diagram for explaining the processing in the image processing means when the impeller is defective.
The configuration and the inspection method of the inspection apparatus 1 according to FIG. 7 are the same as those described with reference to FIGS.
As described above, the configuration and the inspection method of the inspection apparatus 1 according to FIG. 7 are the same as the configuration and the inspection method of the inspection apparatus 1 according to FIGS. can do.

Next, FIGS. 8 to 11 showing examples different from the inspection apparatus and the inspection method of FIGS. 1 to 4 in the configuration of the inspection apparatus, the inspection target position of the impeller, and the processing in the image processing means will be described.
In the inspection apparatus 1, reference numeral 10 denotes a place where the impeller 2 that is rotatably supported is installed.
Reference numeral 11 denotes a base, which can be adjusted to an arbitrary angle θ4. A known mechanism may be used as the angle adjusting means of the base 11. The installation angle θ4 of the base 11 is an angle at which the outline of the impeller is easily extracted in consideration of image processing (binarization processing) after imaging the impeller (irradiation to the free end surface 7 from the illumination means 20). For example, the angle θ4: 0 to 30 ° may be set. 52 indicates the presence of an adjusting mechanism for adjusting the angle of the base 11.
50 rotates the impeller 2 installed at the place 10 where the impeller 2 is installed so that the imaging means 15 can image the free end surfaces 7 of all the blades 4 provided around the impeller 2. The rotation means made to move is shown. As the rotation means 50, for example, a known servo motor is used. As the rotating means 50, any other arbitrary motor may be used. In that case, the rotation angle of the motor corresponds to the number of divisions of the blades 4 in the impeller, and the rotation may be continuous or intermittent. It is preferable that the imaging unit 15 is synchronized with the imaging. Reference numeral 51 denotes a known photoelectric sensor that can determine the initial imaging position of an impeller that is optionally installed at the installation location 10.
In addition, in order to synchronize imaging and rotation, it is possible to cope with control using ON / OFF of the photoelectric sensor 51 instead of the pulse number control of the motor.

Next, 15 is an image pickup means for picking up the impeller 2 placed at the installation place 10, and the image pickup means 15 is arranged so that the entire free end surface 7 of all the blades 4 of the impeller 2 can be picked up. It is arranged.
Next, a captured image 30 by the imaging unit 15 will be described (see FIG. 9). Reference numerals 2a, 3a, 4a, and 7a denote corresponding portions of the impeller 2, the shaft member 3, the blade 4, and the free end surface 7, respectively. 9A relates normally, and the captured image 30 shown in FIGS. 9B and 9C relates to a defective product, and 7p denotes a deformed free end face corresponding portion.

  Reference numeral 20 that appears in FIG. 8A denotes illumination means for irradiating the impeller 2 placed at the installation place 10. The installation position of the illuminating means 20 is set so that light can be emitted when the angle θ5 is about 20 ° to 85 °.

Next, reference numeral 53 denotes an image processing unit connected to the image pickup unit 15 and includes a binarization unit 54 and a determination unit 55 described later. The image processing means 53 can perform the image processing (S60 to S80) shown in FIG.
The binarization processing unit 54 can binarize the captured image relating to the entire area of the free end surface 7 of each blade 4 based on each of the captured images 30 captured by the imaging unit 15.
The binarized image 31 binarized by the binarization processing means 26 will be described (see FIG. 10).
Reference numeral 37 denotes a bright portion corresponding to the free end surface corresponding portion 7a (hereinafter also referred to as “free end surface corresponding bright portion”). Reference numeral 40 denotes an example of a characteristic portion (a portion in which the deformation appears characteristically when the blade is deformed) in the free end face corresponding light portion 37. Reference numeral 41 denotes a dark part. 37p appearing in FIGS. 10B and 10C indicates a bright portion corresponding to the deformed tip portion 7p appearing in FIGS. 9B and 9C.

