JPH0282374A - Void ratio measuring method for porous substance - Google Patents

Abstract

PURPOSE:To highly accurately measure the void ratio of a porous substance by dividing and extracting a cross section area from a picture obtained with photographing the cross section of the porous substance, making area other than the cross section area into the void area, and calculating the area ratio of these areas. CONSTITUTION:A polystyrene sintered body cross section is photographed by a scanning type electron microscope (SEM) 3, and the obtained microscopic picture is inputted to a picture processor (IP) 4. To detect the even density area for data stored into a picture memory in the IP 4 by an engineering station 9, an edge is detected by a primary differentiation operator, and an inverted binarization processing is executed. A picture processing to eliminate the small areas of the obtained picture is executed, and the binary picture whose even density area has the value of 1 is obtained. Next, the even density area is segmented from the original picture with the binary picture as a mask, the extracted area is combined with four types of characteristic quantities, the cross section area is discriminated from the void area, the area ratio between the cross section area and void area is calculated, and displayed on a monitor 10 with the area ratio of the void area as the void ratio. Thus, the void ratio of the porous substance can be highly accurately measured.

Description

[Detailed description of the invention] Ru.

Since the cross-sectional area has a uniform concentration, the cross-sectional area can be extracted by first performing first-order differential processing and inverting and binarizing the differential value, but even the uniform-concentration area within the void area is extracted.

A uniform density area is extracted from the original image using the above-mentioned binarized image as a mask, and only the cross-sectional area is separated using the following feature amount, and the area ratio with respect to the other void areas is calculated. The determination based on this feature amount is as shown in Table 1.

Feature Quantity ■Average Concentration Table 1 FIG. 2 shows a block diagram of a measuring device for carrying out an embodiment of the present invention.

CC image of the cut surface of the porous body enlarged with optical microscope 1
Send an image signal obtained by photographing with a D (charge-coupled device) camera 2 or an image signal obtained by photographing with a SEM (scanning electron microscope) 3 to an IP (image processing device) 4;
It is converted into a digital image by an A/D converter provided in IP4. The digital image is stored in the image memory in the IF 5, displayed on the high-definition monitor 5, and copied as a hard copy 6 if necessary. The image is disk 7
Alternatively, it can be recorded on the hard disk 8.

To measure the porosity of the porous body, the data stored in the image memory provided at rp is subjected to image processing shown in the flowchart of FIG. 3 using an EWS (engineering workstation) 9. The results processed by the EWS 9 are displayed on a monitor 10, and printed out from a printer 11 if necessary.

The example of image processing in this example is the measurement of a polyethylene sintered body, and the image input from the SEM3 is: - Edge detection using a first-order differential operator - Binarization - Small area - Discrimination between cross-sectional areas and void areas based on the average density of the area, the peak width of the area concentration histogram, and the edge strength of the area outline - Processing of porosity (area ratio) calculation to calculate the porosity quickly and with high accuracy It is measured with. The four types mentioned above are defined as feature quantities, and an appropriate combination of these feature quantities can be selected depending on the object, so it can be applied to many cases of porous materials, and the porosity can be determined with high precision. can be measured.

FIG. 3 is a flowchart showing an outline of the measurement procedure according to the embodiment of the present invention.

FIG. 4 shows a processed image of a polystyrene sintered body processed according to the flowchart shown in FIG.

First, in step S1, a cut surface of the polystyrene sintered body is photographed by SEM3, and the obtained microscopic image is
Enter in F4. The original image shown in FIG. 4 (Δ) is a microscopic image, and the magnification is 150 times.

1 in step S2 to extract a uniform density area.
Edge detection is performed using a differential operator, and inversion binarization processing is performed in step S3. An image obtained by such processing is shown in FIG. 4(B). The uniform density area is shown in white.

Although there are small regions in the images obtained by the processing in steps S2 and S3, since the cross-sectional region is larger than a predetermined size, image processing to remove the small regions is performed in step s4.

Through the above processing, a binary image is obtained in which the uniform density area has a value of 1, as shown in FIG. 4(C).

Next, in step S5, a uniform density region is cut out from the original image (see FIG. 4(A)) using the binary image shown in FIG. 4('C) as a mask, and four types of feature quantities are In combination, the cross-sectional area and void area are determined.

In step S6, the area ratio between the cross-sectional area and the void area,
That is, the porosity is calculated. Then, in step S7, the result is displayed on the monitor 10 using the area ratio of the void region as the porosity.

The detailed contents of the image processing performed as described above are as follows.

(1) Detection of a uniform density region In order to detect a region where the density distribution is -like, the concentration is differentiated because there are few changes in the density 7 flame, and regions with small differential values are extracted.

In the case of digital images, the differential is actually determined as a difference. A horizontal difference o(x, y) and a vertical difference v(x, y) are determined by the one-differential operator shown in FIGS. 5(A) and (8). FIG. 5(A) is an operator for differences in the horizontal direction, and FIG. 5(B) is an operator for differences in the vertical direction. If +(x, y) is the density value falling at the point (x, y) in the original image, then u (x,
y) and v (x, y) are as follows.

