JP2020165918A - Track detection device, track detection method, and track detection program - Google Patents

Abstract

To provide a track detection device, a track detection method, and a track detection program that can reduce an amount of calculation for detecting a track without lowering detection accuracy.SOLUTION: A track detection device 12 includes: a tomographic image acquisition unit 20 that acquires a tomographic image group 56 that includes a tomographic image 571 obtained by imaging each of a plurality of cross sections of an emulsion layer 51 having different depth directions, and obtained by a cross section of the emulsion layer 51 closest to the incident side of μ particles, and a tomographic image 57n obtained by a cross section of the emulsion layer 51 farthest to the incident side of the μ particles, wherein the images each have a plurality of pixels 58 disposed. The track detection device 12 includes: a track candidate line identification unit 24 that identifies a plurality of track candidate lines C that each pass through a reference point 60 of the tomographic image 571 in different directions and each pass through different pixels 58 of the tomographic image 57n; and a detection unit 26 that detects a track T in the emulsion layer 51 from the plurality of track candidate lines C.SELECTED DRAWING: Figure 5

Description

The present disclosure relates to a track detection device, a track detection method, and a track detection program.

Conventionally, in non-destructive inspections, etc., it has been possible to detect the tracks of charged particles such as muons contained in cosmic rays by using a nuclear dry plate containing a charged particle track recording layer that records the tracks of charged particles as latent images. (See Patent Document 1).

In the technique described in Patent Document 1, after developing a nuclear dry plate, the charged particle track recording layer of the nuclear dry plate is continuously imaged while changing the depth of the focal plane, and a plurality of tomographic images are acquired. Then, in the technique described in Patent Document 1, a distribution of indicators indicating the existence of tracks obtained by superimposing tomographic images is obtained by a so-called shift method in which a plurality of tomographic images are shifted by a shift amount according to the angle of the tracks. Based on this, the track is detected.

Japanese Unexamined Patent Publication No. 2010-101676

In the above-mentioned shift method, each tomographic image is shifted by a shift amount according to the track angle of the track assumed as a track of charged particles, and the tracks having the corresponding track angle have substantially the same x and y coordinates on each tomographic image. Tracks are detected based on the presence or absence of black spots at the same coordinates of each tomographic image.

However, when the above-mentioned shift method is applied to all the expected track angles of the track, the amount of calculation may become enormous. On the other hand, if the track angle of the assumed track is simply reduced, there is a concern that the track detection accuracy will decrease.

The present disclosure has been made in view of the above circumstances, and is a track detection device, a track detection method, and a track detection program that can reduce the amount of calculation for detecting a track without lowering the detection accuracy. The purpose is to provide.

In order to achieve the above object, the track detection device of the first aspect of the present disclosure is a track in a charged particle track recording layer that records the track of a charged particle as a latent image by transmitting the incident charged particle. It is a track detection device that detects, and is obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions, and by the cross section of the charged particle track recording layer closest to the incident side of the charged particles. Acquire a plurality of tomographic images in which a plurality of pixels are arranged, including the obtained first tomographic image and the second tomographic image obtained by the cross section of the charged particle track recording layer farthest to the incident side of the charged particle. The tomographic image acquisition unit, the track candidate line identification unit that specifies a plurality of track candidate lines that pass through different pixels of the second tomographic image in different directions, and the charged particles. It is provided with a detection unit that detects tracks in the track recording layer from a plurality of track candidate lines.

According to the track detection device of the first aspect, it is possible to reduce the amount of calculation for detecting the track without lowering the detection accuracy.

In the track detection device of the second aspect of the present disclosure, in the track detection device of the first aspect, the track candidate line identification portions have different directions of passing through the reference point, and each of them passes through the second tomographic image. Select a track candidate line from the virtual track candidate lines of.

According to the track detection device of the second aspect, it is possible to easily identify the track candidate line.

Further, in the track detection device of the third aspect of the present disclosure, in the track detection device of the second aspect, the track candidate line identification unit uses the same pixel in the second tomographic image among the plurality of virtual track candidate lines. When there are multiple virtual track candidate lines that pass through, only one of the virtual track candidate lines is included in the plurality of track candidate lines, and when there is one virtual track candidate line that passes through the same pixel in the second tomographic image, the same pixel Include virtual track candidate lines that pass through in multiple track candidate lines.

According to the track detection device of the third aspect, it is possible to more appropriately reduce the amount of calculation for detecting the track without lowering the detection accuracy.

Further, the track detection device of the fourth aspect of the present disclosure is the track detection device of the third aspect, in the case where the track candidate line identification unit has a plurality of virtual track candidate lines passing through the same pixel in the second tomographic image. Of a plurality of virtual track candidate lines passing through the same pixel, a virtual track candidate line passing through the position closest to the center of the same pixel is defined as one virtual track candidate line.

According to the track detection device of the fourth aspect, the detection accuracy can be improved.

Further, the track detection device according to the fifth aspect of the present disclosure is the track detection device according to any one of the first to fourth aspects, and the track candidate line identification unit is formed by a plurality of track candidate lines. When the angle is equal to or less than a predetermined angle, at least one track candidate line is excluded from the specified plurality of track candidate lines.

According to the track detection device of the fifth aspect, the amount of calculation for detecting the track can be further reduced without lowering the detection accuracy.

Further, the track detection device of the sixth aspect of the present disclosure is the track detection device of the fifth aspect under the condition that the in-plane angle parallel to the plane on which the charged particles of the charged particle track recording layer are incident is large. One of the conditions that is considered to be small is predetermined as a deletion condition, and the track candidate line identification unit specifies a plurality of track candidates that satisfy the deletion condition among a plurality of track candidate lines. Exclude from the candidate line.

According to the track detection device of the sixth aspect, it is possible to identify the track candidate lines that are regularly arranged by thinning out the specified track candidate lines.

Further, in the track detection device of the seventh aspect of the present disclosure, in the track detection device of the fifth aspect, the track candidate line identification unit has a plurality of track candidate lines having the same zenith angle, and the angles formed by each other are predetermined. When there are multiple track candidate lines that are less than or equal to the specified angle, the distribution of the remaining multiple track candidate lines can be determined by excluding some of the specified virtual track candidate lines. Is the state closest to the rotational symmetry with respect to the axis of 0 degrees.

According to the track detection device of the seventh aspect, the track candidate lines arranged more regularly can be specified by thinning out the specified track candidate lines.

Further, the track detection device according to the eighth aspect of the present disclosure is the track detection device according to any one of the first to seventh aspects, and the detection unit is a predetermined number of adjacent tomographic images. Tracks are detected from the track candidate lines based on the track images corresponding to the latent images included in the plurality of synthetic tomographic images obtained by combining the above.

According to the track detection device of the eighth aspect, the amount of calculation for detecting the track can be further reduced without lowering the detection accuracy.

Further, in the track detection device of the ninth aspect of the present disclosure, in the track detection device of the eighth aspect, the synthetic tomographic image shows the adjacent faults according to the direction in which the track candidate line is incident on the charged particle track recording layer. It is an image synthesized by shifting the positions of the images.

According to the track detection device of the ninth aspect, the detection accuracy can be further improved.

Further, the track detection method according to the tenth aspect of the present disclosure is a track detection method for detecting tracks in a charged particle track recording layer that records the tracks of charged particles as latent images by transmitting incident charged particles. The first fault obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions and obtained by the cross section of the charged particle track recording layer closest to the incident side of the charged particle. A plurality of tomographic images in which a plurality of pixels are arranged are acquired, including an image and a second tomographic image obtained by a cross section of a charged particle track recording layer farthest on the incident side of the charged particle, and a first tomographic image is obtained. A process of identifying a plurality of track candidate lines that pass through different pixels of the second tomographic image and each passing through a different pixel of the second tomographic image, and detecting tracks in the charged particle track recording layer from the plurality of track candidate lines. including.

Further, the track detection program according to the eleventh aspect of the present disclosure can be applied to a computer that detects tracks in a charged particle track recording layer that records the tracks of charged particles as latent images by transmitting incident charged particles. The first tomographic image obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions and obtained by the cross section of the charged particle track recording layer closest to the incident side of the charged particle, and the charge. A plurality of tomographic images in which a plurality of pixels are arranged are acquired, including a second tomographic image obtained by a cross section of the charged particle track recording layer farthest on the incident side of the particles, and a reference point of the first tomographic image is set. In order to execute a process of identifying a plurality of track candidate lines each passing through different pixels of the second tomographic image and detecting tracks in the charged particle track recording layer from the plurality of track candidate lines. belongs to.

Further, the track detection device of the present disclosure has a processor and a memory, and a track that detects a track in a charged particle track recording layer that records the track of the charged particle as a latent image by transmitting the incident charged particle. A detection device, obtained by a processor imaging each of a plurality of cross sections of a charged particle track recording layer having different depth directions, and obtained by a cross section of the charged particle track recording layer closest to the incident side of the charged particle. A plurality of tomographic images in which a plurality of pixels are arranged are acquired, including the first tomographic image obtained and the second tomographic image obtained by the cross section of the charged particle track recording layer farthest on the incident side of the charged particle. , The directions passing through the reference points of the first tomographic image are different from each other, and a plurality of track candidate lines each passing through different pixels of the second tomographic image are identified, and the tracks in the charged particle track recording layer are set to a plurality of track candidate lines. Detect from.

