JP7079396B2 - Surface texture analysis method, surface texture comparison evaluation method, and surface shape measurement system - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Act https: // www. jstage. jst. go. jp / article / jjsde / advpub / 0 / advpub_2017 / 2760 / _pdf / -char / ja

The present invention relates to a method for analyzing and evaluating the surface texture of a measured surface, and a surface shape measuring system having a function of executing the method, and more particularly, radon for the load curve data of the surface shape profile of the measured surface. The present invention relates to a method for analyzing and evaluating the surface texture of the measured surface based on the projected image data obtained by performing the conversion, and a surface shape measuring system having a function of executing the method.

As a device for measuring the surface shape of the measurement surface, a stylus type device (Patent Document 1), a non-contact type device using laser light (Patent Document 2), a stylus type sensor and a non-contact type Various types of sensors have been developed, such as those equipped with the above sensors (Patent Document 3). Some of these surface shape measuring devices have a function of analyzing and evaluating surface texture based on a fine surface shape profile measured by using a stylus or an optical sensor. Surface roughness such as arithmetic mean roughness Ra, maximum height Rz, and root mean square height Rq is often used as a typical analysis / evaluation parameter.

However, the surface roughness as described above may not be enough to sufficiently express the characteristics of the surface texture. For example, the sliding surface of an automobile engine or machine tool is formed with a smooth portion (plateau region) as shown in FIG. 1 by using plateau honing or grinding in order to improve friction characteristics and wear characteristics. It may be a so-called plateau structure surface in which rough portions (valley regions) are mixed. Since the surface roughness of such a plateau structure surface differs greatly between the plateau region and the valley region, the surgical average roughness Ra, the maximum height Rz, the root mean square height Rq, etc. of the entire plateau structure surface are obtained. However, it is difficult to properly evaluate its characteristics. Therefore, in JIS / ISO (Non-Patent Documents 1, 2, and 3), two regions having different characteristics are separately evaluated by using a load curve on a normal probability paper. Here, the load curve is a curve showing the ratio of the actual portion and the void portion with respect to the height direction of the profile in which the planar unevenness is measured, and is a probability distribution function of the cross-sectional length of the real portion. The normal probability paper is expressed by converting the percentage% on the horizontal axis of the load curve into the standard deviation σ at equal intervals, and by finding the slope of the plotted data, the standard deviation of that part is approximated. It has the characteristic that it can be obtained. The surface shape profile of the plateau structure surface of FIG. 1 is converted into a load curve on a normal probability paper and plotted as shown in FIG. The plateau structure surface has plateau regions and valley regions with different statistical properties, but in the load curve on the normal probability paper, these two regions appear as two separate linear feature regions R1 and R2. Since the roughness parameter corresponding to the root mean square height Rq can be obtained from the slope of the load curve on the normal probability paper, the roughness parameter Rpq of the plateau region and the roughness of the valley region can be obtained by obtaining the slope of each linear feature region. The parameter Rvq can be obtained separately.

JIS / ISO (Non-Patent Document 3) proposes a method of fitting a hyperbola to a load curve as a specific evaluation method of a plateau structure surface. The procedure of this method is as follows. First, the first hyperbola fitting operation is performed on the load curve on the normal probability paper. Second, find the bisector of the asymptote of the fitted hyperbola, and find the intersection of the bisector and the load curve. This intersection is a temporary displacement point from the plateau region to the valley region. Third, the quadratic derivative is sequentially calculated at each point of the load curve in the plateau region and the valley region, starting from the load length ratio of the displacement point. The value of the obtained quadratic derivative is normalized by the standard deviation of the value of the quadratic derivative, and its absolute value is taken. The positions where this value is larger than 6 are defined as the upper limit value UPL of the plateau region and the lower limit value LVL of the valley region, respectively. Fourth, the vertical axis of the load curve is enlarged / reduced and normalized so that the region within the range of UPL and LVL becomes a square. Fifth, the hyperbola fitting operation is executed again on the normalized load curve, and the asymptote of the obtained hyperbola is obtained. Sixth, the bisector is obtained three times for the asymptote, and the intersection of the two bisectors at both ends and the load curve is obtained. Let the two intersections be the lower limit LPL of the plateau region and the upper limit UVL of the valley region, respectively. Seventh, the least squares straight line fitting is performed on the range of the plateau region (UPL-LPL) and the range of the valley region (UVL-LVL) obtained in the fourth and sixth. The roughness parameter Rpq of the plateau region and the roughness parameter Rvq of the valley region are derived from the slopes of the two straight lines that are applied, and the load length ratio that is an index of the boundary between the plateau region and the valley region is derived from the intersection of the two straight lines. Derive Rmq.

