JPH0474661B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a surface tension measuring device. More specifically, the present invention can automatically input image data and calculate surface tension with improved accuracy, and can also simultaneously calculate basic physical property data regarding surface tension such as contact angle and density. The present invention relates to a surface tension measurement device that can detect local differences in surface tension and also calculate the surface tension of a droplet on an inclined solid plate.

(Prior art and its problems) Surface tension is one of the basic physical properties of a molten material, and various measurement methods have been used so far. The sessile drop method is one of the well-known methods, and is widely used in fields such as metallurgy as a method for measuring the surface tension of materials with high melting points, such as metal plates, without contact. It's coming.

In this type of sessile drop method, a sessile drop is usually copied onto a photographic plate or photographic paper, for example as shown in Figure 1, and the length of a predetermined part of this image is measured manually using a measuring tape. , Bashforth
- Find the surface tension by comparing it with Adams' table of numbers.

Point P(x,
In y), it is known that the following relationship holds true from the balance between the pressure caused by gravity and surface tension.

γ(1/R+1/R')=ρg(h-y)+2γ/b (1) γ is the surface tension of the droplet, ρ is the density of the droplet, g is the gravitational acceleration, and h is the peak T of the droplet. The height, b, is the radius of curvature of the contour curve at the apex of the droplet, R is the radius of curvature of the contour curve at point P, and R' is the principal radius of curvature of the droplet perpendicular to the contour surface at point P. It is also known that R' can be expressed as R'=x/sinφ (2) using the angle φ formed by the normal vector at point P.

However, it is impossible to measure the three values of R, b, and φ related to slight changes in the contour curve from the image illustrated in Fig. 1, and the above equations (1) and (2) Surface tension cannot be directly determined using . Therefore, the Bashforth-Adams table is used and the surface tension is calculated backwards.

This Bashforth-Adams table numerically solves the values of R, b, and φ for a given surface tension γ,
x and h− at a point P with a predetermined normal angle φ
This represents the relationship with y.

As a procedure for calculating surface tension, since it is generally difficult to accurately measure the normal angle φ on a curve, first, the value at point B at φ = 90°, that is,
Measure the lengths of BC and TC in Figure 1, then correct them with the values of EO and TO, and
Calculate the surface tension γ by comparing it with the Adams table.

However, calculating the surface tension as described above is a complicated procedure, and furthermore, since a predetermined region is manually measured from a quadratic curve on a plane as shown in Figure 1, there is a possibility of reading error. The disadvantage is that it expands. That is, since the apex T of the contour curve and the point B at φ=90° are both determined by drawing, the accuracy of determining these reference points itself is extremely poor, and BC, TC, EO,
Even when measuring the length of a predetermined part such as a TO, it is impossible to accurately measure the part due to the blurring of the outline that inevitably occurs due to the optical characteristics of a light emitting body or subject that has a curved surface. This, combined with the ambiguity of the reference point, was a factor that increased the error. Moreover, since it requires manual intervention, individual differences in reading cannot be avoided.

In order to overcome the above-mentioned drawbacks of manually measuring surface tension, attempts have been made in recent years to measure surface tension using a sessile drop method using a computer. For example, a curve-fitting method using a graphic digitizer has been proposed.

However, even in this case, reading errors and individual differences of the person operating the graphic digitizer intervene when reading the outline, and the drawbacks of conventional methods cannot be overcome, and it is not possible to automate the measurement. I haven't even reached it yet. On the other hand, there is also a method in which a digitized video signal is quantized into binary values on an image storage device, divided into regions, and a contour is extracted by detecting boundary pixels. However, in this method, the resolution of pixels is determined by the digitization ability of the video signal and the ability of the image storage device, so the distance, curvature, etc. between two points on a curve from a quadratic curve on a plane can be calculated. The reality is that it is difficult to accurately measure and accurately calculate surface tension.

In addition, conventional methods for measuring surface tension using computers assume the left-right symmetry of the droplet and the perfect horizontality of the solid plate, and in order to set these, various improvements have been made to the equipment. The work involved is extremely troublesome, and no definitive measure for improving measurement accuracy has yet been found.