According to the present disclosure, it is possible to reduce the amount of calculation for detecting tracks without lowering the detection accuracy.

It is a side sectional view of an example of the nuclear dry plate of an embodiment. It is a block diagram which shows an example of the structure of the nondestructive inspection system of embodiment. It is a figure explaining the tomographic image generated from the nuclear dry plate of an embodiment. It is a block diagram which shows an example of the hardware configuration of the track detection apparatus of embodiment. It is a block diagram which shows an example of the functional structure of the track detection device of embodiment. It is a figure for demonstrating the track angle of a track. It is a figure for demonstrating the track passing through a plurality of tomographic images included in the tomographic image group obtained by an emulsion layer. It is a figure for demonstrating the example of the virtual track candidate line passing through a reference point. It is a figure for demonstrating the example of the track candidate line passing through a reference point. It is a flowchart which shows an example of the track detection process executed by the track detection apparatus of 1st Embodiment. It is a flowchart which shows an example of the track candidate line identification process executed by the track detection process shown in FIG. It is a flowchart which shows an example of the track candidate angle derivation process executed in the track candidate line identification process shown in FIG. It is a figure which shows an example of the list used in the track candidate angle derivation process shown in FIG. It is a figure which shows an example of the ID arrangement table used in the track candidate angle derivation process shown in FIG. It is a flowchart which shows an example of the detection process executed by the track detection process shown in FIG. It is a figure for demonstrating the detection of the track by the shift method in the detection process of 1st Embodiment. It is a figure for demonstrating the detection of the track by the shift method in the detection process of 1st Embodiment. It is a flowchart which shows an example of the track candidate angle derivation process executed in the track detection apparatus of 2nd Embodiment. It is a figure for demonstrating an example of exclusion (thinning out) of a track candidate line in track candidate angle processing of 2nd Embodiment. It is a figure for demonstrating an example of exclusion (thinning out) of a track candidate line in track candidate angle processing of 2nd Embodiment. It is a figure for demonstrating composition of a tomographic image. It is a figure for demonstrating composition of a tomographic image. It is a flowchart which shows an example of the detection process executed in the track detection process of 3rd Embodiment. It is a figure for demonstrating the synthetic tomographic image obtained by synthesizing a tomographic image. It is a figure for demonstrating the detection of the track by the shift method in the detection process of 3rd Embodiment. It is a figure for demonstrating the detection of the track by the shift method in the detection process of 3rd Embodiment.

Hereinafter, examples of embodiments for carrying out the technique of the present disclosure will be described in detail with reference to the drawings.

[First Embodiment]
In the non-destructive inspection using the nuclear emulsion 50, the nuclear dry plate 50 is installed around the inspection target, the track T of muons recorded on the nuclear dry plate 50 is detected during the installation, and the detected track T is analyzed. As a result, the density of the inspection target can be obtained.

FIG. 1 shows a side sectional view of an example of the nuclear dry plate 50 of the present embodiment. As shown in FIG. 1, the nuclear dry plate 50 of the present embodiment includes two emulsion layers 51A and 51B and one film base 52, and the emulsion layers 51A and 51B are coated on both surfaces of the film base 52. .. The film base 52 is made of, for example, plastic and has a thickness of 300 μm. Each of the emulsion layers 51A and 51B contains, for example, silver halide and has a thickness of 60 μm. The emulsion layers 51A and 51B of the present embodiment are an example of the charged particle track recording layer in the present disclosure, which records the track of the charged particle as a latent image by transmitting the incident charged particle. As the track recording material other than silver halide contained in the track recording layer, for example, an iron salt photosensitive material containing ferric oxalate or a diazo photosensitive material containing an aromatic diazonium salt can be used. .. In the present embodiment, when each of the emulsion layers 51A and 51B is generically referred to without distinction, it is referred to as "emulsion layer 51".

When the muons contained in the cosmic rays pass through the nuclear emulsion 50, the latent image along the track T of the muons is recorded in the emulsion layer 51 by the electrons generated by the ionization action in the emulsion layer 51. In the example of FIG. 1, μ particles are incident from the emulsion layer 51A side, and a track T is formed by the incident μ particles. In the example of FIG. 1, the portion of the track T recorded as a latent image in the emulsion layer 51 is indicated by a solid line, and the portion other than the portion recorded as a latent image in the emulsion layer 51 is indicated by a broken line. There is. The muon of this embodiment is an example of a charged particle according to the disclosed technique. The charged particles are not limited to muons. For example, the charged particles may be electrons or protons secondarily generated from γ-rays, neutron rays, or the like. Further, the charged particles are not limited to those that directly expose the nuclear emulsion plate 50. For example, the charged particles collide with and scatter the protons in the track recording material of the nuclear emulsion plate 50, and the protons are detected to form the tracks. It may be one.

First, the configuration of the non-destructive inspection system 1 of the present embodiment using the nuclear emulsion 50 will be described. FIG. 2 shows a block diagram showing an example of the configuration of the non-destructive inspection system 1 of the present embodiment. As shown in FIG. 2, the non-destructive inspection system 1 includes a tomographic image generation device 10, a track detection device 12, and an analysis device 14. The tomographic image generation device 10, the track detection device 12, and the analysis device 14 are each connected to the network N, and can communicate with each other via the network N. In the present embodiment, the tomographic image generation device 10, the track detection device 12, and the analysis device 14 have been described as separate devices, but two or all of these are combined into one device. Needless to say, it may be in the form of.

The tomographic image generation device 10 is a device that generates a tomographic image of the emulsion layer 51 of the nuclear emulsion 50 on which a latent image is formed by muons. By developing the nuclear dry plate 50 on which the latent image is formed as described above, the latent image remains in the nuclear dry plate 50 as silver particles (black spots) having a size of about 1 μm. The track T of the muon particles that have passed through the nuclear emulsion 50 is recorded as a three-dimensional arrangement of silver particles (black dots). The three-dimensional arrangement of the silver particles (black spots) is extracted by observation using an optical microscope. The black dots of this embodiment are an example of the track image of the present disclosure.

The tomographic image generation device 10 includes an optical microscope and an imaging device (both not shown). The objective lens of the optical microscope is arranged on the surface side where the muons of the emulsion layer 51A are incident. The tomographic image generator 10 continuously takes an image while moving the nuclear dry plate 50 in the in-plane direction of the emulsion layer 51 to take an image of the entire surface on which the muons of the emulsion layer 51 are incident. Further, the nuclear dry plate 50 is formed by moving the objective lens in the stacking direction of the emulsion layer 51 and the film base 52 in the nuclear dry plate 50 (vertical direction in FIG. 1, hereinafter referred to as “depth direction”). By changing the focal position, a plurality of images having different depths of the emulsion layer 51 are captured.

In this way, as shown in FIG. 3, the tomographic image generator 10 has different depth directions from the emulsion layer 51A and includes a plurality of pixels 58 (see FIG. 7), n (n is 2 or more). A tomographic image group 56A containing) (integer) tomographic images 57 _{1 to} 57 _n is generated. Also, from the emulsion layer 51B, the depth direction are different, n tomographic image group 56B including the tomographic image ₅₇ 1 to 57 _n are generated. Tomographic image 57 ₁ side is out of the tomographic image 57 ₁ and the tomographic image 57 _n, a closest cross section entrance side of the track T, as an example, corresponds to the uppermost layer of the tomographic image group 56. Further, the tomographic image 57 _n side is out of the tomographic image 57 ₁ and the tomographic image 57 _n, a farthest section on the entrance side of the track T, as an example, corresponds to the lowest layer of the tomographic image group 56. Tomographic image 57 ₁ of the present embodiment corresponds to an example of the first tomographic image of the present disclosure, the tomographic image 57 _n of the present embodiment corresponds to an example of a second tomographic image of the present disclosure. In the following, when the tomographic images 57 _{1 to} 57 _n of the tomographic image group 56A and the tomographic image group 56B are generically referred to without distinction, they are referred to as "tomographic image 57". Further, when the tomographic image group 56A and the tomographic image group 56B are collectively referred to without distinction, they are referred to as "tomographic image group 56". The image data of the plurality of tomographic images 57 generated by the tomographic image generator 10 is transmitted to the track detection device 12.

The track detection device 12 is a device that acquires image data of the tomographic image 57 generated by the tomographic image generation device 10 and detects the track T in the emulsion layer 51 of the nuclear emulsion plate 50 based on the acquired tomographic image 57. is there. The detection result of the track T by the track detection device 12 is transmitted to the analysis device 14.

The analysis device 14 is a device that acquires the detection result of the track T by the track detection device 12 and derives the density and the like of the inspection object by analyzing the track T.

