JPH0513258B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to an electron beam diffraction image analysis device, and in particular, an electron beam diffraction image automatic measurement device that can process an electron beam diffraction image to automatically measure lattice spacing, surface index, etc. with high precision and high speed. It is related to the device. When observing the microstructure of a metal using an electron microscope, precipitates and matrices can be identified by analyzing electron beam diffraction images. Conventionally, when performing this analysis, the diffraction image is photographed, the photograph is enlarged and projected using an optical magnifying device such as a projector, and the distance between the transmitted beam spot and the diffraction spot is calculated on this projected image. In order to analyze the electron beam channeling pattern, the angle between the diffraction spot and the transmission spot, the radius of the diffraction ring, and even the distance between lines and the angle formed by the lines are manually measured. Such manual measurements require a great deal of effort and time, and are extremely inefficient. In addition, analysis can only be performed after taking a photograph, and it is impossible to identify substances while observing them, which has many disadvantages in electron microscopic observation. In order to improve these problems, an automatic electron diffraction image analysis system has been developed, and was presented, for example, at the 39th Academic Conference of the Japanese Society of Electron Microscopy, page 6 (May 1981). However, this automatic analysis system uses an expensive high-sensitivity television imaging device called an image carrier to improve accuracy, making the entire system expensive. Furthermore, although the time required for analysis is on the order of several minutes, in practical terms it is necessary to process on the order of several seconds when contamination of the sample is taken into account. Furthermore, since a high-sensitivity television imaging device is used, if noise occurs for some reason, it will also be captured at high sensitivity, which can have a serious impact on analysis and is cumbersome to handle. There are drawbacks. The purpose of the present invention is to eliminate the above-mentioned drawbacks, to be able to analyze electron beam diffraction images with high precision without being affected by noise, to process quickly in the order of several seconds, and to have a simple configuration. The purpose of this invention is to provide a simple and inexpensive automatic electron diffraction image measuring device. The automatic measuring device for electron beam diffraction images of the present invention includes an electron microscope that forms an electron beam diffraction image of a sample, and an analog image signal of a plurality of images by capturing the electron beam diffraction image multiple times under the same conditions. an imaging device that
An analog-to-digital conversion unit capable of converting this analog image signal into a multi-gradation digital image signal and removing noise by adding and averaging the digital image signals of multiple images for each pixel, and this analog-to-digital converter. an image processing section that extracts the features of image elements including points, circles, and lines characteristic of the electron beam diffraction image from the digital image signal output from the section to obtain a diffraction image; An arithmetic processing unit that calculates basic data including angle, radius, and line distance, and based on the basic data obtained by this arithmetic processing unit, calculates the crystal structure and crystal structure of the sample by comparing it with pre-stored Miller index and interplanar spacing. The present invention is characterized in that it includes an analysis section that analyzes orientation. The present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram showing the overall configuration of an automatic electron beam diffraction image measuring device according to the present invention. Image input section 1
comprises an electron microscope 1a and an imaging device 1b for capturing an electron beam diffraction image projected on a fluorescent plate, and supplies an analog image signal output from the imaging device 1b to an analog-digital converter 2.
This analog-to-digital converter 2 is a characteristic feature of the device of the present invention, and converts analog image signals into digital signals at a multi-tone level of 64 or more levels, preferably 256 levels, and converts images of two or more screens at high speed. These images can be superimposed and converted into digital images. Since this multi-tone digital image signal is obtained by quantizing brightness and darkness ranging from black to white, it is also referred to as a gray-level digital image signal in this specification. That is, several screens are captured one after another by the imaging device 1b, and digital image signals for one screen can be obtained by averaging the gray level digital values of each pixel of these screens. Such an analog-to-digital converter 2 includes, for example, an analog-to-digital converter, a gray memory capable of storing gray-level digital values for one screen,
For each pixel, the circuit is configured to calculate the average value of the digital value output from the analog-digital converter and the digital value stored in the gray memory, and to store this average value in the gray memory. can do. Or also,
The analog-digital converter 2 may be configured to determine that signals having a difference of more than a certain value for the same pixel are noise, and perform superimposition to remove the signals. In this specification, the analog-to-digital conversion section is assumed to be capable of performing such analog-to-digital conversion and superimposition of screens. By using the analog-to-digital converter 2 capable of such superimposition, it is possible to suppress noise in diffraction images with very low brightness and contrast, and to emphasize gray information at intermediate brightness.