The determination unit 29 compares one binarized image with another binarized image for all the binarized images 31 obtained by the binarization processing unit 54, and determines whether the impeller blade shape is good or bad. Can be determined. Specifically, the process (S420) shown in FIG. 11 is performed.
In S420, any one binarized image, for example, binarized by first imaging and binarizing the binarized image 31 relating to the entire area of the free end surface 7 in all the blades 4 obtained in S410. An image is selected, and difference processing with other binarized images is performed using the selected image as a template. And based on the characteristic part 40 in the free end surface corresponding light part 37, it can be determined whether the blade shape of the impeller is good or bad. As a result of the processing of S420, in all the binarized images, when the detection of the characteristic portion in the free end surface corresponding bright part 37 is equal to or less than the predetermined value, it is determined that the blade shape of the impeller is good, and the binarized image If there is a binarized image in which the detection of the characteristic part in the free end face corresponding bright part 37 is determined to exceed a predetermined value, it is determined that “the impeller blade shape is defective (with deformation)” is there.
In the case of performing the above inspection, since the free end surfaces of all the many blades provided around the impeller 2 are imaged and only the free end surface corresponding bright portions 37 in the blades are inspected, the amount of determination processing is small. The feature is that the time required for inspection is shortened. This means that even if there are many blades around the impeller, the inspection can be performed very quickly.

An inspection method for determining the quality of the blade shape of the impeller using the configuration shown in FIG. 8 will be described with reference to the flowchart shown in FIG.
First, the impeller 2 is installed at the installation place 10.
Next, the impeller 2 placed at the installation place 10 is irradiated with the illumination means 20, and the entire area of the free end face 7 of the blade 4 of the impeller 2 is imaged (S60). As described above, the captured image 30 as shown in FIG. 9A is generated. The data of the captured image 30 is sent to the image processing means 53.
Next, the binarization processing unit 54 of the image processing unit 53 receives the data of the captured image 30 and binarizes the captured image related to the entire free end surface 7 of the blade 4 of the impeller 2 (S70). . By this image processing, a binary image 31 as shown in FIG. 10A is generated.
Subsequently, the rotation means 50 rotates at an angle corresponding to the number of divisions of the blades 4 in the impeller (an angle at which the free end surface 7 of the adjacent blade comes to the imaging position), and the blade adjacent to the previously imaged blade is illuminated. 20 and image the entire area of the free end face 7 of the blade 4 (S60). The captured image 30 is generated for the blades as described above, and the data is sent to the image processing unit 53. In the binarization processing means 54, the captured image relating to the entire area of the free end surface 7 of the blade 4 of the impeller 2 is binarized, and the binarized image 31 is generated (S70). Subsequently, the processing of S60 and S70 is repeated for all the blades 4 provided around the impeller 2 while rotating the rotating means 50.
Next, in the determination means 29, the process which determines the quality of the blade shape of an impeller is performed (S80). Specifically, the process of S420 is performed as described above. As a result, as described above, the quality of the blade shape of the impeller is determined.

Next, FIG. 12 will be described which shows an example different from the processing related to the image processing means of FIG.
The configuration according to FIG. 12 is the same as described above, that is, as described with reference to FIGS. 8, 9, and 10, the impeller 2 is installed at the installation location 10, and the processing of S60 and S70 is performed. Subsequently, an example in which the processing of S80, that is, the processing of S520 to S530 is performed will be shown.
The determination means 55 calculates the coordinates of the feature 40 in the free end face corresponding bright part 37 for all the binarized images 31 (S520), the coordinates of the feature 40 related to one binarized image, and the like. By comparing with the coordinates of the characteristic part related to the binarized image, it is possible to determine the quality of the impeller blade shape (S530).

Next, FIG. 13 which shows an example different from the processing related to the image processing means in FIG. 11 in the processing in the determination means will be described.
The configuration according to FIG. 13 is the same as described above, that is, as described with reference to FIGS. 8, 9, and 10, the impeller 2 is installed at the installation location 10, and the processes of S60 and S70 are performed. Subsequently, an example in which the processing of S80, that is, the processing of S620 to S630 is performed will be described.
The determination means 55 obtains a curve by the least square method based on the free end face corresponding bright part 37 for all the binarized images 31 (S620), and the curve relating to one binarized image and the other binary values The quality of the impeller blade shape can be determined based on the characteristic portion 40 in the free end face corresponding bright portion 37 by comparing with the curve relating to the digitized image (S630).