Since the above method is very complicated, attempts have been made to devise an algorithm or the like for applying a hyperbola in order to alleviate this (Non-Patent Documents 4 and 5).

Japanese Unexamined Patent Publication No. 2000-97691 Japanese Unexamined Patent Publication No. 2012-2573 Japanese Unexamined Patent Publication No. 2016-200399

JIS B 0671-1 JIS B 0671-2 JIS B 0671-3 Evaluation of surface texture of two-layer structure (1st report, hyperbola approximation of load curve), Journal of Design Engineering, 44,5 (2009), 301-308 Evaluation of surface texture of two-layer structure (2nd report, derivation of parameters), Journal of Design Engineering, 44, 11 (2009), 624-629

As described above, the method of fitting a hyperbola to a load curve requires complicated preprocessing for the fitting operation of the hyperbola, and the method requires complicated calculation. Therefore, the method is executed by a computer. In order to do so, it is necessary to implement an esoteric algorithm. Although it is expected that the complexity of the calculation will be reduced to some extent by the methods shown in Non-Patent Documents 4 and 5, it still includes the fitting operation of the hyperbola and requires the implementation of a complicated algorithm.

Therefore, the present invention provides a surface texture analysis method, a comparative evaluation method, and a surface shape measurement system for executing these methods, which do not require complicated preprocessing such as a conventional double curve fitting method or implementation of an esoteric algorithm. The purpose is to do.

That is, the present invention
The step of performing Radon transform on the load curve data of the surface shape profile of the measurement surface,
A step of specifying a data concentration point in the r-θ plane of the projection image data obtained by the Radon transform, and
To provide a surface texture analysis method including.

As will be described in detail later, the data concentration point in the r-θ plane of the projected image data obtained by the radon transformation appears in relation to the linear feature region of the load curve data of the surface shape profile. By specifying it, it becomes possible to analyze the linear feature region. Further, in the case of a surface having two or more regions having different surface roughness, such as a plateau structure surface, the data concentration points corresponding to each region appear separately, so that the data concentration points are based on each data concentration point. It is also possible to separate and analyze each region having different surface roughness. That is, in the surface texture analysis method using the Radon transform, the surface shape is appropriately analyzed even in the case of a surface having two or more regions having different surface roughness, such as a plateau structure surface. Is possible. In addition, since it does not include the operation of fitting the hyperbola to the load curve, it is possible to eliminate the need for complicated preprocessing and implementation of an esoteric algorithm as in the conventional method. Although the present invention is particularly effective in analyzing the surface of the plateau structure, the target measurement surface is not limited to the surface of the plateau structure.

Preferably, it may further include a step of calculating linear parameters that apply to the linear feature region on the load curve data based on the position of the identified data concentration point on the r-θ plane. can. Further, a step of evaluating the surface texture of the measurement surface based on the parameter of the straight line can be further included. Further, the step of evaluating the surface texture may include a step of calculating the roughness parameter of the linear feature region from the inclination parameter among the parameters of the linear line.

Specifically, the measurement surface is a plateau structure surface including a plateau region and a valley region, and the load curve data is used as load curve data on normal probability paper.
The step of calculating the parameter of the straight line is the step of calculating the parameter of the valley region straight line applicable to the first linear feature region corresponding to the valley region on the load curve data, and the plateau on the load curve data. A second linear feature corresponding to the region can include a step of calculating the parameters of the plateau region straight line that applies to the region.

Preferably, the valley region roughness parameter (Rvq) is calculated from the slope parameter of the valley region straight line parameter, and the plateau region roughness parameter (Rpq) is calculated from the slope parameter of the plateau region straight line parameter. ) Can be included. Further, it is possible to further include a step of calculating the intersection of the valley region straight line and the plateau region straight line on the normal probability paper to obtain the boundary (Rmq) between the valley region and the plateau region.