In other words, since the original image appears in the same position every screen, it becomes a large value when added together, whereas noise occurs randomly and does not appear in the same position every screen, so it does not become a large value when averaged. . In this way, an electron diffraction image with extremely low brightness and contrast can be input with high precision without being affected by noise. Therefore, it has become possible to perform high-speed analysis of low-contrast electron diffraction images, which were impossible to analyze using conventional equipment. The image signal converted into a digital quantity with the brightness difference, that is, the contrast emphasized as described above, is supplied to the image processing section 3, where a transmitted spot and a circular or point-like diffraction image are found. Furthermore, line elements are extracted in the channeling pattern or Kikuchi pattern. This image processing is also a feature of the present invention, and the processing method will be explained below. First, the case of processing a spot-like diffraction image will be explained. In this case, first find the center of the transmission spot. Since this transmission spot appears as a density gap with a certain extent, its center of gravity is temporarily set as the center point of the transmission spot. On the other hand, diffraction spots can be detected separately from transmission spots by utilizing the characteristic that the spread of the density gap is relatively smaller than that of transmission spots. The center of gravity of the diffraction spot obtained in this way is regarded as the diffraction spot point. Next, a spot point near a point that is symmetrical to the tentatively determined transmission spot point is selected. Several sets of diffraction spot points that are symmetrical to the tentative transmission spot points obtained in this way are found, and the intersection of the straight lines connecting these sets of diffraction spot points is determined as the true transmission spot point. Transmission spot points and diffraction spot points can be found in the manner described above. Furthermore, in the case of a ring-shaped diffraction image, there is a very large density gap on the ring, so this is detected, these points are regarded as points on the ring, and circular approximation is performed. In addition, in the case of channeling or Kikuchi patterns, line element tracking filter processing is used to
Extract Kikuchi lines, which are several sets of black and white parallel lines. Generally, the Kikuchi pattern has extremely low contrast, but in the apparatus of the present invention, since several screens are superimposed as described above, the contrast is emphasized and the Kikuchi line can also be detected accurately. Image processing is performed as described above to extract transmission spot points, diffraction spots, diffraction rings, and Kikuchi lines, but the contrast of the image signal supplied from the analog-digital converter 2 is emphasized,
Since noise is suppressed and the S/N ratio is high, image processing can be performed in a short time, and the image processing time is about 3 to 15 seconds. Next, the diffraction image extracted by the image processing section 3 is sent to the arithmetic processing section 4, where it is processed to determine the distance between the transmission spot and the diffraction spot, the angle between the diffraction spots, the radius of the ring, and the distance between the Kikuchi lines. Basic data such as the distance between is calculated. Next, the basic data regarding the diffraction image obtained by the arithmetic processing section 4 in this manner is sent to the analysis section 5. The analysis section 5 determines parameters such as surface spacing and surface index from these basic data, and identifies the sample material based on these parameters.
The configuration is such that the measurement results, basic data, parameters, etc. can be outputted by the output section 6 as appropriate. Next, a measurement example performed by the above-mentioned automatic electron diffraction image measurement apparatus of the present invention will be described. FIG. 2 shows the result of circular approximation by the image processing section 3 of a ring-shaped electron beam diffraction image obtained using an Au vapor deposited film as a sample. As a result of processing this circular approximation pattern by the arithmetic processing section 4 and the analysis section 5, the lens length of the electron microscope 1a in this case is
It turned out to be 556.1mm. The processing time for such a ring-shaped electron beam diffraction image was completed in about 1 second. Further, by processing such a ring-shaped electron beam diffraction image, the lattice spacing of each crystal constituting the polycrystal can be determined. FIG. 3 shows the results of processing an electron beam diffraction pattern obtained using a single crystal of pure iron as a sample in the image processing section 3 to find transmission spot points and diffraction spot points. As mentioned above, the transmission spot points are the intersections of straight lines connecting pairs of diffraction spot points. Next, the distance between the transmission spot point and the diffraction spot point and the angle between the diffraction spot points are determined from the result of image processing in this way in the arithmetic processing unit 4, and the analysis unit 5 calculates the interplanar spacing and index of pure iron. The results are shown in Tables 1 and 2, respectively. The transmission spot O and the diffraction spot P ₁ to _{P 4} representing the interplanar spacings r ₁ to r ₄ and angles in these tables are shown in FIG. 4, which correspond to those shown in FIG. 3. ing.