1-1(x,y)-1(x+I,y-1)+21(x+
I,y)+I(x+1.y+1)(1(x-1,y-1
)+21(x-1,y)+I(x-1,y+1))-(
1) V(x,y)-1(x-1,y-1)+21(x,
y-1)O(x+1.y-1)-(I(x-1,y+1
)+2I(x,y+1)+I(x+1.y+1))...
・(2) From this, the size of the concentration gradient is E(x, y)a
(x, y) = N votes y)'+V(x, y)' −
(3) becomes.

The size of the concentration gradient (edge strength) by one differential
(x, y) is obtained. In order to separate the uniform density area from the background in the image, the following inversion binarization process is performed.

In the inverted binarized image, in addition to the cross-sectional area and the above-mentioned void areas (2) and (2), even small areas with locally uniform density are detected, so areas smaller than a predetermined area are removed. Since the detected regions are not connected to each other and are independent regions, various feature quantities can be calculated for each region.

(2) Discrimination based on feature values Since the processed image obtained in (1) is a binary image, the sum (AND) of logical operations is performed for each IM of the original image and binary image.
A uniform density region is cut out, a feature amount is calculated for each region, and each region is discriminated based on the difference in feature amount between the cross-sectional region and the void region.

(2) Discrimination based on the average density of the area By removing areas with a small average density, dark-looking areas deep within the void area (2) are removed.

■ Discrimination based on the peak width of the concentration distribution in the area When determining the concentration distribution in the area, since the cross-sectional area is a plane, the concentration distribution is uniform and has a sharp peak, but the concentration distribution in the void area ■ is gently sloped. Areas with a more dispersed distribution have a more dispersed distribution. As shown in FIG. 6, the half-value width W2 or the l/n value width wn is defined as a characteristic quantity of the concentration distribution, and the cross-sectional area and the void area are determined based on these values.

(2) Discrimination based on edge strength of area contour The density is uniform in the cross-sectional area, but the density changes at places where the cross-sectional area is interrupted. This location is called a contour. Since the concentration change is rapid at the contour of the cross-sectional area, 1
The magnitude of the concentration gradient (edge strength) when subdifferentiated is large, and the gap region (3) has a gentle slope, so the edge strength is small in that contour, so it can be determined by the edge strength of the contour.

■ Discrimination based on the shape factor of the region Looking at the shape of each region, the cross-sectional region has a smooth curved outline, while the void region ■ has a sharply uneven contour. are defined and determined based on their values.

Shape factor 1 - Perimeter length / (Perimeter length of circumscribed figure) or Shape factor 2 - 4π x Area / (Perimeter length) 2 The criteria for determining each section and void area is the representative number as follows. It is set experimentally from several images.

In measuring the porosity of a polyethylene sintered body in this example, several representative images of the sintered body sample are selected, and after detecting uniform density regions in each image, the shape factor of each region is calculated. Regarding factor 1, the area where the value is large is the void area, and the area where the value is small is the cross-sectional area, so the intermediate value between them is used as the criterion.

[Effects of the Invention] As explained above, in the present invention, a cross-sectional area is separated and extracted from an image obtained by photographing a cross-section of a porous body, and the area ratio of these areas is calculated by treating the other areas as void areas. By calculating this, the void area of the porous body can be measured with high precision.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing a cross section of a porous body, Fig. 2 is a configuration diagram of a measuring device in an embodiment of the present invention, Fig. 3 is a flowchart showing a processing procedure in an embodiment of the present invention, and Fig. 4 is a diagram showing the implementation of the present invention. FIG. 5 is a diagram showing a processed image in the example. FIG. 5 is a diagram showing the first-order operator in the embodiment of the present invention. FIG. 6 is a diagram showing the density distribution in the embodiment of the present invention. 1... Optical microscope, 2... CCO (charge coupled device) camera, 3... SE
M (scanning electron microscope), 4...IP (image processing device), 5...high definition monitor, 6...hard copy 7...optical disk, 8...hard disk, 9...EWS (engineering), lO...monitor, 11...printer. Ng work station'#t, shadow 1 匈tata pLl''E い杢 θ hitting surface をj (Fig. 1-- ≧ 1 prisoner showing a night in public 3. Relationship with the case of the person making the amendment Patent applicant Asahi Kasei Kogyo Co., Ltd. 4. Agent address: 3rd floor, Seiko Building 6, 5-1-31 Akasaka, Minato-ku, Tokyo 107 (all date of delivery: December 1988) 20th) 6. Column 7 of "4 Brief Description of Drawings" of the specification to be amended, Contents of the amendment (1) "Diagrams showing processed images" on page 14, lines 15 to 16 of the specification is corrected to "a photograph showing a processed image of a particulate structure." (2) Figure 4 of the drawings is corrected as shown in the attached sheet (no change in content). ``2''

Claims (1)

[Claims] 1) A step of imaging a cut surface obtained by cutting a porous body at a predetermined plane, and a step of separating and extracting a cross-sectional region and a cross-sectional region of the void region in the image obtained by the imaging. A method for measuring the porosity of a porous body, the method comprising the steps of: calculating the area ratio of these regions by defining a region other than the cross-sectional region as a void region.