In particular,
The step of identifying the data concentration point is
A step of obtaining a class showing a peak when the projected image data is used as a first histogram having a first class width, and
The r-θ is based on the distribution of the second histogram when the data included in the class among the projected image data is used as the second histogram having the width of the second class narrower than the width of the first class. Steps to determine the position of the data concentration point on a plane,
Can be included.

Further, the present invention
The step of performing Radon transform on the load curve data of the surface shape profile of the reference measurement surface and acquiring the reference projection image data,
The step of performing Radon transform on the load curve data of the surface shape profile of the comparative measurement surface and acquiring the comparative projection image data,
A step of evaluating the difference between the surface texture of the reference measurement surface and the surface texture of the comparative measurement surface based on the reference projection image data and the comparative projection image data.
Provided is a method for comparing and evaluating surface properties including.

Similar to the above-mentioned surface texture analysis method, the surface texture comparison and evaluation method also makes it possible to eliminate the need for complicated preprocessing and implementation of an esoteric algorithm as in the conventional hyperbola fitting method. ..

Preferably,
The step to evaluate the difference is
A step of specifying a data concentration point in the r-θ plane of the reference projection image data, and
A step of specifying a data concentration point in the r-θ plane of the comparative projection image data, and
A step of comparing the position of the data concentration point on the reference measurement surface with the position of the data concentration point on the comparative measurement surface in the r-θ plane.
Can be included.

The present invention further provides a surface texture analysis program for causing a computer to execute the above-mentioned surface texture analysis method, and a surface texture comparison evaluation program for causing a computer to execute the above-mentioned surface texture comparison and evaluation method.

The invention of the present application further
A surface shape measuring means for measuring the surface shape profile of the measurement surface,
A data conversion unit that converts the surface shape profile measured by the surface shape measuring means into load curve data, and a data conversion unit.
An arithmetic unit that executes at least one of the surface texture analysis program and the surface texture comparison evaluation program, and
To provide a surface shape measuring system.

Hereinafter, an embodiment of a surface texture analysis method, a surface texture comparison evaluation method, and a surface shape measurement system according to the present invention will be described with reference to the accompanying drawings.

It is a figure which shows the surface shape profile of the plateau structure surface. It is a figure which shows the load curve on the normal probability paper of the surface shape profile of FIG. It is a conceptual diagram for demonstrating the Radon transform for the load curve data. It is a figure which shows the projection image data obtained by performing Radon transform on the load curve data of FIG. It is a schematic diagram which shows the surface shape measurement system which concerns on one Embodiment of this invention. It is the first figure which shows the operation of the surface texture analysis program in the surface shape measurement system of FIG. FIG. 2 is a second diagram showing the operation of the surface texture analysis program in the surface shape measurement system of FIG. It is a figure which shows the relationship between the load curve of FIG. 2 and a plateau area straight line and a valley area straight line obtained by calculation. It is a figure which shows the operation of the surface property comparison evaluation program in the surface shape measurement system of FIG. It is a figure which shows the load curve of the surface shape profile of another plateau structure surface.

In the surface texture analysis method according to the embodiment of the present invention, the projected image data of the load curve is acquired by performing the radon conversion on the load curve data of the surface shape profile of the measurement surface, and the measurement is performed based on the projected image data. It analyzes and evaluates the surface texture of the surface.

である。したがって、対象物ｆ(ｘ,ｙ)の投影像ｐ(ｒ,θ)は、

の積分変換によって求めることができる。この積分変換がラドン変換である。 Here, the radian transformation is an integral transformation that is generally used when obtaining a projected image in X-ray CT, and specifically, it is applied to an object f (x, y) existing on xy orthogonal coordinates. On the other hand, the object f (x, y) is projected in the rs coordinate system rotated by an angle θ [rad] with respect to the xy orthogonal coordinate system. The projection of an object can be defined as the line integral value of the object on a straight line perpendicular to the projection plane. The coordinates rotated by θ [rad] from the X axis are

Is. Therefore, the projected image p (r, θ) of the object f (x, y) is

It can be obtained by the integral transform of. This integral transform is the Radon transform.

Performing the radon conversion on the load curve data conceptually regards the load curve as an object, and as shown in FIG. 3, changes the projection angle within a predetermined angle range and changes the projection angle to the projection surface at each projection angle. It means to get a projected image of the load curve projected on.