Furthermore, in all methods of measuring surface tension using a computer, surface tension is calculated using coordinate data over the entire contour of a sessile drop.
If a chemical reaction or diffusion progresses, creating non-uniform areas in the structure of the melt, and locally changing the surface tension, not only will this change not be able to be detected, but it will also affect the measurement. There was a problem that this could cause errors.

Measuring the surface tension in this local area is useful for clarifying reaction processes that change over time, such as the reaction at the interface between molten metal and ceramic in composite materials, and the reaction between molten material and gas in a high-purity atmosphere. This is expected to provide valuable information in the field of surface tension, and it has been desired to realize a surface tension measuring device that can sufficiently deal with local differences in surface tension.

This invention was made in view of the above circumstances, and eliminates the drawbacks of accuracy, automation, and local area measurement in conventional surface tension measurements, and automatically performs image data input and surface tension calculation calculations. It is possible to calculate basic physical data related to surface tension of contact angle and density at the same time, and it is also possible to detect local differences in surface tension. This invention relates to a new surface tension measurement device that can also measure the surface tension of droplets on a solid plate.

(Means for Solving the Problems) The present invention solves the above problems by providing an optical input device that two-dimensionally inputs the shape of a molten droplet placed on a solid flat plate, and an optical input device that digitally inputs image information. A surface tension measuring device comprising: a supply device for supplying digitized image data to an arithmetic device; an image storage device for storing digitized image data; and an arithmetic storage device for storing data processed by the arithmetic device. When extracting contour coordinate data from droplet image data input to the image storage device, image intensity data near the contour coordinate data stored in the image storage device is used to correct the data at a resolution of a unit pixel or higher;
Obtain the contour coordinates, store the contour curve in an arithmetic storage device as a line figure from this contour coordinate data, and calculate the relationship between each point on this line figure using the coordinate data of multiple adjacent points. Calculate the quadratic curve that minimizes the sum of squares of the distances, calculate the curvature and normal vector direction at the contour coordinates, and use Laplace's formula from the curvature and normal vector direction corresponding to each point in the contour coordinates. To provide a surface tension measuring device, which calculates a plot regression line based on the plot, calculates the surface tension from the slope of this straight line, and displays the calculated surface tension on a display device.

In this invention, a calculation function that extrapolates the value of the calculated normal vector to the point where height = 0 of the contour coordinates and a calculation function that extrapolates the horizontal value of the contour coordinates to the position where height = 0 are provided. The contact angle and the extrapolated value are used to integrate the volume as a rotating body of the contour, and the density of the droplet is calculated from the droplet weight value measured in advance, respectively. Includes a surface tension measuring device that measures the coordinates of each point on the contour curve, minute changes in the line elements at that position, and the curvature of the contour apex, and calculates the specific surface tension at each point on the contour curve based on these. It is something to do.

Further, the present invention changes the reference coordinates in the range between the horizontal maximum value and the minimum value of the contour coordinates, corrects the symmetry of the contour figure, and adjusts the local curvature corresponding to each point of the contour coordinates. From the normal vector direction
The present invention also provides a surface tension measuring device characterized in that a plotted regression line is determined based on the Laplace equation, the surface tension is calculated from the slope of this straight line, and the surface tension is displayed on a display device.

(Example) The surface tension measuring device of the present invention will be described in more detail below by showing examples along with the drawings.

FIG. 2 is a configuration diagram showing an example of the configuration of the surface tension measuring device of the present invention.

In this example, the surface tension measuring device includes a droplet imaging section A, a video digital signal storage section B, and a calculation section C.
and a display section D.

The droplet photographing unit A is a part that photographs an optical image of a droplet. Inside the photography room 1, there are droplets 2 to be measured.
is placed on a smooth solid flat plate 3. Photography room 1
is connected to an atmosphere control device 4 and a temperature control device 5, so that the temperature of the droplet 2 and the atmosphere can be controlled as desired. The optical input device 6 is arranged at substantially the same horizontal position as the solid flat plate 3, and has a horizontal optical axis. This optical input device 6 inputs an optical image of the droplet 2, and for example, an optical camera, a video camera, etc. can be used. If the temperature is lower than that at which the droplet 2 emits light, a light emitting device 7 is provided at a position opposite to the optical input device 6 with respect to the droplet 2, and a silhouette image of the droplet 2 is photographed on the optical input device 6. You can also. In this case, any light emitting device such as a laser light emitting device can be used as the light emitting device 7. Further, when measuring at a temperature at which the droplet 2 emits light, a filter 8 may be provided immediately before the optical input device 6 in order to suppress diffused reflection in the photographing room 1.