Next, the hardware configuration of the track detection device 12 of the present embodiment will be described with reference to FIG. As shown in FIG. 4, the track detection device 12 includes a CPU (Central Processing Unit) 30, a memory 31 as a temporary storage area, and a non-volatile storage unit 32. Further, the track detection device 12 includes a display unit 33 such as a liquid crystal display, an input unit 34 such as a keyboard and a mouse, and a network I / F (InterFace) 36 connected to the network N. The CPU 30, the memory 31, the storage unit 32, the display unit 33, the input unit 34, and the network I / F 36 are connected to the bus 39. An example of the analyzer 14 is a server computer.

The storage unit 32 is realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like. The track detection program 37 is stored in the storage unit 32 as a storage medium. The CPU 30 reads the track detection program 37 from the storage unit 32, expands it into the memory 31, and executes the expanded track detection program 37. In addition, the track angle data 38, which will be described in detail later, is stored in the storage unit 32.

As described above, when the CPU 30 executes the track detection program 37, each of the tomographic image acquisition unit 20, the preprocessing unit 22, the track candidate line identification unit 24, the detection unit 26, and the output unit 28 shown in FIG. Functions as. FIG. 5 shows a block diagram showing an example of the functional configuration of the track detection device 12 of the present embodiment.

The tomographic image acquisition unit 20 acquires the image data of the tomographic image 57 transmitted from the tomographic image generation device 10, and outputs the image data of the acquired tomographic image 57 to the preprocessing unit 22.

The preprocessing unit 22 performs image processing as preprocessing on the input image data of the tomographic image 57. The pre-processing is a process for facilitating the detection of the track T from the tomographic image 57 in the track candidate line identification unit 24 in the subsequent stage, and is a process for facilitating the detection of black spots representing the track T from the tomographic image 57. .. Examples of the preprocessing include various corrections such as offset correction, gain correction, and image unevenness. The image data of the tomographic image 57 after the preprocessing is output to the track candidate line identification unit 24.

The track candidate line identification unit 24 becomes a candidate for a track line representing the track T passing through the tomographic image 57 according to the assumed track angle of the track T based on the tomographic image 57 represented by the input image data. The track line candidate is specified, and information representing the specific result is output to the detection unit 26. The "track angle" is an incident angle at which muons are incident on the emulsion layer 51A. In the present embodiment, as shown in FIG. 6, the zenith angle θ and the surface of the emulsion layer 51 (tomographic image 57). It is an angle represented by a combination of in-plane angles ω parallel to, and is also called the direction in which muons are incident. As shown in FIG. 6, the in-plane angle ω has a predetermined direction of the in-plane direction of ω = 0 degrees, and the counterclockwise direction is positive with respect to the axis of the zenith angle θ = 0 degrees. It is expressed by the angle of ω = 0 degrees or more and ω = less than 360 degrees.

When the track angle (θ, ω) of the track T is other than (0,0), in other words, when the track T is obliquely incident on the tomographic image 57, as an example, as shown in FIG. 7, the tomographic image group 56 The position of the track T passing through each tomographic image 57 in each tomographic image 57 is different. As an example, FIG. 7 shows a case where the tomographic image group 56 includes four tomographic images 57 (n = 4). In the example shown in FIG. 7, tracks T which passes through the reference point 60 of the tomographic image 57 ₁ of the pixel 58 (x1, y1) is the reference point in each of the tomographic images 57 _2, a tomographic image 57 _3, and the tomographic image 57 ₂ 60 It is a position different from the position corresponding to, and passes through a different position in each tomographic image 57. In the present embodiment, as an example, the reference point 60 is set as the center point of the pixel 58.

As the assumed track angle of the track T, the track line by the virtual track T when the zenith angle θ and the in-plane angle ω are comprehensively determined is defined as the virtual track candidate line. In this case, as in the example shown in FIG. 8, as a reference point 60 of the tomographic image 57 ₁ of the pixel 58 is the uppermost layer (x1, y1), the same pixel 58 in the tomographic image 57 _n is a bottom layer (xn , Yn), there are a plurality of (five in FIG. 8) virtual track candidate lines VC.

The detection unit 26 in the latter stage performs a calculation on a pixel-by-pixel basis of the tomographic image 57 based on the black spots formed on the tomographic image 57 according to the track T, and tracks from the track candidate line specified by the track candidate line identification unit 24. The track line due to T is detected. Therefore, when an operation is performed using each of a plurality of virtual track candidate lines VC passing through the same pixel 58 of the tomographic image 57 _n , which is the lowest layer, an operation corresponding to the same black point is performed, in other words, an equivalent operation. Will be repeated. Therefore, track candidate lines identifying unit 24 of the present embodiment, as shown in FIG. 9, a plurality of virtual tracks candidate line passing through the reference point 60 of the tomographic image 57 ₁ of the pixel 58 is the uppermost layer (x1, y1) VC By identifying one track candidate line C passing through the same pixel 58 of the tomographic image 57 _n , which is the lowest layer, the amount of calculation is increased by repeating the same calculation in the detection unit 26 in the subsequent stage. Suppress.

As an example, in the track candidate line identification unit 24 of the present embodiment, one virtual track candidate line VC close to the center point 61 of each pixel 58 is selected from among a plurality of virtual track candidate line VCs passing through the tomographic image 57 _n . It is specified as a track candidate line C. In the example shown in FIG. 9, the track candidate line identification unit 24 identifies the track candidate line C passing through each of the nine pixels 58 of the tomographic image 57 _n from the virtual track candidate line VC.

Thus, in this embodiment, track angle track candidate lines (hereinafter, referred to as "track candidate angle") (theta, omega) is the size of the pixels 58 in the tomographic image 57, from the tomographic image 57 ₁ to tomographic images 57 _n It is determined according to the distance of the zenith angle θ and the step of the in-plane angle ω. Tracks candidate lines identifying unit 24 of the present embodiment, the track angle (theta, omega), with the size of the pixel 58 in the tomographic image 57, the distance and the zenith angle theta and a surface from the tomographic image 57 ₁ to tomographic images 57 _n Information associated with the three conditions of the notch of the internal angle ω is stored in the storage unit 32 as track angle data 38.

The detection unit 26 detects the track T from the plurality of track candidate lines specified by the track candidate line identification unit 24. As an example, the detection unit 26 of the present embodiment detects the track T by using the so-called shift method. Specifically, the detection unit 26 shifts each tomographic image 57 by a shift amount according to the track angle (θ, ω) of the track T, and the track T having the corresponding track angle (θ, ω) is each tomographic image. Assuming that the tomographic images 57 are located at substantially the same x and y coordinates, the track T is detected based on the presence or absence of black spots at the same coordinates of each tomographic image 57, and information representing the detection result is output to the output unit 28.

The output unit 28 outputs information representing the detection result of the detection unit 26, that is, information representing the track T, to the analysis device 14 via the network N by the network I / F36.

Next, the operation of the track detection device 12 of the present embodiment will be described. First, when the CPU 30 executes the track detection program 37, the track detection process shown in FIG. 10 is executed. The track detection process shown in FIG. 10 is executed, for example, at the timing when the user of the track detection device 12 instructs the track detection to be input by the input unit 34.

In step S100 of FIG. 10, the tomographic image acquisition unit 20 acquires the image data of the tomographic image 57 from the tomographic image generator 10 as described above. As described above, the image data of the tomographic image 57 acquired by the tomographic image acquisition unit 20 is preprocessed by the preprocessing unit 22.

In the next step S102, the track candidate line identification unit 24 performs the track candidate line identification process shown in FIG. 11 in detail later, and as described above, is based on the preprocessed tomographic image 57. , The track candidate line, which is a candidate for the track line representing the track T, is specified.

In the next step S104, the detection unit 26 performs the detection process shown in FIG. 15, which will be described in detail later, and detects the track T from the track candidate line C identified in the step S102 as described above. , Ends this track detection process.

The track candidate line identification process executed in step S102 of the track detection process (see FIG. 10) described above will be described in detail with reference to FIG.

In step S120 of FIG. 11, the track candidate line specifying unit 24 determines whether or not to derive the track candidate angle (θ, ω). In the present embodiment, the track candidate angle (θ, ω) based on the track angle data 38 stored in the storage unit 32 is used to identify the track candidate line C, and the newly derived track candidate angle (θ, ω) is used. ) May be used. As an example, in the track candidate line specifying unit 24 of the present embodiment, when the user instructs the input unit 34 to derive the track candidate angle (θ, ω), and the information stored in the track angle data 38, If the above conditions do not match, it is determined that the track candidate angles (θ, ω) will be derived. As an example of the case where the above conditions do not match, in the track angle data 38, the size of the pixel 58 to which the track candidate angle (θ, ω) is associated and the pixel 58 of the tomographic image 57 acquired in the above step S100. There is a case where the difference in size of the pixels 58 exceeds the permissible range.

When deriving the track candidate angles (θ, ω), the determination in step S120 becomes an affirmative determination, and the process proceeds to step S122. In step S122, the track candidate line identification unit 24 executes the track candidate angle derivation process shown in FIG. 12 as an example, which will be described in detail later, and then ends the track candidate line identification process to detect the track as shown in FIG. The process proceeds to step S104.