The projected image data obtained by performing the radian conversion in the range of the projection angle θ from 0 [rad] to π [rad] with respect to the load curve data (Fig. 2) on the normal probability paper of the plateau structure surface (Fig. 1). , Fig. 4 is obtained by plotting on the r−θ plane where the horizontal axis is the position r on the projection plane and the vertical axis is the projection angle θ. Taking a cross section parallel to the horizontal axis gives a projected image p ( _{ri, θ j )} at each projection angle in FIG. As can be seen from FIG. 4, two data concentration points D and E appear in the projected image data on the r−θ plane. The fact that the data is concentrated on the r-θ plane means that a lot of load curve data looks linear when the load curve is viewed in the direction of the projection angle θ from the position r on the projection plane at the projection angle θ. That's what it means. The data concentration point D appears at a position where the projection angle θ is about π / 4, which corresponds to the projected image at the position A in FIG. Therefore, the data concentration point D indicates the linear feature region R1 corresponding to the valley region. Similarly, the data concentration point E appears at the position B in FIG. 3 and indicates the linear feature region R2 corresponding to the plateau region. On the other hand, the data concentration point does not appear in the projected image at the position as shown in C in FIG. 3, which does not face any of the linear feature regions. That is, the data concentration points appear in the projected image data in the r−θ plane as many as the number of linear feature regions of the load curve.

正規確率紙上の負荷曲線では、プラトー領域および谷領域にそれぞれ対応する直線的特徴領域Ｒ１，Ｒ２の傾きにから二乗平均平方根高さＲｑに相当するプラトー領域の粗さパラメータＲｐｑ及び谷領域の粗さパラメータＲｖｑを求めることができる。プラトー領域に対応するデータ集中点Ｅのｒ－θ平面上での位置を（ｒ_ｐ、θ_ｐ）とし、谷領域に対応するデータ集中点Ｄのｒ－θ平面上での位置を（ｒ_ｖ、θ_ｖ）とすると、

となる。なお、負荷曲線は単調減少であり、粗さパラメータＲｐｑ、Ｒｖｑは正の値で扱うため、粗さパラメータＲｐｑ、Ｒｖｑの符号は正としている。プラトー領域と谷領域の境界Ｒｍｑは、プラトー領域直線と谷領域直線の交点のｘ軸の座標から求めることができる。プラトー領域直線と谷領域直線の交点のｘ軸の座標ｔを極座標表現すると、

となる。ここで、正規確率紙のｘ軸は標準偏差であるが、プラトー領域と谷領域の境界Ｒｍｑは１００分率であるため、標準正規分布関数に数式５を代入し、１００分率にするためにそれを１００倍すると、境界Ｒｍｑは以下のようになる。

When the positions of the data concentration points D and E on the projection plane data are specified, the linear parameters applicable to the linear feature regions R1 and R2 on the load curve data are calculated based on the positions, and the following linear equation is obtained. Can be decided.

In the load curve on the normal probability paper, the roughness parameter Rpq of the plateau region and the roughness of the valley region corresponding to the root mean square height Rq from the slopes of the linear feature regions R1 and R2 corresponding to the plateau region and the valley region, respectively. The parameter Rvq can be obtained. The position of the data concentration point E corresponding to the plateau region on the r-θ plane is (r _p , θ _p ), and the position of the data concentration point D corresponding to the valley region on the r-θ plane is (r _v ). , Θ _v )

Will be. Since the load curve is monotonically decreasing and the roughness parameters Rpq and Rvq are treated as positive values, the signs of the roughness parameters Rpq and Rvq are positive. The boundary Rmq between the plateau region and the valley region can be obtained from the coordinates of the x-axis of the intersection of the plateau region straight line and the valley region straight line. When the x-axis coordinate t of the intersection of the plateau area straight line and the valley area straight line is expressed in polar coordinates,

Will be. Here, the x-axis of the normal probability paper is the standard deviation, but the boundary Rmq between the plateau region and the valley region is 100%, so in order to substitute Equation 5 into the standard normal distribution function to make it 100%. When it is multiplied by 100, the boundary Rmq becomes as follows.