The video digital signal storage section B is a section that digitizes the video signal of the droplet 2 from the droplet imaging section A and stores it in the image storage device 9. To digitize video information, the optical input device 6 is directly connected to the image digitizer 10, and the output signal is sent to the image storage device 9.
Once the image recording media 11 is introduced,
It is also possible to adopt any method in which the image is evacuated to the image input device 12 and introduced into the image digitizer 10 via the image input device 12. There are no particular limitations on the image recording media 11, and any media such as photographic film, photographic plates, photographic paper, and video tapes can be used. There are also no particular limitations on the image input device 12, including image pickup tubes, drum scanners, etc. An appropriate device such as a flying spot scanner, an image scanner, or a solid-state image sensor can be used. The image storage device 9 has a two-dimensional image logical space having an x-axis and a y-axis corresponding to the horizontal and vertical directions of the droplet image, respectively, and image intensity data is stored on each coordinate of this space. .

The calculation unit C includes a CPU 13 and a calculation control memory A.
14, Arithmetic control memory H15, Arithmetic control memory B
16, arithmetic control memory C17, arithmetic control memory D
18, arithmetic control memory E19, arithmetic control memory F
20 and an arithmetic storage device 21 are provided.

The CPU 13 accesses the image storage device 9 based on the control information stored in the arithmetic control memory A14, and converts the image intensity data into information in the arithmetic storage device 21. Further, the CPU 13 controls the arithmetic storage device 21 based on the control information of the arithmetic control memory H15.
is accessed, the information is converted, the surface tension value is calculated, and the result is recorded in the arithmetic storage device 21. Various control processing information is stored in each arithmetic control memory from the arithmetic control memory B16 to the arithmetic control memory F20, and the control information is transmitted to the CPU 13.

There are no particular restrictions on the display device 22 provided in the display section D, and it may be a device that can be directly read by humans such as a CRT display or a printer, or a device that can graphically display numerical values such as a blotter or graphic display. A device capable of recording on an external storage device such as a magnetic disk, a magnetic drum, a magnetic tape, or an optical disk, or any one or a combination thereof can be appropriately employed.

FIG. 3 shows the control memory A14 illustrated in FIG.
FIG. 2 is a block diagram showing a system flow of control processing according to the present invention.

First, filtering processing is performed on the image data on the image storage device 9 to remove noise. Next, image intensity change data is obtained by performing digital gradient calculation, salt-and-pepper noise is removed by a median filter, and then the coordinates of the peaks of the image intensity gradient are sequentially searched, and the corresponding logic in the arithmetic storage device 21 is Store as sample point coordinates in space. At this time, if necessary, as in the system flow illustrated in Figure 4, a threshold intensity is determined using the intermediate value of the image intensity, etc., and binarization is performed using this threshold value as a boundary value. method can also be adopted.

Next, as shown in FIG.
1, and label the boundary pixels. Calculate the coordinates of a point located on the contour using the coordinates of the sample point of the labeled boundary pixel, the intensity data of the corresponding sample point pixel on the image storage device 9, and its neighboring pixels, It is stored on the arithmetic storage device 21 as a two-dimensional array of position vectors.