Next, FIG. 14 which shows an example different from FIGS. 8 and 11 in terms of the binarization process will be described.
FIG. 14 provides an inspection method in which the binarization processing step is omitted and the determination process is performed directly on the captured image 30 when the free end face corresponding portion 7a in the captured image 30 is clear.
Reference numeral 55 shown in FIG. 14A denotes a determination unit in the image processing unit 53. For all the captured images 30 captured by the imaging unit 15, the free end surface corresponding part 7 a related to one captured image 30 and other captured images. Compared with the free end surface corresponding part 7a according to the image 30, the quality of the blade shape of the impeller can be determined based on the characteristic part 40 in the free end surface corresponding part 7a. Specifically, the process (S720) shown in FIG. 14B is performed.
In S720, an arbitrary one captured image, for example, the first captured image is selected from the captured images 30 related to the entire area of the free end surface 7 in all the blades 4 obtained in S710, and the other image is used as a template. Each of the captured image and pattern matching is performed. As a result, in all the captured images, when matching with the first captured image as a template (complete match or a match of a predetermined ratio or more), it is determined that there is a correlation, and “the impeller blade shape is good” If there is a captured image that does not match, it is determined that the blade shape of the impeller is defective (with deformation).
As described above, since the binarization processing step is omitted and the determination processing step is performed directly by the captured image 30, the inspection can be performed quickly.
Steps related to the description using FIGS. 8 to 14 described above (steps of inspection relating to the entire free end surface 7 of the blade 4) and steps related to the description using FIGS. 8 to 14 described above (impeller) 2), the blade shape of the impeller 2 may be inspected. Combining these steps further improves the detection accuracy of defective products. Even when the inspection is performed by combining the processes in this way, only the bright part corresponding to the tip of the blade or only the free end surface 7 of the blade 4 is to be inspected, so the amount of calculation processing is small and the time required for the inspection is small. short.

(A) is the schematic for demonstrating the blade shape inspection apparatus of an impeller, (B) is a block diagram for demonstrating the process of an image process means. It is a figure for demonstrating a captured image, (A) is a healthy article, (B) is a figure for demonstrating a defective article. (A) is a figure for demonstrating the binarized image which binarized the image of FIG. 2 (A), (B) is the binarized image which binarized the image of FIG. 2 (B). Illustration for explaining. (A) is a block diagram for explaining the processing in the image processing means, (B) is a schematic diagram for explaining the processing in the image processing means, and (C) is an image processing means in the case where the impeller is defective. Schematic for demonstrating the process in FIG. (A)-(D) is a figure for demonstrating the example different from FIG. 4, (A) is a block diagram for demonstrating the process in an image processing means, (B) demonstrates the process in an image processing means. (C) is the schematic for demonstrating the process in the image processing means about the case where an impeller is a defective article, (D) is a figure for demonstrating the example different from (A). (E) and (F) are diagrams for explaining an example different from FIG. 4, (E) is a block diagram for explaining the processing in the image processing means, and (F) shows the processing in the image processing means. The schematic for doing is shown. (A)-(D) are the figures for demonstrating the example different from FIG. 4, (A) is a block diagram for demonstrating the process in an image processing means, (B) demonstrates the process in an image processing means. (C) is the schematic for demonstrating the process in the image processing means about the case where an impeller is a defective article, (D) is a figure for demonstrating the example different from (A). (E) and (F) are schematic diagrams for explaining an example different from the straight line S1 of FIGS. 5 (A) to (D) and FIGS. 6 (A) to (D), (E) is a front view, (F) is the elements on larger scale of (E). FIGS. 1 to 3 are diagrams for explaining an example of a different position to be inspected, (A) is a diagram for explaining a captured image (sound product), and (B) is a captured image (defective product). (C) is a diagram for explaining a binarized image obtained by binarizing the image of (A), and (D) is a binarized image obtained by binarizing the image of (B). FIG. 4E is a schematic diagram for explaining processing in the image processing means, and FIG. 4F is a schematic diagram for explaining processing in the image processing means when the impeller is defective. FIGS. 4A and 4B are diagrams for explaining an example different from FIGS. 1 to 4, in which FIG. 4A is a schematic diagram for explaining an impeller blade shape inspection device, and FIG. 4B is a block diagram for explaining processing of an image processing unit; Figure. It is a figure for demonstrating a captured image, (A) is a healthy article, (B), (C) is a figure for demonstrating a defective article. (A) is a figure for demonstrating the binarized image which binarized the image of FIG. 9 (A), (B) is the binarized image which binarized the image of FIG. 9 (B). FIG. 10C is a diagram for explaining a binarized image obtained by binarizing the image of FIG. 9C. The figure for demonstrating the process in an image processing means. The figure for demonstrating the example different from FIG. The figure for demonstrating the example different from FIG. FIGS. 8A and 11B are diagrams for explaining an example that is partially different from FIGS. 8 and 11, in which FIG. 8A is a block diagram for explaining image processing means, and FIG. 11B is a diagram for explaining processing of the image processing means;