【table】

[Table] As is clear from Tables 1 and 2 above, the measured values and the theoretical values almost match, indicating that it is possible to identify substances with high accuracy. Note that this process was completed in about 8 seconds. FIG. 5 shows the results of processing a Kikuchi pattern obtained from a steel thin film as a sample in the image processing unit 3. Further, FIG. 6 shows the orientation analyzed by the arithmetic processing section 4 and the analysis section 5 based on the image shown in FIG. 5. This process finished in about 10 seconds. As mentioned above, according to the device of the present invention, surface spacing, angle, index, direction, etc. can be measured at high speed and with high precision, but this is largely due to the overlapping of the screens in the analog-digital converter 2. This effect is shown in FIGS. 7 and 8. That is, FIG. 7 shows the Kikuchi pattern when the two screens are not superimposed, and FIG. 8 shows the Kikuchi pattern when the two screens are superimposed. As is clear from comparing these images, by overlapping two or more screens, noise is removed and the contrast between light and dark is emphasized, resulting in a sharp image. The invention is not limited to the embodiments described above, but can be modified in many ways. For example, in the above-mentioned embodiment, the electron beam diffraction image formed on the fluorescent plate of the electron microscope was directly imaged by the imaging device. The image may be captured using an imaging device. Further, the number of screens to be superimposed by the analog-digital converter may be arbitrary as long as it is two or more, and in practice it is preferable to configure the system so that the optimum number can be selected by determining various situations. As is clear from the above, according to the automatic electron diffraction image measurement device of the present invention, the image signals obtained by repeatedly capturing electron diffraction images with an imaging device are superimposed and converted into digital signals. A high S/N image with the influence of noise removed is obtained, and by processing this in the image processing section, transmission spot points, diffraction spot points, diffraction rings,
Kikuchi lines, etc. can be extracted at high speed and with high precision. Based on this, useful data such as surface spacing, surface index, and orientation can be obtained in a short time, making it possible to identify materials with high precision. can be done. Furthermore, since an image pickup device with normal sensitivity can be used, the entire device can be constructed at low cost and is easy to handle.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the configuration of one embodiment of the automatic electron diffraction image measuring device of the present invention, and Fig. 2 is a ring-shaped diffraction image obtained using a metal structure made of a deposited Au film as a sample. Figure 3 shows the results of processing with the apparatus. Figure 3 shows the results of processing a spot-shaped diffraction image obtained from a metal structure made of a single crystal of pure iron. Figure 4 shows the method for measuring interplanar spacing and angle. Figure 5 is a diagram showing the results of processing a Kikuchi pattern obtained from a metal structure made of a thin film of steel. Figure 6 is a diagram showing the analysis results of the Kikuchi pattern. Figure 7 shows a Kikuchi pattern obtained using a metal structure made of a steel thin film as a sample without superimposition, and Fig. 8 shows a Kikuchi pattern obtained using a metal structure made of a steel thin film as a sample when two images are superimposed. FIG. 1... Image input section, 1a... Electron microscope, 1b
...imaging device, 2...analog-digital conversion section, 3...image processing section, 4...arithmetic processing section, 5...
...Analysis section, 6...Output section.

Claims (1)

[Claims] 1. An electron microscope that forms an electron diffraction image of a sample; an imaging device that captures the electron diffraction image multiple times under the same conditions and outputs analog image signals of multiple images; an analog-to-digital converter capable of converting into a digital image signal of the same color and removing noise by adding and averaging the digital image signals of a plurality of images for each pixel; and a digital image output from the analog-to-digital converter. an image processing unit that obtains a diffraction image by extracting features of image elements including characteristic points, circles, and lines in the electron beam diffraction image from the signal;
An arithmetic processing unit that calculates basic data including distances between points, angles, radii, and line distances of this diffraction image, and a comparison with pre-stored mirror indices and surface spacing based on the basic data obtained by this arithmetic processing unit. 1. An automatic electron beam diffraction image measurement device comprising: an analysis section for analyzing the crystal structure and crystal orientation of a sample.