FIG. 5 is a conceptual diagram showing an example of the method for calculating the coordinates of the contour points described above. The intersection points of the grid formed by dotted lines in the figure indicate sample points.
White circles indicate boundary pixels. As you can see at first glance, if we represent the contour points using boundary pixels, then x
It is impossible to achieve accuracy that has a resolution greater than the interval between discrete sample points in both the direction and the y direction. Therefore, using the coordinates of the boundary pixel Bn (x ₁ , y ₁ ) and the coordinates of the nearby out-of-area pixel E _o (x _j , y _j ) shown as a diamond in FIG. 5, we can calculate the intensity of each pixel. By using I _Bo , I _Eo and the threshold intensity I _s , the coordinates (x _o , y _o ) of the point P _o located on the contour are determined using the following formula.

x _o ＝I _s −I _Bo /I _Eo −I _Bo (x _j −x ₁ )＋x ₁ (3) y _o ＝I _s −I _Bo ／I _Eo −I _Bo (y _j −y ₁ )＋y ₁ ( 4) The point P _o obtained in this way is indicated by a black circle in Fig. 5. If the intensity gradation is L (the gradation currently generally used in image processing is 256 gradations), the resolution and coordinate accuracy will be multiplied by L, and by connecting the points P _o , the original contour curve A smooth contour figure close to . FIG. 12 shows an example of this. In addition to this example, it is also possible to calculate contour points in the same way using the intensities of multiple pixels adjacent to a boundary pixel instead of one pixel near the boundary pixel, which is effective when improving accuracy. Become.

Such a contour figure allows calculation of local curvature. It is known that the curvature of a line figure such as a digital image is defined as follows. In other words, if the difference between the direction from the pixel P _o on the curve to the m-th pixel P _o+n along the curve and the direction m pixels earlier to the pixel P _on is C _n (P _o ), then an appropriate correction C(P _o )=max(C ₁ (P _o ), C ₂ (P _o ), ..., C _K (P _o )) (5 ) is defined as At this time, if pixels are located only on discrete sample coordinates, the number of possible curvature values is less than the number of vector directions that can be taken in a 2K x 2K space, and considering vectors in the same direction, K The curvature is quantized at as few levels as 31 at =3, 47 at K=4, and 79 at K=5. on the other hand,
In the line figure obtained by calculating the contour coordinates used in this invention, the size of the space in which the vector can take is increased to (2KL)×(2KL), so it is possible to obtain almost continuous curvature values. Furthermore, this line diagram can be extended to a method for calculating local curvature, which will be described later.

Each of the figures from FIG. 6 to FIG. 10 is a block diagram showing, as a system flow, an example of control processing by the arithmetic control memory H15 illustrated in FIG. Note that the control processing shown in FIGS. 6 and 7 shows a simple method for improving the accuracy of the conventional Bashforth-Adams method.

FIG. 6 is the simplest example, in which two points are set for the contour figure on the arithmetic storage device obtained by the method described above: the point with the maximum height and the point with the maximum width of the contour. and T, A, and B shown in Figure 1.
Assuming that the end points of the contour line in contact with the surface of the solid flat plate are D and E shown in the same Figure 1, the distances of CB, TC, TO, and OD in Figure 1 are calculated from these coordinates and stored in advance in the arithmetic control memory H15. This example shows how to calculate surface tension by comparing and interpolating with the sequence of numerical solutions obtained by solving equation (1) using γ/ρ as a parameter, that is, the Bashforth-Adams numerical table. be.

FIG. 7 shows an example in which the curvature and the direction of the normal vector are used to determine the points T, A, and B shown in FIG. 1.

In Figure 1, T is the point where the curvature is minimum, A
and B are points where the direction of the normal vector is ±90°. If these points are different from the points T, A, and B obtained by the method described above, the accuracy of measurement can be improved by using the respective intermediate points. At this time, calculation of the curvature is controlled by the calculation control memory B16.

The function of the calculation control memory B16 is to cause the CPU to perform calculations for calculating curvatures and normal vectors that are not quantized and change smoothly along line elements on the contour figure. The first is a function of accessing the contour graphic vectors on the arithmetic storage device, that is, the arithmetic results controlled by the arithmetic control memory A14. This prevents quantization of the curvature values. The second function is to have the CPU perform calculations to obtain the curvature and normal vector that smoothly change along the line element.

This will be explained using FIG. 11.

