JPS6319086A - Particles number counting method - Google Patents

Abstract

PURPOSE:To eliminate the need for a complicated processing by referring to the label of an adjoining picture element, labelling respective picture elements, executing the calculation based upon an adjoining condition for respective labels and counting the number of the particles. CONSTITUTION:A binary image is successively scanned and labelled to respective picture elements in accordance with the label of the adjoining picture element. The adjoining label relation between respective stuck labels is preserved at a label unit. The number of the pair, in which two labels different from the type of the label used for labeling are adjoining, and the difference are counted and the counting value is made into the number of the label.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to particles (connected figures) used when analyzing or inspecting the three-dimensional structure of an object based on information of a binary image of a cross section or a projection plane. Concerning counting methods.

[Prior art]

As a method to count the number of particles included in a binary image,
A method of degenerating connected graphic pixels into one point (a degeneracy operation for preserving topology) is known. FIG. 9 is a schematic diagram showing a part of a binary image, and three particles (
Connected figures) (X).

(y) and (z) exist. Here, a particle (connected figure) refers to a figure formed by connecting pixels representing a figure. Then, the graphic pixels constituting the particles (x), (y), and (z) are sequentially degenerated in the direction indicated by the arrow in the figure.
In the end, the force, which is approximately one pixel in the middle of each particle, is
(See FIG. 10). Next, as a result of degenerating the figure pixels, the number of remaining pixels is calculated to the 10th
In the figure, there are three (force, force, and force), and the counted value is the number of particles.

In addition to the above-mentioned method, there is also a method of calculating the number of particles by performing circuit processing on images of two adjacent scanning lines. 11th
The figure is a schematic diagram showing a part of a binary image, and three particles (connected figures) (s), (t),
(u) exists. FIG. 12 is a circuit diagram showing a circuit for implementing this method. First, a binary image is scanned in order from the top line to obtain an image signal A by line scanning, which is input to the counter 10a. The line scanning image signal A is also input to a delay device 11, from which an image signal delayed by one line from the input signal is obtained.

This delayed signal and the image signal A are input to the AND gate 12. The AND gate 12 calculates the logical product of both image signals, and the logical product image signal B is input to the counter 10b. Next, the counters 10a and 10b count the number of "1" (high level) portions of the image signals A and B in each line, and calculate the counted value as an+bl (n
is the scanning line number). After scanning an appropriate number of lines, the subtractor 13 subtracts the total number of bn from the total number of an, and the subtracted value is used as the number of particles.

In FIG. 11, the eighth line from the top line (that is, n
-8), the image signals A and B become as shown in FIG. 13, and the count value a of each line. .

Similarly, bn is as shown in FIG. Count value an,
The total number of bn is 8 and 5, respectively, so the number of particles is 8-5=
It becomes 3.

[Problem that the invention seeks to solve]

Figure 14 is a schematic diagram of a binary image, similar to Figure 9, showing a particle (W) with a hole (background pixel) [h] inside.
It is shown. By the way, if the above-mentioned method for degenerating graphic pixels is applied to such a holed particle, the periphery of the hole will remain in a loop shape, as shown in Figure 15, and one point (one pixel) will be There were disadvantages in that degeneracy was not possible and the number of particles could not be measured.

For particles with holes, there is a method of degenerating the graphic pixels while filling the holes, or by cutting and separating a part so that the shape is not annular.

However, in any case, there is a problem in that real-time processing is impossible because it is a degenerate process and it is an iterative process. Another problem is that a large amount of memory is required to store the degenerate images.

On the other hand, with the second method, real-time processing is possible, but for particles with complex shapes (for example, perforated particles), it is difficult to distinguish between newly appearing components and existing same components during particle determination. There was a point.

The present invention has been made in view of the above circumstances, and its purpose is to label each pixel of a binary image while scanning it. Scanning and labeling are performed based on the labels of adjacent shapes that have already been added, and during this labeling, the labels of the shape pixels adjacent to that label are saved for each attached label, and then the labels are calculated based on the number of label types used. By reducing the number of adjacent pairs of two types of labels and using the difference as the number of particles, even particles with holes do not require complicated processing, require a small amount of memory, and can be processed in real time. The objective is to provide a number counting method.

[Means for solving problems]

The particle number counting method according to the present invention labels the pixels representing the figure with the number of particles formed by connecting the pixels representing the figure in a binary image consisting of pixels representing the figure and pixels representing the background. In the counting method, the labeling includes:
While sequentially scanning each pixel of the binary image, check the label of the adjacent pixel in the scanning method, and if the adjacent pixel is labeled, give it the same label as the adjacent pixel, and if it is not labeled, do not use it. The adjacency relationship between each attached label is saved for each label, and the difference between the number of types of labels used for labeling and the number of adjacent pairs of different labels is calculated. It is characterized in that the value is the number of particles.

[Effect]

In the present invention, a binary image is sequentially scanned, each pixel is labeled according to the label of an adjacent pixel, adjacent labels are saved for each attached label, and the label used is based on the saved state. Subtract the number of adjacent pairs of two types of labels from the number of types. Then, the subtracted value becomes the number of particles.

〔Example〕

The present invention will be explained below based on drawings showing embodiments thereof.

FIG. 1 is a schematic diagram of a part of a binary image, in which the dark pixels are figure pixels and the white pixels are background pixels. In FIG. 1, (a) and (b).

There are six particles (connected figures): (c), (d), (e), and (f).

First, each line in the figure is sequentially scanned in the horizontal direction (from left to right) starting from the top line, and the pixels of the figure are labeled. The top line has no graphics, so no pixels are labeled. Then, as soon as each pixel is labeled, the attached label is stored in the label buffer. FIG. 2 shows the contents of the label buffer of the top line after the top line has been scanned. Note that FF is not labeled i
! This shows the contents of the label buffer of i elements, and since none of the pixels on the top line are labeled, all the pixels in the label buffer shown in FIG. 2 are FF.