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Impeller blade shape inspection apparatus, 2 ... Impeller, 3 ... Shaft member, 4 ... Blade, 5 ... Tip, 6 ... Tip, 7 ... Free End face, 8 ... front, 9 ... rotation axis, 10 ... installation location, 11 ... base, 12 ... support member, 15 ... imaging means, 20 ... illumination means , 21 ... irradiation light, 25 ... image processing means, 26 ... binarization processing means, 27 ... tip end detection means, 28 ... calculation means, 29 ... determination means, 30・ ・ ・ Captured image, 31 ... Binary image, 32 ... Bright part corresponding to impeller, 33 ... Bright part corresponding to shaft member, 34 ... Bright part corresponding to blade), 35 ... Tip Light-corresponding part 36, light-corresponding tip part 37, light-corresponding free-end face part 40, characteristic part 41, dark part 50, rotating means (servo motor) 51・ ・Photoelectric sensor, 52 ... adjustment mechanism, 53 ... image processing unit, 54 ... binarization processing unit, 55 ... judging means

Claims (8)

Imaging the front of the impeller from the direction of the rotational axis of the impeller placed at the installation location;
A process of binarizing a captured image related to the front surface of the impeller from the rotational axis direction of the impeller imaged in the imaging step, and obtaining a binarized image;
A step of detecting the positions of the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the impeller-corresponding bright part based on the binarized image;
For each of the detected tip-corresponding bright parts, calculating a positional relationship between each tip-corresponding bright part and a predetermined reference part in the impeller-corresponding bright part;
And a step of comparing the calculated positional relationship with a predetermined specified value to determine whether the impeller blade shape is good or bad. Imaging the front of the impeller from the direction of the rotational axis of the impeller placed at the installation location;
A process of binarizing a captured image related to the front surface of the impeller from the rotational axis direction of the impeller imaged in the imaging step, and obtaining a binarized image;
A step of detecting the positions of the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the impeller-corresponding bright part based on the binarized image;
The step of calculating the `` distance between adjacent tip-corresponding bright parts '' for the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the detected impeller corresponding bright part,
For all of the above calculated “distances between adjacent tip-corresponding bright parts”, differences in “distances between adjacent tip-corresponding bright parts” are calculated, and the difference between the difference and a predetermined specified value And a step of judging whether the impeller blade shape is good or bad. Imaging the front of the impeller from the direction of the rotational axis of the impeller placed at the installation location;
A process of binarizing a captured image related to the front surface of the impeller from the rotational axis direction of the impeller imaged in the imaging step, and obtaining a binarized image;
A step of detecting the positions of the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the impeller-corresponding bright part based on the binarized image;
Based on the binarized image, the coordinates of the position of the center of gravity and the coordinates of the minimum value existing in the concave portion located between the tip-corresponding bright parts of all the blade-corresponding bright parts are calculated, Calculating a straight line passing through the coordinates of the center of gravity position;
For the detected tip-corresponding bright part, calculating the midpoint coordinates of the tip-corresponding bright part in all the tip-corresponding bright parts, and calculating a straight line passing through the midpoint coordinates and the coordinates of the center of gravity position; ,
Calculating the angle formed by the straight line passing through the coordinates of the minimum value and the coordinates of the center of gravity, and the straight line passing through the coordinates of the midpoint coordinates and the coordinates of the center of gravity;
And a step of comparing each of the calculated angles with a predetermined specified value to determine whether the blade shape of the impeller is good or bad. Imaging the front of the impeller from the direction of the rotational axis of the impeller placed at the installation location;