By performing Radon transform on the load curve data in this way and specifying the data concentration point of the projected image data in the r-θ plane, the plateau region and the valley region on the plateau structure surface are separated and analyzed, respectively. Roughness parameters Rpq and Rvq can be obtained. It is also possible to obtain the boundary Rmq between the plateau region and the valley region. These parameters make it possible to appropriately analyze and evaluate the surface texture of the plateau structure surface. This surface texture analysis method does not require an operation for applying a hyperbola, which is required in the conventional method, and can be executed by a relatively simple algorithm.

As shown in FIG. 5, the surface shape measuring system 1 according to the embodiment of the present invention includes a surface shape measuring device 10 for measuring the surface shape and a computer computing device 50 communication-connected to the surface shape measuring device 10. And prepare. The computer arithmetic unit 50 is equipped with a surface texture analysis program for executing the above-mentioned surface texture analysis method.

The surface shape measuring device 10 includes a stylus type sensor unit (surface shape measuring means) 16 attached to a vertical moving mechanism 14 fixed to the device frame 12. The surface shape measuring device 10 moves the measured object W in the horizontal direction by the horizontal plane moving mechanism 20 in a state where the stylus 18 of the sensor unit 16 is in contact with the measurement surface S of the measured object W, and measures at that time. The amount of displacement of the stylus 18 that moves up and down along the unevenness of the surface S is detected. The detected displacement amount is transmitted to the computer arithmetic unit 50 together with the position information of the horizontal plane movement mechanism 20, and the computer arithmetic unit 50 stores them as a surface shape profile in the memory 52. The above-mentioned surface texture analysis program is also stored in the memory 52. The computer arithmetic unit 50 further includes a data conversion unit 54 and an arithmetic unit 56, and the data conversion unit 54 and the arithmetic unit 56 execute a surface texture analysis program.

Taking the case where the plateau structure surface shown in FIG. 1 is used as the measurement surface, the operation of the surface shape measurement system will be described below with reference to FIGS. 6 and 7.

The surface shape of the measured surface S of the object W to be measured is measured by the surface shape measuring device 10 as described above, and the surface shape profile is stored in the memory 52 of the computer arithmetic unit 50 (S10). Next, the data conversion unit 54 of the computer arithmetic unit 50 converts the surface shape profile stored in the memory 52 into load curve data on the normal probability paper (S20). FIG. 2 shows the load curve data obtained from the surface shape profile data of 8001 points on the plateau structure surface of FIG. 1 plotted on a normal probability paper.

Next, the arithmetic unit 56 of the computer arithmetic unit 50 performs the Radon transform on the load curve data based on the above equation 2 (S30). The projection image data (FIG. 4) obtained by the Radon transform is stored in the memory 52. Next, the data concentration points D and E in the r−θ plane of the projected image data are specified (S40). The positions of the data concentration points D and E on the r−θ plane are specified using a histogram. Specifically, the projection image data obtained at each projection angle θ is converted into a first histogram in which the first class width is set so that the number of classes at the position r is 400 (S41). Next, in the first histogram, the projection angle θ indicating the peak and the class of the position r are specified (S42). Further, the data included in the class indicating the peak of the projected image data is converted into a second histogram in which the second class width is set so that the number of classes at the position r is 20 (S43). The width of the second class is inevitably narrower than the width of the first class. The position r of the data concentration point is determined based on the distribution of the second histogram (S44). In the present embodiment, the median value of the second histogram is adopted as the value of the position r, but other values such as the peak value and the average value can also be adopted. In the present embodiment, the number of classes is fixed at 400 in the first histogram, but the number of classes can be arbitrarily changed. However, if the number of classes is too small or too large, it may not be possible to detect peaks properly. Alternatively, the first class width may be fixed to a predetermined value so that the number of classes varies according to the size of the data range. These are the same in the second histogram.

Once the positions of the data concentration points D and E in the r-θ plane are specified, the following applies to the linear feature region R1 corresponding to the valley region on the load curve data based on the positions of the data concentration points D and E. Calculation of the parameters of the valley region straight line Lv and the parameters of the plateau region straight line Lp applicable to the linear feature region R2 corresponding to the plateau region is executed based on the above equations 3 to 5 (S50). As shown in FIG. 8, the valley region straight line Lv and the plateau region straight line Lp obtained by this calculation are in good agreement with the linear feature regions R1 and R2. Finally, the surface texture of the measured surface S is evaluated based on the linear parameter (S60). Here, the linear parameters are specifically the slope a and the intercept b in the above equation 3, but since the slope a is a roughness parameter corresponding to the root mean square height Rq as described above, the slope a is used. Finding means finding the roughness parameter (Equation 4). Further, the intersection of two straight lines Lv and Lp can be obtained by obtaining the intercept b in addition to the slope a, and the boundary Rmq (formula 6) between the plateau region and the valley region can be obtained from the x-axis coordinate t (formula 5) of the intersection. Can be asked. In the embodiment, the surface texture is analyzed and evaluated by obtaining the roughness parameter Rpq of the plateau region, the roughness parameter Rvq of the valley region, and the boundary Rmq of the plateau region and the valley region.