Generally, the radius of curvature R at points x and y of the curve f(x, y) = 0 is R = (f _x ² + f _y ² ) ^3/2 / 2f _xy f _x f _y −f _xx f _y ² −f _yy f It is known to be expressed as _x ² (6). Note that the subscripts x and y here represent the differentials of x and y.
However, it is also known that when the (x, y) coordinates are digitized, calculations involving higher-order differentials, such as in the above equation, do not yield satisfactory values due to differential noise. Therefore, the above-mentioned equation (5) is generally used as the curvature of the digital line figure. This method is effective as a quantity that reflects the degree of change when the direction of the line vector changes rapidly over a short period of time, but when the curvature changes slowly, such as when measuring surface tension, The drawback is that it picks up subtle errors associated with digitalization and shows inaccurate curvature. That is, in Fig. 11, when finding the curvature of point P _o , the angles of P _o+n , P _o , and P _on are the sharpest in this example when m = 1; Both the sign and the value of the value will be far from the actual value. Furthermore, even if m is adopted that maximizes the angles of P _o+n , , P _o , and P _on , if there is a point where three points are located on a straight line, as in the case of m=2 in the figure,
Even if m is increased further, the curvature will be represented by 180°. Moreover, the curvature obtained by this method is a concept different from the curvature of a digital line segment, and cannot be converted into an analog and used for calculating surface tension. There is also a circle approximation method, which assumes that a curve is a combination of figures that form part of a circle, finds the curvature of the unit circle, and uses it as the curvature at that point. The reality is that it often does not reflect the actual curvature.

In this invention, a quadratic curve f _o (x, y) = a _o x ² + b _o xy + C _o y ₂ d _o x + e _o y + 1 = 0 (7) is considered for each point P _o , and the point P _o The neighborhood of ±
For m points (m≧3), find the parameters a _o , bo _, c _o , d _o , e _o so that the sum of the distances from each point to the curve is the smallest, and calculate R = ( (2ax+by+d) ² + (2cy+bx+e) ² ) ^3/2 /2b
(2ax+by+d)(2cy+bx+e)-2a(2cy+bx+e) ²
−2c (2ax+by+d) ² (8) (subscript n omitted) and φ _o = arctan (2a _o x _o + b _o y _o + d _o /2c _o y _o + b _o x _o + e _o ) (9). The values of R _o and φ _o are determined by this.

An example of the results is shown in FIG. 13, and for comparison, FIG. 14 shows the results calculated by the conventional method. Although an inaccurate curvature appears in FIG. 14, it is confirmed that the inaccuracy of the curvature is significantly reduced in FIG. 13 using a quadratic curve. Note that any method such as the least squares approximation method or the perturbation method can be used to calculate the parameters a _o , b _o , C _o , d _o , and e _o .

FIG. 8 shows an example of accessing the arithmetic control memory B16 and the arithmetic control memory C17 from the arithmetic control memory H15, and calculating the surface tension by the CPU based on the obtained control information. The control information held in the arithmetic control memory C17 has the following functions.

The arrangement of the radius of curvature R _o and the direction φ _o of the normal vector stored in the calculation storage device as the calculation result based on the calculation control memory B16 is accessed, and the following calculations are performed to obtain the obtained arrangement (u _o , v _o ) is stored on the arithmetic storage device.

u _o = g (h - y _o ) (10) v _o = 1/R _o + sin (φ _o )/x _o (11) For this arrangement of (u _o , v _o ), the arithmetic control memory C17 Perform a regression calculation to find a regression line expressed as v=Au+B (12). A corresponds to the reciprocal of specific surface tension (γ/ρ), and from this A,
Using the known density value ρ given in advance to the arithmetic control memory C17, γ=ρ/A (13) is determined, and the surface tension value is stored on the arithmetic storage device.

This method is based on the following principle.
That is, if we bring the term related to curvature in equation (1) to the left side, we get γ(1/R+1/R'-2/b)=ρg(hy) (14) Furthermore, by moving γ to the right-hand side, we get ( Using the relationship of formula 2), 1/R+sin(φ)/x-2/b=(ρ/γ)g(h-
y) (15). Here, b is the curvature of the vertex, so the point
Does not depend on the position of P _o . Also, assuming that (ρ/γ) is uniform throughout the droplet, it also does not depend on the point P _o . Therefore, by setting A=(ρ/γ), B=2/b (16) and converting the values of R and φ at (x, y), u=g(h-y), v=1/ When R+sin(φ)/x (17), the relationship v=Au+B (18) holds true. That is, A and B correspond to the slope and intercept of the regression line in equation (12).