On the other hand, when the track candidate angle (θ, ω) is not derived, in other words, when the track angle data 38 is used, the determination in step S120 of the track candidate line identification process becomes a negative determination, and the process proceeds to step S124.

In step S124, the track candidate line specifying unit 24 specifies the track candidate angle (θ, ω) from the track angle data 38. Specifically, track candidate line specifying unit 24, corresponding to the tomographic image 57 tomographic image obtaining unit 20 obtains the distance and size of the pixels 58 in the tomographic image 57, from the tomographic image 57 ₁ to tomographic images 57 _n And the track candidate angles (θ, ω) associated with the notches of the zenith angle θ and the in-plane angle ω are acquired from the track angle data 38. By ending step S124, the main track candidate line identification process is completed, and the process proceeds to step S104 of the track detection process shown in FIG.

The track candidate angle derivation process executed in step S120 of the track candidate line identification process (see FIG. 11) described above will be described in detail with reference to FIG.

In step S200 of FIG. 12, the track candidate line specifying unit 24 generates a list corresponding to all combinations of the zenith angle θ and the in-plane angle ω assumed as the track candidate angles (θ, ω). In other words, a list is generated according to the zenith angle θ and the in-plane angle ω of the track angles (hereinafter referred to as “virtual track candidate angles”) (θ, ω) of all the virtual track candidate lines VC. FIG. 13 shows an example of the list 70. In step S200, a list 70 in which data is input is generated in the columns of "line ID", "θ", and "ω" in the list 70 shown in FIG.

In general, muons fall from the zenith, so when the nuclear dry plate 50 is placed in a state where the normal direction is the zenith angle θ = 0 degrees, the amount of muons that reach the nuclear dry plate 50 is the zenith. There are many cases where the angle θ = less than 90 degrees, and especially within the range where the zenith angle θ = 45 degrees or less. Therefore, the track candidate line identification unit 24 of the present embodiment comprehensively virtual tracks for each degree within a range where the zenith angle θ is 0 degrees or more and 45 degrees or less and the in-plane angle ω is 0 degrees or more and less than 360 degrees. Candidate angles (θ, ω) are defined. When the zenith angle θ is 0 degrees, the obtained track candidate lines C are the same regardless of the angle of the in-plane angle ω. Therefore, in the track candidate line specifying unit 24 of the present embodiment, only the case where the zenith angle θ is 0 degrees and the in-plane angle ω is 0 degrees is included in the virtual track candidate angles (θ, ω). Therefore, the virtual track candidate angles (θ, ω) are 16201 (45 × 360 + 1 = 16201), and the lines ID1 to ID16201 are sequentially associated with each combination as shown in FIG.

Tracks candidate lines identifying unit 24 in the next step S202, as a reference point 60 of the tomographic image 57 ₁ of the uppermost, through all virtual tracks candidate lines VC having a virtual track candidate angle (theta, omega) is the bottom tomographic images 57 _n pixels 58 derives the position in the tomographic image 57 _n, input to each of the field "pixel coordinates x" and "pixel coordinate y 'of table 70. In the present embodiment, the coordinates expressed in pixel units are used as the positions of the pixels 58 in the tomographic image 57. In the list 70 shown in FIG. 13, the virtual track candidate line VC having the line ID “3” and the virtual track candidate angle (1, 1) is the pixel coordinates (5, 0) in the tomographic image 57 _{n at} the bottom. It shows that it passes through the pixel 58 at the position of.

In the next step S204, the track candidate line identification unit 24 determines the center point 61 of the pixel 58 of the lowest tomographic image 57 _n through which the virtual track candidate line VC passes, and the passing position for all the virtual track candidate line VCs. The distance is derived and entered in the "Δd" field of the list 70. Δd is the virtual track candidate angle (theta, omega), is derived from the distance of the size of a pixel 58 in the tomographic image 57, and from the tomographic image 57 ₁ to tomographic images 57 _n.

In the next step S206, the track candidate line identification unit 24 resets each value in the ID sequence table 72 shown in FIG. 14 to “0”. The ID sequence table 72, which shows an example in FIG. 14, is a table in which the coordinates (positions) of the pixels 58 in the tomographic image 57 _n of the lowermost layer are adopted as items. In the ID sequence listing 72 of the present embodiment, "0" is input when there is no virtual track candidate line VC passing through the pixel 58. Further, as will be described in detail later, in the ID arrangement table 72, when one virtual track candidate line VC passing through the pixel 58 exists, the line ID of the virtual track candidate line VC passing through the pixel 58 is input. Further, in the ID sequence listing 72, when there are a plurality of virtual track candidate line VCs passing through the pixel 58, one of the plurality of virtual track candidate line VCs passing through the pixel 58 passes through the position closest to the center point 61. The line ID of the virtual track candidate line VC of is input.

In the next step S208, the track candidate line identification unit 24 sets the attention ID to 1 (attention ID = 1).

In the next step S210, the track candidate line identification unit 24 acquires the pixel coordinates (x, y) corresponding to the attention ID from the list 70. In the example shown in FIG. 13, when the attention ID is "1", the pixel coordinates (0,0) are acquired from the list 70, and when the attention ID is "3", the pixel coordinates (5,0) are listed. Obtained from Table 70.

In the next step S212, the track candidate line identification unit 24 determines whether or not the pixel coordinate (x, y) value of the ID array table 72 corresponding to the attention ID is 0 ((x, y) = 0). To do. When the value of the pixel coordinates (x, y) is 0, the determination in step S212 becomes an affirmative determination, and the process proceeds to step S214. That is, if there is no virtual track candidate line VC that passes through the pixel 58 of the pixel coordinates (x, y) acquired in step S210 in addition to the virtual track candidate line VC corresponding to the attention ID, the determination in step S212 is affirmative. The determination is made, and the process proceeds to step S214.

In step S214, the track candidate line identification unit 24 proceeds to step S228 after inputting "1" indicating validity in the "valid / invalid" field corresponding to the attention ID in the list 70. Further, the track candidate line identification unit 24 inputs the number of the attention ID to the value of (x, y) in the ID sequence listing 72 corresponding to the pixel coordinates (x, y) acquired in step S210. The "valid / invalid" shown in the list 70 indicates whether the virtual track candidate line VC is valid or invalid as the track candidate line C, and if it is "valid", the above-mentioned "1" is input as in, and if it is invalid, "0" is input.

On the other hand, when the values of the pixel coordinates (x, y) are not 0, the determination in step S212 becomes a negative determination, and the process proceeds to step S216. That is, if there is a virtual track candidate line VC that passes through the pixel 58 of the pixel coordinates (x, y) acquired in step S210 in addition to the virtual track candidate line VC corresponding to the attention ID, the determination in step S212 is denied. The determination is made, and the process proceeds to step S216.

In step S216, the track candidate line identification unit 24 has a virtual value of Δd corresponding to the attention ID in the list 70 and a line ID input to the pixel coordinates (x, y) of the attention ID in the ID sequence table 72. Compare with Δd of the track candidate line VC. In the next step S218, the track candidate line identification unit 24 determines whether or not the Δd of the attention ID is smaller than the Δd of the line ID compared in the step S216.

When the Δd of the attention ID is smaller than the Δd of the line ID compared in step S216, the determination in step S218 becomes an affirmative determination, and the process proceeds to step S222. When step S218 is affirmative in this way, the virtual track candidate line VC corresponding to the attention ID has more pixels 58 than the virtual track candidate line VC corresponding to the line ID compared in step S216. This is a case of passing through a position close to the center point 61.

In step S222, the track candidate line identification unit 24 inputs "1" indicating validity in the "valid / invalid" field corresponding to the attention ID in the list 70. In the next step S224, the track candidate line identification unit 24 sets the value entered in the “valid / invalid” column corresponding to the line ID compared in the above step S216 in the list 70 as “1” indicating validity. Therefore, it is set to "0" indicating invalidity.

In the next step S226, the track candidate line identification unit 24 inputs the number of the attention ID to the value of (x, y) in the ID sequence listing 72 corresponding to the pixel coordinates (x, y) of the attention ID, and then the track candidate line identification unit 24. The process proceeds to step S228.

On the other hand, when the Δd of the attention ID is not smaller than the Δd of the line ID compared in step S216, in other words, when the Δd of the attention ID is Δd or more of the line ID compared in step S216, the step S218 The determination becomes a negative determination, and the process proceeds to step S220. When step S218 is negatively determined in this way, the virtual track candidate line VC corresponding to the attention ID has more pixels 58 than the virtual track candidate line VC corresponding to the line ID compared in step S216. When passing through a position far from the center point 61, or when the virtual track candidate line VC corresponding to the attention ID and the virtual track candidate line VC corresponding to the line ID compared in step S216 pass through the same position.

In step S220, the track candidate line identification unit 24 enters "0" indicating invalidity in the "valid / invalid" field corresponding to the attention ID in the list 70, and then proceeds to step S228.