A process of binarizing a captured image related to the front surface of the impeller from the rotational axis direction of the impeller imaged in the imaging step, and obtaining a binarized image;
A step of detecting the positions of the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the impeller-corresponding bright part based on the binarized image;
Based on the binarized image, the coordinates of the center of gravity position and the coordinates of the minimum value existing in the recesses located between the tip-corresponding bright parts of all the blade-corresponding bright parts are calculated, and the respective minimum values are calculated. Calculating a straight line passing through the coordinates and the coordinates of the center of gravity position;
For each of the detected tip-corresponding bright portions, a straight line passing through the coordinates of the respective minimum values and the coordinates of the center of gravity is used as a symmetry axis, and two tip-corresponding bright portions located on both sides of the symmetry axis. A blade shape inspection method for an impeller, comprising: performing pattern matching and determining whether the impeller blade shape is good or bad. Where to place the impeller,
An imaging means for imaging the impeller placed at the installation location, the arrangement of the imaging means includes an imaging means arranged to be able to image the front of the impeller from the rotational axis direction of the impeller, further,
Binarization processing means for binarizing a captured image related to the front surface of the impeller from the rotational axis direction of the impeller imaged by the imaging means;
Tip detection means for detecting the positions of the tip-corresponding bright parts of all the blade-corresponding bright parts provided around the impeller-corresponding bright part based on the binarized image obtained by the binarization processing means;
For all the tip-corresponding bright parts detected by the tip-part detecting means, calculation means for calculating the positional relationship between each tip-corresponding bright part and a predetermined reference part in the impeller-corresponding bright part,
Comprising image processing means 25 having judgment means configured to judge the quality of the impeller blade shape by comparing the positional relationship calculated by the computing means with a predetermined specified value. Car blade shape inspection device. A step of imaging an impeller that is rotatably installed at an installation location, the step of imaging the entire area of the free end surface of each blade of the impeller, and
Binarizing the captured image relating to the entire free end face of each impeller imaged in the imaging step, and obtaining the binarized image relating to the entire free end face of all the blades;
Comparing one binarized image with another binarized image for all the binarized images, and determining whether or not the shape of the impeller is good. Shape inspection method. A step of imaging an impeller that is rotatably installed at an installation location, the step of imaging the entire area of the free end surface of each blade of the impeller, and
Binarizing the captured image relating to the entire free end face of each impeller imaged in the imaging step, and obtaining the binarized image relating to the entire free end face of all the blades;
For all the above binarized images, the difference between one binarized image and the other binarized image is processed, and the quality of the blade shape of the impeller is determined based on the characteristic part in the free end surface corresponding bright part. A blade shape inspection method for an impeller. A place to install an impeller supported rotatably;
An imaging means for imaging the impeller placed at the installation location, wherein the imaging means is arranged by an imaging means arranged so as to be capable of imaging the entire free end surface of the impeller blades, and the imaging means, Rotating means adapted to rotate the impeller installed at the place where the impeller is installed so that the free end surfaces of all the blades provided around the impeller can be imaged.
Based on each of the picked-up images picked up by the image pick-up means, a binarization processing means for binarizing the picked-up image relating to the entire free end surface of each blade, and all binarization obtained by the binarization processing means. Comparing one binarized image with another binarized image for the binarized image, the image processing unit includes a determining unit configured to determine whether the impeller blade shape is good or bad. An impeller blade shape inspection device.