In the example shown in FIG. 9, the arithmetic control memory H15 controls the CPU, and similarly to the example illustrated in FIG. The value of surface tension (γ/ρ) is calculated and stored in the arithmetic storage device. In this example, the value of the density ρ is not given in advance, but the value is calculated in the device based on the control information in the arithmetic control memory D18. The density ρ is calculated. At the same time, the contact angle can also be calculated based on the control information in the arithmetic control memory E19.

The control principle of the arithmetic control memory D18 follows the principle shown below.

According to the definition of curvature, the radius of curvature R is expressed as R=ds/dφ (19) using a line element _ds . Also, from the definition of a line element, the relationship between the direction of a line element and the normal vector is as follows: dx／dφ＝(ds／dφ)cosφ (20) dy／dφ＝−(ds／dφ)sinφ (21) That is, dx =Rcosφdφ (22) dy=-Rsinφdφ (23) Using A and B obtained from equation (12), R=1/Ag(h-y)+B obtained from equation (1) −(sinφ/x) (24) Using the relationship, △
Find x, △y, and substitute X _oew , y oew , φ _oew obtained from x oew = x + △ _x , y _oew = y + △y, φ _oew = φ + △ _φ into equation (24),
Find the value of R _oew . This R _oew is sequentially calculated again. Through these processes, it is possible to find the shape of the contour up to the point where it touches the solid flat plate, and by using this and integrating it as a rotating body,
The volume of the droplet can be determined. The density ρ is calculated from the weight value that has been determined in advance. In addition, with this method, at the same time, φ at the position touching the solid plate
can also be determined, and from that value, the contact angle can be determined as the angle of the tangent. With conventional methods, accurate values could not be obtained due to blurring of the optical image due to reflections, etc. particularly near the solid flat plate, but this invention makes it possible. The control method for calculating the contact angle is controlled by the calculation control memory E19.
is stored in.

The control function of the arithmetic control memory H15 illustrated in FIG. Calculate the curvature and normal vector at each point on the contour figure from the arrangement of figure vectors,
The sequences R _o and φ _o corresponding to each point P _o on the contour figure
is stored on the arithmetic storage device as
Also, based on the control information of the calculation control memory F20,
By accessing P _o (x _o , y _o ), R _o , and φ _o on the arithmetic storage device and having the CPU perform calculations, the specific surface tension (γ/ρ) _o for each P _o can be calculated. be. The function of _the calculation control _memory F20 is to calculate _the local specific surface tension ( γ/ρ) _o . That is, the coordinates (x _o , y _o ), curvature R _o and normal vector φ corresponding to the point P _o on each contour figure stored in the arithmetic storage device as the calculation result based on the control of the arithmetic control memory B _o , perform calculations based on the following formula, and store the results in the calculation storage device as a sequence of specific surface tensions.

(γ/ρ)=g(h−y _o )/1/R+1/R′−2/b (25) However, R′=x/sinφ (26), and this equation converts equation (1) It can be obtained by deforming, but in that case, if φ approaches 0,
Since the value of R' is infinite, R'=R is used.

The above calculation results are once stored in the calculation storage device and then displayed on the display device. When displaying, the contour figure, surface tension, density, and contact angle can be displayed simultaneously or separately.

Further, in the surface tension measuring device of the present invention, a control memory separate from the above-mentioned one having a function of correcting the left-right asymmetry of the droplet can be provided in the arithmetic control memory H15. On both the left and right sides of the droplet, search x _r where the curvature R _o is maximum, x 〓 where normal vector φ is 0, and x _y where height is maximum, and find the maximum and minimum values in the horizontal direction. Calculate x _t that minimizes the deviation of the regression line by changing the reference coordinate of x within a range of values to improve the symmetry of the droplet.

In this way, for example, even if the solid plate on which the droplet is placed is tilted and the symmetry of the droplet is broken, the contour curve of the droplet on the horizontal plane can be determined, and the surface tension Measurement accuracy can be improved.