For example, when the attention ID is "3", the pixel coordinates (x, y) of the attention ID are "(5,0)" and Δd is "0.080" according to the list 70. .. According to the ID arrangement table 72, the line ID input to the pixel coordinates (5,0) is “2”. According to the list table 70, the line ID corresponds to “2”, the value of Δd is “0.078”, and the distance passes closer to the center point 61 than the virtual track candidate line VC corresponding to the above attention ID. Therefore, the determination in step S218 is a negative determination. In the next step S220, the track candidate line identification unit 24 inputs "0" indicating invalidity in the "valid / invalid" field corresponding to the attention ID = line ID "3".

In the present embodiment, if the Δd of the attention ID and the Δd of the line ID compared in the step S216 are equal in the step S218, a negative determination is made and the input is made earlier by shifting to the step S220. The mode in which the virtual track candidate line VC corresponding to the line ID is adopted as the track candidate line C has been described. However, when Δd of the attention ID and Δd of the line ID compared in step S216 are equal, which virtual track candidate line VC is adopted as the track candidate line C is not limited to this embodiment. For example, it may be determined which is adopted according to the magnitude of the zenith angle θ or the magnitude of the in-plane angle ω.

In step S228, the track candidate line identification unit 24 determines whether or not the attention ID is less than the maximum value of the line ID (attention ID <maximum value). As described above, in the present embodiment, since the maximum value of the line ID is “16201”, if the value of the attention ID is less than “16201”, the determination in step S228 becomes an affirmative determination, and the process proceeds to step S230. In step S230, the track candidate line identification unit 24 sets a value obtained by adding 1 to the attention ID as a new attention ID (attention ID = attention ID + 1), then returns to step S210 and repeats the processes of steps S210 to S228.

On the other hand, if the attention ID is not less than the maximum value of the line ID, in other words, when the attention ID reaches the maximum value of the line ID, the determination in step S228 becomes a negative determination, and the process proceeds to step S232. In step S232, the track candidate line identification unit 24 determines the virtual track candidate angles (θ, ω) corresponding to the line ID, in which the value in the “valid / invalid” column of the list 70 is “1” indicating valid. After specifying as the track candidate angle (θ, ω) of the track candidate line C, the main track candidate angle derivation process is completed. By terminating the track candidate angle derivation process, the track candidate line identification process shown in FIG. 11 is also completed, and the process proceeds to step S104 of the track detection process shown in FIG.

Incidentally, the data of the track candidate angle derived trajectory candidate angle derived by processing (theta, omega), with the size of the pixel 58 in the tomographic image 57, the distance from the tomographic image 57 ₁ to tomographic images 57 _n, zenith angle The track angle data 38 may be stored in the storage unit 32 in association with the three conditions of θ and the step of the in-plane angle ω. When the track detection process is performed from the next time onward by storing the track angle data 38 in this way, if the above three conditions are satisfied, the track candidate line identification unit 24 uses the track angle data 38 to perform the track candidate line. Since C can be specified, the amount of calculation for characterizing the track candidate line C can be reduced.

The detection process executed in step S104 of the track detection process (see FIG. 10) described above will be described in detail with reference to FIG.

In step S140 of FIG. 15, the detection unit 26 shifts each tomographic image 57 from among the plurality of track candidate lines C identified in step S102 by a shift amount corresponding to a certain track candidate angle (θ, ω). Let me. A case where the track candidate line C having the track candidate angles (θ, ω) shown in FIG. 16A passes through the tomographic image group 56 including five tomographic images 57 (n = 5) will be described as a specific example. In this case, as shown in FIG. 16B, the detection unit 26 shifts each tomographic image 57 by a shift amount according to the track candidate angle (θ, ω) and transforms the coordinates of each pixel 58 to convert the track candidate line. It is assumed that C is perpendicular to the tomographic image 57, in other words, the track candidate line C is aligned with the normal in the tomographic image 57.

In the next step S142, the detection unit 26 adds the pixel values of the pixels 58 at a certain coordinate of each tomographic image 57. In the present embodiment, as shown in FIG. 16B, in a state where each tomographic image 57 is shifted by a shift amount according to the track candidate angle (θ, ω), the track T is detected in the region where all the tomographic images 57 overlap. The area at the end of each tomographic image 57 where some tomographic images 57 do not overlap is excluded from the detection target of the track T. Incidentally, the ends of each tomographic image 57 are excluded in the detection of the thus track T region, to a thickness from the tomographic image 57 ₁ to tomographic images 57 _n, from the synthesized tomographic image 59 ₁ of the tomographic image 57 _n Since the size is relatively large, the area is relatively small, so that the effect of excluding the edge region from the detection target is suppressed.

Therefore, the detection unit 26 adds the pixel values of the tomographic images 57 to the pixels 58 at a certain coordinate in the detection target of the track T. In other words, the detection unit 26 adds the pixel values of the pixels 58 of each tomographic image 57 through which the track candidate line C passes.

In the next step S144, the detection unit 26 determines whether or not the addition value obtained in the step S142 is equal to or less than a predetermined threshold value (addition value ≤ threshold value). In the present embodiment, the smaller the pixel value, the darker the image, and the larger the pixel value, the brighter the image. In other words, the smaller the pixel value, the blacker the image and the black spots, and the larger the pixel value, the whiter the image and the white spots. Therefore, when a black spot is formed on the tomographic image 57 by the track T, the pixel value of the pixel 58 that has become the black spot becomes small. In the present embodiment, a threshold value for determining that a black spot is formed in the tomographic image group 56 by the track T is set in advance based on the addition value of the pixels when all the tomographic images 57 are black spots. deep. The relationship between the pixel addition value and the threshold value when all the tomographic images 57 are black spots depends on the characteristics of the emulsion used. This is because the sensitivity of the emulsion depends on how much the muon exposes the passed emulsion to light, that is, how many black spots are generated. Furthermore, as the zenith angle θ increases, the passing distance of muons increases, so the number of sunspots generated increases. In the experiment, when the zenith angle θ is 0 degrees, the addition value of the pixels when all the tomographic images 57 are black points, whereas the addition value of the pixels when 80% of the tomographic images 57 are black points. Good results were obtained by setting.

If the added value is not less than or equal to the threshold value, in other words, if the added value exceeds the threshold value, the determination in step S144 becomes a negative determination, and the process proceeds to step S148. On the other hand, when the added value is equal to or less than the threshold value, the determination in step S144 becomes an affirmative determination, and the process proceeds to step S146.

In step S146, the detection unit 26 identifies the track candidate line C having the track candidate angles (θ, ω) in step S140 and having the coordinates in step S142 as the track T.

In the next step S148, the detection unit 26 determines whether or not the processing of steps S142 to S146 has been performed for all the coordinates (pixels 58) in the detection target of the track T. If there are coordinates (pixels 58) that have not been processed in steps S142 to S146, the determination in step S148 is a negative determination, and the process proceeds to step S150. In step S150, the detection unit 26 changes the coordinates (pixel 58) to be detected of the track T, then returns to step S142, and repeats the processes of steps S142 to S146.

On the other hand, when the processing of steps S142 to S146 is performed for all the coordinates (pixels 58) to be detected by the track T, the determination in step S148 becomes an affirmative determination, and the process proceeds to step S152.

In step S152, the detection unit 26 determines whether or not the processing of steps S140 to S150 has been performed on all of the plurality of track candidate lines C identified in step S102. If there is a track candidate line C for which the processes of steps S140 to S150 have not been performed yet, the determination in step S152 becomes a negative determination, and the process proceeds to step S154. In the next step S154, the detection unit 26 changes the track candidate line C to be detected by the track T, then returns to step S140, and based on the track candidate angle (θ, ω) of the changed track candidate line C, the above The processing of steps S140 to S152 is repeated.

On the other hand, when the processes of steps S140 to S150 are performed for all of the plurality of track candidate lines C specified in step S102, the determination in step S152 becomes an affirmative determination, and the process proceeds to step S156.

In step S156, the detection unit 26 outputs information representing all the track T specified as the track T in step S146 to the output unit 28 as a detection result, and then ends this detection process. When the present detection process is completed, step S104 of the track detection process (see FIG. 10) is completed, and the track detection process is completed. Although not shown in the present application, after this process, a process of collecting duplicate tracks is performed. When the black spot has a size occupying a plurality of pixels 58 on the synthetic tomographic image 59 or the tomographic image 57, one track T of the nuclear emulsion 50 may be detected as a plurality of track Ts. However, since these detected plurality of track T have a feature that they pass through extremely close positions such as adjacent pixels 58 and the directions of the track T are almost the same, they are detected in duplicate. It can be extracted as a track T. Tracks T detected in duplicate are subjected to a process of collecting them as one track T having an average passing position and flight direction of those track Ts.