FIG. 3 is a schematic diagram showing the positional relationship of adjacent pixels to be examined when labeling a pixel, and the pixel marked O in the figure is the pixel to be processed for labeling. Considering the scanning direction, pixels A, A, T, and D have been labeled after scanning, and the attached labels are stored in the label buffer.

Pixel O is a figure, and the neighboring four pixels (A.

If a label other than FF, which is a figure, is stored in the label buffer for any one of (a, tsu, d), that label is used as the label of pixel O and is labeled on pixel O.

Here, the priority pixel order of the label buffer that is referred to when labeling the pixel is set.

It is checked whether it is a figure in the order of pixel work, A, A, and 3. If it is a figure, the contents of the label buffer are used as the label of pixel O. On the other hand, if pixel O is a figure and all four pixels are background pixels, a new unused label is used as the label of pixel O and the pixel O is labeled.

Then, the label attached to the pixel O is stored in the label buffer.

If pixel O is a background pixel, pixel O is not labeled. However, if pixel O is a background and pixel A is a figure, the contents of the label buffer of pixel A are saved as the contents of the label buffer of pixel O. Further, if pixel O is a background and pixel A is also a background, the content of the label buffer of pixel O becomes FF.

In other words, the label of each pixel is the label of the adjacent pixel after processing if a label is attached, and an unused label if there is no label. The label of the labeled pixel is then stored in a label buffer, and the label stored in this label buffer is used as the label of the adjacent pixel during the above-mentioned labeling.

In addition, here, if pixel A is a figure and the top of the pixel is the background,
Since the contents of the label buffer of pixel A are saved as the contents of the label buffer on the pixel, even if the contents of the label buffer to be referenced are three pixels (pixel priority order, A, T), pixel O has the same contents as pixel A. is labeled. Therefore, the contents of the label buffer to be referenced are actually 3 pixels (pixel I).

It will be a good thing.

On the other hand, this labeling is performed in parallel with the label adjacent state (
If there is an adjacent label, the adjacent optical label is stored in an adjacent label memory (not shown).

FIG. 4.5 is a schematic diagram of the label when scanning and labeling up to the seventh line in FIG. 1 is completed, and a schematic diagram of the contents of the label buffer for the seventh line. FIG. 6 is a schematic diagram showing the contents of the adjacent label memory saved when scanning and labeling up to the seventh line are completed. For example, in parallel with the process of attaching the label C to the pixel shown in Figure 4, it is checked that the pixel is adjacent to the already labeled pixel (label A), Label A is stored and saved in the adjacent label memory as an adjacent label.

From the eighth line onward, the above-described labeling process and adjacent label storage process are also performed every time each pixel is scanned. Figure 7 is a schematic diagram showing the state in which all the pixels in Figure 1 have been labeled, and Figure 8 is a schematic diagram of the adjacent label memory when all pixels in Figure 1 have been scanned and added. It is a schematic diagram showing the contents.

Next, the number of particles is counted based on the contents stored in the adjacent label memory. According to the adjacent state shown in FIG. 8, first, the number of types of labels used is A, B, and C.

There are 12 pieces: D, E, G, H, 1, J, L, M.
Also, the number of adjacent pairs of two types of labels is (A, C).

(C,D), (G,H), (H,I), (
J, K).

There are 6 pairs (L, M). Therefore, the number of particles in FIG. 1 is 12-6=6 (pieces), and the number of particles can be counted.

〔effect〕

As described in detail above, in the present invention, while scanning each pixel, each pixel is labeled by referring to the label of the adjacent pixel, and the adjacent state of each label is saved. Since the number of particles is counted by calculating the number of pairs of adjacent labels (number of pairs of adjacent labels), even if particles with complicated shapes (for example, particles with holes) are present, complicated processing unlike conventional methods is not necessary.

Furthermore, there is no need to store a large amount of image content, and the memory capacity can be small.

Furthermore, the present invention has excellent effects, such as being able to count the number of particles in the time it takes to scan all pixels just once, and allowing real-time processing.

[Brief explanation of the drawing]

1st. 9.11.14 is a schematic diagram of a binary image, Figure 2 is a schematic diagram of the contents of the label buffer on the top line of Figure 1,
Fig. 3 is a schematic diagram of an image for explaining labeling in the method of the present invention, Fig. 4 is a schematic diagram of labels up to line 7 in Fig. 1, and Fig. 5 is a label buffer for line 7 in Fig. 1. 6 is a schematic diagram showing the contents of the adjacent label memory at the end of scanning the 7th line in Figure 1. Figure 7 is a schematic diagram of the label when all pixels in Figure 1 have been scanned. , 8th
The figure is a schematic diagram showing the contents of the adjacent label memory when all pixels in Figure 1 have been scanned, Figures 10 and 15 are schematic diagrams of images with particles degenerated using the conventional method, and Figures 12 and 13 are They are a circuit diagram and an image signal diagram of a conventional method. Agent Patent Attorney Noboru Kono

Claims (1)

[Claims] 1. Labeling the pixels representing the figure with the number of particles formed by connecting the pixels representing the figure in a binary image consisting of pixels representing the figure and pixels representing the background. In the counting method, the labeling includes sequentially scanning each pixel of the binary image and checking labels of scanned adjacent pixels;
If the adjacent pixel is labeled, it is given the same label as the adjacent pixel, and if it is not labeled, it is given an unused label, and the adjacency relationship between each attached label is saved for each label, A particle number counting method characterized by calculating the difference between the number of types of labels used for labeling and the number of adjacent sets of different labels, and using the calculated value as the number of particles.