The surface shape measurement system 1 is also equipped with a surface texture comparison and evaluation program that executes the surface texture comparison and evaluation method described below.

The surface texture comparison and evaluation method utilizes the above-mentioned surface texture analysis method. Specifically, as shown in FIG. 9, the surface shape profile of the reference measurement surface as a reference is first measured (S110), the surface shape profile is converted into load curve data on normal probability paper (S120), and then the load is applied. Radon transform is performed on the curve data to acquire the reference projection image data (S130). Next, the comparative projection image data is acquired in the same manner for the comparative measurement surface to be compared (S140-S160). These processes are the same as the processes in S10-S30 of FIG. Next, the difference between the surface texture of the reference measurement surface and the surface texture of the comparison measurement surface is evaluated based on the reference projection image data and the comparison projection image data (170). In the embodiment, when evaluating the difference, the process described above as S40-S60 is executed to identify the data concentration point, and the parameters Rpq, Rvq, and Rmq on the reference measurement surface and the comparative measurement surface are set. Obtained and compare each parameter on the comparative measurement surface with each parameter on the reference measurement surface. In the evaluation of the difference (S170), it is not always necessary to compare the above parameters. For example, it is possible to directly compare the position of the data concentration point on the r-θ plane of the reference projection image data with the position of the data concentration point on the r-θ plane of the comparative projection image data to evaluate the difference in measurement properties. Further, when the surface shape profile of the plateau structure surface is a load curve on normal probability paper, two linear feature regions usually appear as shown in FIG. 2, but in some cases, a third linear feature region appears as shown in FIG. Feature region R3 may appear. Such a third linear feature region R3 often appears when deep scratches or cracks are formed on the surface. As described above, the number of linear feature regions in the load curve can be specified by the number of data concentration points in the r−θ plane. Therefore, it is also possible to evaluate the difference in surface texture by comparing the number of data concentration points, that is, the number of peaks when the projected image data is converted into a histogram. Also, instead of comparing specific values such as roughness parameters and the number of peaks, the shape of the load curve on the reference measurement surface and the load curve on the comparative measurement surface are compared to evaluate the difference in surface texture. May be good. Furthermore, the measurement surface can be classified based on a predetermined criterion based on the number of peaks and the shape of the load curve.

The surface property evaluation method is effective, for example, in quality control at a production site. Specifically, the projection image data and the roughness parameter of the surface shape profile of the standard product are acquired in advance. Then, the surface shape profile of the products to be produced in sequence is measured to obtain projection image data and roughness parameters, and the surface properties of the standard product and the surface properties of the product are compared and evaluated as described above. .. This makes it possible to control quality variations. It is also possible to determine the pass / fail of the product by determining whether or not the surface texture of the product is within a predetermined range with respect to the surface texture of the reference product.

In the surface shape measuring system 1, as an auxiliary function, it is also possible to obtain a straight line applicable to the linear feature portion of the load curve data by manual operation. In this method, the load curve data of the measured surface texture profile is displayed on the monitor 58, and the operator manually specifies the range of the linear feature area that is considered to correspond to the plateau area judging from the shape of the load curve. do. The calculation unit 56 performs a straight line fitting operation by the least squares method on the data included in the manually specified range, and obtains the parameters of the straight line applicable to the linear feature area. The arithmetic unit 56 further obtains the roughness parameter Rpq based on the linear parameter. Similarly for the valley region, the roughness parameter Rvq is obtained by manually specifying the range of the linear feature region. After this, the boundary Rmq is also calculated. Each obtained parameter is displayed on the monitor 58.