Regarding the method of determining whether or not the track candidate line C is the track T in the track detection process (see FIG. 15) executed by the detection unit 26 of the track detection device 12, each tomographic image 57 through which the track candidate line C passes. Although the method based on the comparison result of comparing the added value of the pixel values of the pixels 58 with the predetermined threshold value has been described, the method is not limited to the above method. For example, the following method may be adopted. First, the detection unit 26 compares each pixel 58 of each tomographic image 57 through which the track candidate line C passes with a threshold value for determining whether or not a black spot is formed in the pixel 58, and whether or not a black spot is formed. Judge whether or not. Further, when the number of pixels 58 (tomographic image 57) determined to be black spots is equal to or greater than the threshold value for determining whether or not the track candidate line C is the track T, the detection unit 26 determines the track candidate line C. You may use the method of determining that is a track T.

In this way track detector 12 of the present embodiment, track candidate lines identifying unit 24, the direction is different each passes through the reference point 60 of the tomographic image 57 ₁ which is the uppermost layer of the tomographic image group 56, and each of the tomographic image A plurality of track candidate lines C passing through different pixels 58 of the tomographic image 57 _n , which is the lowest layer of the group 56, are specified. According to the present embodiment, as compared with the case where the track T is detected using the track candidate line C in which the assumed zenith angle θ and the in-plane angle ω are comprehensively determined and determined as the track candidate angles (θ, ω). Therefore, the number of track candidate lines C used for detecting the track T can be reduced.

Therefore, according to the track detection device 12 of the present embodiment, it is possible to reduce the amount of calculation for detecting the track without lowering the detection accuracy.

[Second Embodiment]
Hereinafter, the second embodiment will be described in detail. In this embodiment, the same configurations and actions as those described in the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

The overall configuration of the non-destructive inspection system 1 and the configuration of the track detection device 12 of the present embodiment are the configurations of the non-destructive inspection system 1 and the track detection device 12 of the first embodiment (FIGS. 2, 4, and 5). Since it is the same as (see), the description thereof will be omitted.

Since a part of the operation of the track detection device 12 is different in the present embodiment, the operation of the track detection device 12 of the present embodiment will be described. In the track detection device 12 of the present embodiment, the track candidate angle executed in step S122 of the track candidate line identification process (see FIG. 11) executed in step S102 of the track detection process (see FIG. 10) of the first embodiment. Since a part of the derivation process is different, the track candidate angle derivation process executed by the track detection device 12 of the present embodiment will be described.

FIG. 17 shows a flowchart showing an example of the track candidate angle derivation process executed by the track detection device 12 of the present embodiment. The track candidate angle derivation process of the present embodiment is different from the track candidate angle derivation process of the first embodiment (see FIG. 12) in that the processes of steps S234 to S250 are performed after step S232.

As shown in FIG. 17, the track candidate line specifying unit 24 in step S234 after step S232 is selected from among the plurality of track candidate lines C specified in step S232, and the track candidate line C of interest (hereinafter, “attention”). One "track candidate line") is determined. Specifically, the track candidate line identification unit 24 has a track candidate line C corresponding to a line ID in which the value entered in the “valid / invalid” column in the list 70 is “1” indicating validity. From, the attention track candidate line C is determined.

In the next step S236, the track candidate line specifying unit 24 selects one other track candidate line C of the attention track candidate line C from the plurality of track candidate lines C specified.

In the next step S238, the track candidate line identification unit 24 determines whether or not the angle formed by the attention track candidate line C determined in step S234 and the track candidate line C selected in step S236 is equal to or less than a predetermined value. judge.

When the angle formed by the attention track candidate line C and the selected track candidate line C exceeds a predetermined value, the determination in step S238 becomes a negative determination, and the process proceeds to step S244. On the other hand, when the angle formed by the attention track candidate line C and the selected track candidate line C is equal to or less than a predetermined value, the determination in step S238 becomes an affirmative determination, and the process proceeds to step S240.

In step S240, the track candidate line specifying unit 24 determines whether or not the selected track candidate line C satisfies the exclusion condition for excluding from the specified track candidate line C. When the angle formed by the track candidate lines C is equal to or less than a predetermined value, the density of the track candidate lines C is higher than that of the other plurality of track candidate lines C. In particular, when the zenith angle θ at the track candidate angle (θ, ω) of the track candidate line C is relatively small, as shown in FIG. 18A, compared with the case where the zenith angle θ at the track candidate angle (θ, ω) is relatively large. Therefore, the density of the track candidate line C tends to be high. In FIG. 18A, when the length of the line segment of the track candidate line C specified in step S232 is the same, the end of the track candidate line C when the tomographic image 57 is viewed from the incident side of the track T. An example of the position of is shown.

Therefore, the track candidate line specifying unit 24 of the present embodiment thins out the track candidate lines C in a high density state, so that the density of the entire specified track candidate lines C is determined by the zenith angle θ and the in-plane angle ω. Instead, bring it closer to an even state. FIG. 18B shows the track candidate line C when the tomographic image 57 is viewed from the incident side of the track T when the lengths of the line segments are the same for the track candidate line C after thinning out by the main track candidate angle derivation process. An example of the position of the end of is shown. As can be seen by comparing FIGS. 18A and 18B, in the track candidate line C group after thinning out, track candidate lines C are regularly provided, and the density is approaching a uniform state.

In the present embodiment, an exclusion condition is provided as a condition for thinning out the track candidate lines C in such a high density state. The exclusion condition may be a condition that can appropriately reduce the amount of calculation without impairing the detection accuracy of the track T.

For example, the condition that the larger in-plane angle ω of the attention track candidate line C and the selected track candidate line C is excluded, and the smaller in-plane angle ω of the track candidate line C and the selected track candidate line C. One of the conditions for excluding is may be used as the exclusion condition. In this case, the track candidate line C can be provided more regularly.

Further, for example, when there are a plurality of track candidate lines C having the same zenith angle θ and the angles formed by each other are equal to or less than a predetermined value, among the specified plurality of track candidate lines C. By excluding a part, the distribution of the remaining plurality of track candidate lines C may be an exclusion condition in which the distribution of the remaining track candidate lines is set to be the closest to the rotational symmetry with respect to the axis having the zenith angle of 0 degrees. In other words, when there are a plurality of track candidate lines C whose angles formed by each other are equal to or less than a predetermined value, the track candidate lines C in a state where the zenith angle θ is closest to the axis of 0 degree rotationally symmetric are used. It may be an exclusion condition that leaves and excludes the other track candidate line C. In this case, the in-plane angle ω can be thinned out more evenly, and the bias of the track candidate line C can be suppressed.

If the selected track candidate line C does not satisfy the exclusion condition, the determination in step S240 becomes a negative determination, and the process proceeds to step S244. On the other hand, when the selected track candidate line C satisfies the exclusion condition, the determination in step S240 becomes an affirmative determination, and the process proceeds to step S242.

The track candidate line specifying unit 24 in step S242 excludes the selected track candidate line C from the specified track candidate line C. As an example, the track candidate line specifying unit 24 of the present embodiment sets the value entered in the "valid / invalid" field associated with the line ID of the list 70 corresponding to the track candidate line C to be excluded. By changing from "1" indicating valid to "0" representing invalid, it is excluded from the specified track candidate line C.

In the next step S244, the detection unit 26 determines whether or not all of the other track candidate lines C of the attention track candidate line C among the specified plurality of track candidate lines C have been selected in the above step S236. .. If there is a track candidate line C that has not yet been selected, the determination in step S244 becomes a negative determination, and the process proceeds to step S246. The track candidate line identification unit 24 in step S246 is a track candidate line C other than the track candidate line C of interest among the plurality of track candidate lines C identified in step S246, and is not yet selected in step S236. After selecting one, the process returns to step S238 and the processes of steps S238 to S244 are repeated.

On the other hand, when all the other track candidate lines C of the attention track candidate line C among the specified plurality of track candidate lines C are selected in the above step S236, the determination in step S244 becomes an affirmative determination, and the process proceeds to step S248. Transition.

In step S248, the track candidate line specifying unit 24 determines whether or not all the specified track candidate lines C are set as the attention track candidate lines C. Specifically, the track candidate line identification unit 24 has a track candidate line C corresponding to a line ID in which the value entered in the “valid / invalid” column in the list 70 is “1” indicating validity. Among them, it is determined whether or not there is a track candidate line C that is not the attention track candidate line C. If there is a track candidate line C that has not yet been set as the attention track candidate line C, the determination in step S248 becomes a negative determination, and the process proceeds to step S250. In step S250, the track candidate line identification unit 24 changes the attention track candidate line C, then returns to step S236, and repeats the processes of steps S236 to S248.

On the other hand, when all the specified track candidate lines C are set as the attention track candidate lines C, the determination in step S248 becomes an affirmative determination, and the main track candidate angle derivation process ends.

As described above, in the track detection device 12 of the present embodiment, the track candidate line identification unit 24 further thins out the track candidate lines C, so that the number of track candidate lines C used for detecting the track T can be further reduced. , It is possible to suppress the bias between the track candidate lines C. Therefore, according to the track detection device 12 of the present embodiment, the amount of calculation for detecting the track can be further reduced without lowering the detection accuracy.