Of course, the invention is not limited to the above examples. Various aspects are possible with respect to details such as the configuration, structure, and arrangement of the device.

(Effects of the Invention) As described above in detail, the surface tension measuring device of the present invention improves the precision of surface tension measurement, and also enables automation, thereby eliminating the need for manpower after image input. Thereby, errors caused by reading errors and individual differences and their expansion can be avoided.

It is possible to measure the surface tension of a droplet with unknown density, as well as the contact angle and density at the same time, and the curvature can be determined with high accuracy, so the local specific surface tension of the droplet can be determined. It also becomes possible.

Furthermore, surface tension can be measured with high precision without depending on the horizontal state of the solid plate on which the droplet is placed.
Restrictions on the equipment and special means for improving the horizontal state of the solid flat plate are not required.

[Brief explanation of the drawing]

FIG. 1 is an image diagram showing an example of the outline of a droplet. FIG. 2 is a configuration diagram showing an example of the configuration of the surface tension measuring device of the present invention. FIG. 3 is a block diagram showing the control processing by the arithmetic control memory A as a system flow. FIG. 4 is a block diagram showing another example of control processing by the arithmetic control memory A as a system flow. FIG. 5 is a conceptual diagram showing an example of a method for calculating contour point coordinates using the neighborhood intensity correction method. Figure 6, Figure 7,
FIG. 8, FIG. 9, and FIG. 10 are block diagrams each showing the control processing by the arithmetic control memory H as a system flow. Figure 11 shows the optimal 2
FIG. 2 is a conceptual diagram showing an example of a curvature calculation method using the next-order function regression method. FIG. 12 is a screen diagram showing a contour figure obtained under the control of the arithmetic control memory A. FIG. 13 is a correlation diagram showing the relationship between the normal vector of each point on the contour figure and the curvature obtained by the control of the arithmetic control memory B, and FIG.
FIG. 2 is a correlation diagram showing the relationship between a normal vector and curvature according to a conventional method. DESCRIPTION OF SYMBOLS 1...Photography room, 2...Droplet, 3...Solid plate, 4...Atmosphere control device, 5...Temperature control device, 6...Optical input device, 7...Light emitting device, 8...Filter, 9...Image storage device, 10... Image digitizer, 11... Image recording media, 12... Image input device, 13... CPU,
14... Arithmetic control memory A, 15... Arithmetic control memory H, 16... Arithmetic control memory B, 17... Arithmetic control memory C, 18... Arithmetic control memory D, 19... Arithmetic control memory E, 20... Arithmetic control memory F, 21 ...Arithmetic storage device, 22...Display device.

Claims (1)