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments. For example, in the above embodiment, the Radon transform is performed on the load curve on the normal probability paper, but the Radon transform can be performed on the linear load curve for analysis and evaluation. The measurement surface targeted by the present invention is not limited to the plateau structure surface, and can be applied to any measurement surface. In the above embodiment, the surface shape measuring device and the computer arithmetic device are separate devices that are communicated with each other, but may be configured as an integrated device. Further, the surface texture analysis program and the surface texture comparison evaluation program do not need to be executed by a dedicated computer corresponding to the surface shape measuring device, and can be installed and executed in a normal personal computer.

1 Surface shape measuring system 10 Surface shape measuring device 12 Device frame 14 Vertical movement mechanism 16 Sensor unit (surface shape measuring means)
18 Stylus 20 Horizontal plane movement mechanism 50 Computer arithmetic unit 52 Memory 54 Data conversion unit 56 Arithmetic unit 58 Monitor W Instrument S Measurement surface

Claims (13)

The step of performing Radon transform on the load curve data of the surface shape profile of the measurement surface,
A step of specifying a data concentration point in the r-θ plane of the projection image data obtained by the Radon transform, and
Surface texture analysis method including. The surface of claim 1, further comprising calculating a linear parameter that applies to the linear feature region on the load curve data based on the position of the identified data concentration point on the r-θ plane. Property analysis method. The surface texture analysis method according to claim 2, further comprising the step of evaluating the surface texture of the measured surface based on the parameters of the straight line. The surface texture analysis method according to claim 3, wherein the step of evaluating the surface texture includes a step of calculating the roughness parameter of the linear feature region from the inclination parameter among the linear parameters. The measurement surface is a plateau structure surface including a plateau region and a valley region, and the load curve data is used as load curve data on normal probability paper.
The step of calculating the parameter of the straight line is the step of calculating the parameter of the valley region straight line applicable to the first linear feature region corresponding to the valley region on the load curve data, and the plateau on the load curve data. The surface texture analysis method according to any one of claims 2 to 4, comprising a step of calculating a parameter of a plateau region straight line applicable to a second linear feature region corresponding to the region. The roughness parameter (Rvq) of the valley region is calculated from the slope parameter of the valley region straight line, and the roughness parameter (Rpq) of the plateau region is calculated from the slope parameter of the plateau region straight line parameter. The surface texture analysis method according to claim 5, which comprises the step of performing. The fifth or sixth aspect of the present invention further comprises a step of calculating the intersection of the valley region straight line and the plateau region straight line on the normal probability paper to obtain the boundary (Rmq) between the valley region and the plateau region. Surface texture analysis method. The step of identifying the data concentration point is
A step of obtaining a class showing a peak when the projected image data is used as a first histogram having a first class width, and
The r-θ is based on the distribution of the second histogram when the data included in the class among the projected image data is used as the second histogram having the width of the second class narrower than the width of the first class. Steps to determine the position of the data concentration point on a plane,
The surface texture analysis method according to any one of claims 1 to 7, which comprises. The step of performing Radon transform on the load curve data of the surface shape profile of the reference measurement surface and acquiring the reference projection image data,
The step of performing Radon transform on the load curve data of the surface shape profile of the comparative measurement surface and acquiring the comparative projection image data,
A step of evaluating the difference between the surface texture of the reference measurement surface and the surface texture of the comparative measurement surface based on the reference projection image data and the comparative projection image data.
A method for comparing and evaluating surface properties, including. The step to evaluate the difference is
A step of specifying a data concentration point in the r-θ plane of the reference projection image data, and
A step of specifying a data concentration point in the r-θ plane of the comparative projection image data, and
A step of comparing the position of the data concentration point on the reference measurement surface with the position of the data concentration point on the comparative measurement surface in the r-θ plane.
9. The surface texture comparative evaluation method according to claim 9. A surface texture analysis program for causing a computer to execute the surface texture analysis method according to any one of claims 1 to 8. A surface texture comparison evaluation program for causing a computer to execute the surface texture comparison evaluation method according to claim 9 or 10. A surface shape measuring means for measuring the surface shape profile of the measurement surface,
A data conversion unit that converts the surface shape profile measured by the surface shape measuring means into load curve data, and a data conversion unit.
An arithmetic unit that executes at least one of the surface texture analysis program according to claim 11 and the surface texture comparison evaluation program according to claim 12.
A surface shape measurement system.

Images