[Third Embodiment]
Hereinafter, the third embodiment will be described in detail. In this embodiment, the same configurations and actions as those described in the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

The overall configuration of the non-destructive inspection system 1 and the configuration of the track detection device 12 of the present embodiment are the configurations of the non-destructive inspection system 1 and the track detection device 12 of the first embodiment (FIGS. 2, 4, and 5). Since it is the same as (see), the description thereof will be omitted.

Since a part of the operation of the track detection device 12 is different in the present embodiment, the operation of the track detection device 12 of the present embodiment will be described. In the track detection device 12 of the present embodiment, since a part of the detection process (see FIG. 15) executed in step S104 of the track detection process (see FIG. 10) of the first embodiment is different, the track detection of the present embodiment is performed. The detection process executed by the device 12 will be described.

As described above, when the track angle (θ, ω) of the track T is other than (0,0), in other words, when the track T is obliquely incident on the tomographic image 57, as shown in FIG. 19 as an example. The position of the track T passing through each tomographic image 57 of the tomographic image group 56 in each tomographic image 57 is different. In Figure 19 as an example, a tomographic image group 56 comprises five tomographic images 57 (n = 5), shows a case through the reference point 63 of the tomographic image 57 ₁ tracks T. In Figure 19, for the following description, the position of track T passes through a tomographic image 57 ₃ and the tomographic image 57 _5, since the shown to be the center of the tomographic image 57 ₃ and the tomographic image 57 _5, fault tomographic image 57 ₁ and the tomographic image 57 ₂ of the image group 56, a tomographic image 57 _3, a tomographic image 57 _4, mutual spacing of the tomographic image 57 _5, appears to be no seemingly equal, they as well as FIG. 16A It is evenly spaced.

20 shows, in the tomographic image group 56 in FIG. 19, when the tomographic image 57 ₁ and the tomographic image 57 ₂ viewed from the direction of arrow A, and an example of the nine pixels 58 centered on the reference point 63 is shown There is. 20 shows, in track angle track T (theta, omega), the zenith angle theta to the maximum value, when the plane angle omega is 0 to 360 degrees, the position where the track T passes through a tomographic image 57 ₂ It is shown by the dotted line 64. As shown in FIG. 20, therefore, tracks T which passes through the reference point 63 of the tomographic image 57 _1, a position passing through the tomographic image 57 _2, irrespective of the track angles (theta, omega), the circles shown by the dotted line 64 It will be inside. It similar to this, in the tomographic image group 56 in FIG. 19, also holds for the tomographic images 57 ₃ and the tomographic image 57 _4.

In this way, when the distance between the adjacent tomographic images 57 is relatively short, the amount of deviation of the position where the track T passes between the adjacent tomographic images 57 is within the range of ± 1 pixel, and the tomographic image 57 as a whole If you look at it, the amount of deviation is very small.

Therefore, the detection unit 26 of the present embodiment is based on the black spots included in the synthetic tomographic image obtained by combining two adjacent tomographic images 57, more specifically, the pixel values of the pixels 58 of the synthetic tomographic image. Track T is detected.

FIG. 21 shows a flowchart showing an example of the detection process executed by the track detection device 12 of the present embodiment. The detection process of the present embodiment is different from the detection process of the first embodiment (see FIG. 15) in that step S139 is provided and steps S141 and S143 are provided instead of steps S140 and S142. ing.

As shown in FIG. 21, in the present embodiment, when the detection process is started, the detection unit 26 generates the above-mentioned synthetic tomographic image obtained by combining and synthesizing two adjacent tomographic images 57 in step S139. As described above, in the present embodiment, the amount of deviation of the position where the track T passes between the two adjacent tomographic images 57 is within the range of ± 1 pixel. Therefore, in the present embodiment, as shown in FIG. 22, of the two adjacent tomographic images 57, the lower tomographic image 57 with respect to the upper tomographic image 57 _m (m is an odd number of 1 or more and n or less). _The composite tomographic image 59E and the composite tomographic images 59A to 59D and 59F to 59I which are superimposed by shifting the _{m + 1} in the x-direction and the y-direction in the plane of the tomographic image 57 are generated. In the following, when the synthetic tomographic images 59A to 59I are collectively referred to without distinction, they are simply referred to as the synthetic tomographic image 59. The detection unit 26 uses only the image of the region where the tomographic image 57 _m and the tomographic image 57 _{m + 1} are superimposed as the composite tomographic image 59, and “superimposition” means the pixel value of the pixels 58 of the tomographic images 57 to be superimposed. It means to add.

As described above, in the present embodiment, nine synthetic tomographic images 59 are generated from two adjacent tomographic images 57.

When the number (n) of the tomographic images 57 included in the tomographic image group 56 is an odd number, the tomographic image 57 _n of the lowest layer is used as it is without being combined with other tomographic images 57.

In the next step S141, the detection unit 26 shifts the composite tomographic image 59 or the composite tomographic image 59 from the plurality of track candidate lines C identified in step S102 by the shift amount according to a certain track candidate angle (θ, ω). The synthetic tomographic image 59 and the tomographic image 57 _n of the lowermost layer are shifted. That is, the detection unit 26 executes a process in which the tomographic image 57 in step S140 of the detection process (see FIG. 15) of the first embodiment is replaced with the synthetic tomographic image 59. Which of the synthetic tomographic images 59A to 59I to use is determined according to the track candidate angles (θ, ω). For example, when the zenith angle θ at the track candidate angle (θ, ω) is relatively small, the synthetic tomographic image 59E is used.

Specifically, the detection unit 26 shown in FIG. 23A, track candidate angle (theta, omega) track candidate line C with the, two synthetic tomographic image 59 (59 ₁ and 59 ₂₎ and the tomographic image 57 ₅ The case of passing through (n = 5) will be described as a specific example. Synthesis tomographic image 59 ₁ is a tomographic image obtained by superimposing a tomographic image 57 ₁ and the tomographic image 57 _2, synthesized tomographic image 59 ₂ is a tomographic image obtained by superimposing a tomographic image 57 ₃ and the tomographic images 57 _4. In this case, the detection unit 26, as shown in FIG. 23B, the synthetic respectively track candidate angles of the tomographic image 59 and the tomographic image 57 _n (θ, ω) is shifted by a shift amount corresponding to convert the coordinates of each pixel 58 As a result, the track candidate line C is perpendicular to the synthetic tomographic image 59 and the tomographic image 57 _n , in other words, the track candidate line C is aligned with the normal lines in the synthetic tomographic image 59 and the tomographic image 57 _n . ..

Detector 26 in the next step S143, the each of the synthesized tomographic image 59 or synthetic tomographic image 59 and the tomographic image 57 _n,, adds the pixel values of the pixels 58 at a certain coordinate. That is, the detection unit 26 executes a process in which the tomographic image 57 in step S142 of the detection process (see FIG. 15) of the first embodiment is replaced with the synthetic tomographic image 59.

Specifically, as shown in FIG. 23B, the detection unit 26 synthesizes the composite tomographic image 59 and the tomographic image 57 _n in a state where each of them is shifted by a shift amount according to the track candidate angle (θ, ω). the area where all overlaps the tomographic image 59 and the tomographic image 57 _n a detection target of the track T. The detection unit 26, in the region of the detection object, adds the pixel value of the combined tomographic image 59 and the tomographic image 57 _n of the pixel 58.

Since the processes of steps S144 to S156 after step S143 are the same as the detection processes of the first embodiment, the description thereof will be omitted.

As described above, in the track detection device 12 of the present embodiment, the detection unit 26 uses track based on the black dots included in the plurality of synthetic tomographic images 59 obtained by combining a predetermined number of adjacent tomographic images 57. Detect T. Since the synthetic tomographic image 59 is used for detecting the track T, the number of images can be reduced to about half as compared with the case where the tomographic image 57 is used as in each of the above embodiments.

Therefore, according to the track detection device 12 of the present embodiment, the amount of calculation for detecting the track can be further reduced without lowering the detection accuracy.

In the present embodiment, a mode in which two adjacent tomographic images 57 are combined to form a synthetic tomographic image 59 has been described, but the number of tomographic images 57 to be combined is not limited to two. Incidentally, as the number of tomographic images 57 to synthesize is increased, the top layer of the tomographic image 57 ₁ and the lowermost layer tomographic images 57 _n of the tomographic image 57 to synthesize, displacement amount of the position of track T passes increases .. As the amount of deviation of the position where the track T passes increases, the number of synthetic tomographic images 59 increases, for example, the amount required for the memory for calculation increases. Also, depending on the distance between tomographic images 57 adjacent the top layer in the tomographic image 57 _n of the tomographic image 57 ₁ and lowermost, the deviation amount of the position of track T passes out of the tomographic image 57 to synthesize the As the distance between the adjacent tomographic images 57 increases, the amount of deviation increases.

Therefore, how many tomographic images 57 are combined to form a composite tomographic image 59 depends on the number of tomographic images 57 included in the tomographic image group 56, the distance between adjacent tomographic images 57, the memory capacity, and the like. It is preferable to select.