[Scope of Claims] 1. An optical input device that two-dimensionally inputs the shape of a molten droplet placed on a solid flat plate, a supply device that digitizes video information and supplies it to an arithmetic device, and a digitized image. In a surface tension measurement device having an image storage device for storing data and an arithmetic storage device for storing data processed by a calculation device, contour coordinate data is acquired from droplet image data input to the image storage device. When extracting, image intensity data near the contour coordinate data stored in the image storage device is used to correct the image with a resolution higher than that of a unit pixel.
Obtain the contour coordinates, store the contour curve in an arithmetic storage device as a line figure from this contour coordinate data, and calculate the relationship between each point on this line figure using the coordinate data of multiple adjacent points. Calculate the quadratic curve that minimizes the sum of squares of the distances, calculate the curvature and normal vector direction at the contour coordinates, and calculate the curvature and normal vector direction corresponding to each point of the contour coordinates based on the Laplace formula. What is claimed is: 1. A surface tension measurement device that calculates a plotted regression line, calculates surface tension from the slope of this straight line, and displays the calculated surface tension on a display device. 2. When extracting contour coordinate data from the droplet image data input to the image storage device, use image intensity data in the vicinity of the contour coordinate data held in the image storage device to correct it with a resolution of more than a unit pixel, Obtain the contour coordinates, store the contour curve in an arithmetic storage device as a line figure from this contour coordinate data, and calculate the relationship between each point on this line figure using the coordinate data of multiple adjacent points. Calculate the quadratic curve that minimizes the sum of squares of the distances, calculate the curvature and normal vector direction at the contour coordinates, and add the calculated normal vector value to the point where the height of the contour coordinates is 0. It has a calculation function to extrapolate and a calculation function to extrapolate the horizontal value of the contour coordinates to the position of height = 0, and uses the contact angle and extrapolation value from the extrapolated value to integrate the volume as a rotating body of the contour. 2. The surface tension measuring device according to claim 1, wherein the density of each droplet is calculated based on the droplet weight value measured in advance and is displayed on a display device simultaneously or separately from the surface tension. 3. When extracting the contour coordinate data from the droplet image data input to the image storage device, the threshold intensity is determined using the image intensity data near the contour coordinate data stored in the image storage device. Binarize the threshold intensity as a boundary value, obtain contour coordinates, store the contour curve in an arithmetic storage device as a line figure from this contour coordinate data, and for each point on this line figure, calculate the number of adjacent points. Using coordinate data of It has a calculation function that extrapolates to the point where height = 0 of contour coordinates and a calculation function that extrapolates the horizontal value of contour coordinates to the position of height = 0, and calculates the contact angle and extrapolated value from the extrapolated value. According to claim 2, the volume is integrated as a rotating body with a contour, and the density of each droplet is calculated based on a droplet weight value measured in advance, and is displayed on a display device at the same time as or separately from the surface tension. surface tension device. 4. An optical input device that two-dimensionally inputs the shape of a molten droplet placed on a solid flat plate, a supply device that digitizes image information and supplies it to a calculation device, and an image memory that stores digitized image data. In a surface tension measurement device comprising a device and a calculation storage device that stores data processed by the calculation device, when extracting contour coordinate data from droplet image data input to the image storage device, image Correction is made at a resolution of more than a unit pixel using image intensity data near the contour coordinate data stored in the storage device,
Obtain the contour coordinates, store the contour curve in an arithmetic storage device as a line figure from this contour coordinate data, and calculate the relationship between each point on this line figure using the coordinate data of multiple adjacent points. A quadratic curve that minimizes the sum of squares of the distances is calculated, the local curvature and normal vector direction at the contour coordinates are calculated, the specific surface tension at each point on the contour curve is calculated, and the display device displays the A surface tension measuring device characterized by displaying. 5. The method according to claim 4, wherein the coordinates of each point on the contour curve, the slight change in line elements at that position, and the curvature of the contour apex are measured, and the specific surface tension at each point on the contour curve is calculated from these. Surface tension measuring device. 6 An optical input device that two-dimensionally inputs the shape of a molten droplet placed still on a solid flat plate, a supply device that digitizes video information and supplies it to a calculation device, and an image that stores digitized image data. In a surface tension measurement device comprising a storage device and an arithmetic storage device that stores data processed by the arithmetic device, when extracting contour coordinate data from droplet image data input to the image storage device, , correct the image intensity data near the contour coordinate data stored in the image storage device at a resolution of more than a unit pixel, obtain the contour coordinates, and store the contour curve from this contour coordinate data as a line figure in the calculation storage device. let me,
For each point on this line figure, use the coordinates of multiple adjacent points to calculate the quadratic curve that minimizes the sum of squares of the distances to these points, and calculate the curvature and normal at the contour coordinates. Calculate the vector direction and
By changing the reference coordinates within the range of the horizontal maximum and minimum values of the contour coordinates, correcting the symmetry of the contour figure, and calculating the Laplace equation from the curvature and normal vector direction corresponding to each point of the contour coordinates. What is claimed is: 1. A surface tension measuring device characterized in that a plot regression line is calculated based on the equation, the surface tension is calculated from the slope of this straight line, and the surface tension is displayed on a display device. 7 Change the reference coordinates within the range of the horizontal maximum and minimum values of the contour coordinates, calculate the plot regression line based on the Laplace formula from the curvature and normal vector direction corresponding to each point of the contour coordinates, and perform the regression. Select the reference coordinate that minimizes the deviation and Laplace again.
7. The surface tension measuring device according to claim 6, wherein a plotted regression line is determined based on the equation, and the surface tension is calculated from the slope of this straight line and displayed on a display device.