As described above, the track detection device 12 of each embodiment detects the track T in the emulsion layer 51 that records the track T of the μ particles as a latent image by transmitting the incident μ particles. Device 12. Track detection unit 12, the depth direction in the emulsion layer 51 is obtained by imaging each of the plurality of different cross-section, and a tomographic image 57 ₁ obtained by the nearest cross-section of the emulsion layer 51 on the incident side of the μ particle The tomographic image acquisition unit 20 includes a tomographic image 57 _n obtained by the cross section of the emulsion layer 51 farthest from the incident side of the μ particles, and acquires a tomographic image group 56 in which a plurality of pixels 58 are arranged. Further, track detector 12 is different each direction through the reference point 60 of the tomographic image 57 _1, and tracks the candidate line specifying unit, each specifying a plurality of tracks candidate line C passing through different pixels 58 of the tomographic image 57 _n 24 and a detection unit 26 for detecting the track T in the emulsion layer 51 from a plurality of track candidate lines C.

Therefore, according to the track detection device 12 of each embodiment having the above configuration, it is possible to reduce the amount of calculation for detecting the track without lowering the detection accuracy.

Further, in the above embodiment, the hardware of the processing unit that executes various processes such as the tomographic image acquisition unit 20, the preprocessing unit 22, the track candidate line identification unit 24, the detection unit 26, and the output unit 28, for example. As the hardware structure, various processors shown below can be used. As described above, the various processors include a CPU, which is a general-purpose processor that executes software (program) and functions as various processing units, and a circuit after manufacturing an FPGA (Fieid Programmable Gate Array) or the like. Dedicated electricity, which is a processor having a circuit configuration specially designed to execute a specific process such as a programmable logic device (PLD), which is a processor whose configuration can be changed, and an ASIC (Application Specific Integrated Circuit). Circuits and the like are included.

One processing unit may be composed of one of these various processors, or a combination of two or more processors of the same type or different types (for example, a combination of a plurality of FPGAs or a combination of a CPU and an FPGA). It may be composed of a combination). Further, a plurality of processing units may be configured by one processor.

As an example of configuring a plurality of processing units with one processor, first, one processor is configured by a combination of one or more CPUs and software, as represented by a computer such as a client and a server. There is a form in which a processor functions as a plurality of processing units. Secondly, as typified by System On Chip (SoC), there is a form in which a processor that realizes the functions of the entire system including a plurality of processing units with one IC (Integrated Circuit) chip is used. is there. As described above, the various processing units are configured by using one or more of the various processors as a hardware structure.

Further, as the hardware structure of these various processors, more specifically, an electric circuit (circuitry) in which circuit elements such as semiconductor elements are combined can be used.

Further, in the above embodiment, the mode in which the track detection program 37 is stored (installed) in the storage unit 32 in advance has been described, but the present invention is not limited to this. The track detection program 37 is provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), and a USB (Universal Serial Bus) memory. May be good. Further, the track detection program 37 may be downloaded from an external device via a network.

1 Non-destructive inspection system 10 Fault image generation device 12 Track detection device 14 Analysis device 20 Fault image acquisition unit 22 Preprocessing unit 24 Track candidate line identification unit 26 Detection unit 28 Output unit 30 CPU
31 Memory 32 Storage unit 33 Display unit 34 Input unit 36 Network I / F
37 Track detection program 38 Track angle data 39 Bus 50 Nuclear emulsion 51, 51A, 51B Emulsion layer 52 Film base 56, 56A, 56B Tomographic image group 57, 57 _{1 to} 57 _n , 57 _m , 57 _{m + 1} tomographic image 58, 58 ( x1, y1), 58 (xn , yn) pixel _59,59A~59I, 59 _1, 59 ₂ synthesized tomographic images 60 and 63 reference point 61 the center point 64 the dotted line 70 list 72 ID sequence listing C tracks candidate lines N network T Track VC Virtual track candidate line θ zenith angle ω In-plane angle

When the track angle (θ, ω) of the track T is other than (0,0), in other words, when the track T is obliquely incident on the tomographic image 57, as an example, as shown in FIG. 7, the tomographic image group 56 The position of the track T passing through each tomographic image 57 in each tomographic image 57 is different. As an example, FIG. 7 shows a case where the tomographic image group 56 includes four tomographic images 57 (n = 4). In the example shown in FIG. 7, tracks T which passes through the reference point 60 of the tomographic image 57 ₁ of the pixel 58 (x1, y1) is the reference point in each of the tomographic images 57 _2, a tomographic image 57 _3, and the tomographic image 57 ₄ 60 It is a position different from the position corresponding to, and passes through a different position in each tomographic image 57. In the present embodiment, as an example, the reference point 60 is set as the center point of the pixel 58.

Claims (11)

A track detection device for detecting a track in a charged particle track recording layer that records the track of the charged particle as a latent image by transmitting the incident charged particles.
A first tomographic image obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions and obtained by a cross section of the charged particle track recording layer closest to the incident side of the charged particles. And a second tomographic image obtained by the cross section of the charged particle track recording layer farthest to the incident side of the charged particle, and a tomographic image acquisition unit that acquires a plurality of tomographic images in which a plurality of pixels are arranged respectively. When,
A track candidate line identification unit that identifies a plurality of track candidate lines that pass through different pixels of the second tomographic image and that pass through different reference points of the first tomographic image.
A detection unit that detects the track in the charged particle track recording layer from the plurality of track candidate lines, and a detection unit.
Track detection device equipped with. The track candidate line identification unit selects the track candidate line from a plurality of virtual track candidate lines, each of which passes through the reference point in a different direction and passes through the second tomographic image.
The track detection device according to claim 1. When there are a plurality of virtual track candidate lines passing through the same pixel in the second tomographic image among the plurality of virtual track candidate lines, the track candidate line identification unit uses only one of the virtual track candidate lines. Included in the track candidate line of
When there is one virtual track candidate line passing through the same pixel in the second tomographic image, the virtual track candidate line passing through the same pixel is included in the plurality of track candidate lines.
The track detection device according to claim 2. When there are a plurality of virtual track candidate lines passing through the same pixel in the second tomographic image, the track candidate line identification unit is located at the center of the same pixel among the plurality of virtual track candidate lines passing through the same pixel. The virtual track candidate line passing through the nearest position is defined as the one virtual track candidate line.
The track detection device according to claim 3. When the angle formed by the plurality of track candidate lines is equal to or less than a predetermined angle, the track candidate line specifying unit excludes at least one track candidate line from the specified plurality of track candidate lines.
The track detection device according to any one of claims 1 to 4. One of the conditions that the in-plane angle parallel to the plane on which the charged particles of the charged particle track recording layer is incident is large and the condition that the internal angle is small is predetermined as a deletion condition.
Among the plurality of track candidate lines, the track candidate line specifying unit excludes the track candidate lines satisfying the deletion condition from the specified plurality of track candidate lines.
The track detection device according to claim 5. The track candidate line identification unit specifies a plurality of track candidate lines when there are a plurality of track candidate lines having the same zenith angle and the angles formed by each other are equal to or less than the predetermined angle. By excluding a part of the candidate lines, the distribution of the remaining plurality of track candidate lines is set to be the closest to the rotational symmetry with respect to the axis whose zenith angle is 0 degrees.
The track detection device according to claim 5. The detection unit detects the track from the track candidate line based on the track image corresponding to the latent image included in a plurality of synthetic tomographic images obtained by combining a predetermined number of adjacent tomographic images. To detect,
The track detection device according to any one of claims 1 to 7. The composite tomographic image is an image synthesized by shifting the positions of the adjacent tomographic images according to the direction in which the track candidate line is incident on the charged particle track recording layer.
The track detection device according to claim 8. A track detection method for detecting a track in a charged particle track recording layer that records the track of the charged particle as a latent image by transmitting the incident charged particle.
A first tomographic image obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions and obtained by a cross section of the charged particle track recording layer closest to the incident side of the charged particles. And a second tomographic image obtained by the cross section of the charged particle track recording layer farthest to the incident side of the charged particle, and a plurality of tomographic images in which a plurality of pixels are arranged are acquired.
A plurality of track candidate lines that pass through the reference points of the first tomographic image in different directions and each pass through different pixels of the second tomographic image are identified.
The track in the charged particle track recording layer is detected from the plurality of track candidate lines.
Track detection method including processing. A computer that detects the tracks in the charged particle track recording layer that records the tracks of the charged particles as latent images by transmitting the incident charged particles.
A first tomographic image obtained by imaging each of a plurality of cross sections of the charged particle track recording layer having different depth directions and obtained by a cross section of the charged particle track recording layer closest to the incident side of the charged particles. And a second tomographic image obtained by the cross section of the charged particle track recording layer farthest to the incident side of the charged particle, and a plurality of tomographic images in which a plurality of pixels are arranged are acquired.
A plurality of track candidate lines that pass through the reference points of the first tomographic image in different directions and each pass through different pixels of the second tomographic image are identified.
The track in the charged particle track recording layer is detected from the plurality of track candidate lines.
A track detection program for executing processing.