JP2006078234A - Gravel measuring instrument and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily measure deposited gravel in a river flowing area. <P>SOLUTION: This gravel measuring instrument includes an imaging means arranged in an observation area where gravel is deposited; an illumination means separated from the imaging means by a required dimension and irradiating the observation area from different directions to differentiate the positions of shadow formed in the photographed image; and an image processing means for recording plurality of image data fetched from the same direction by the imaging means under the irradiation from different directions not only to separate and discriminate the gravel in the observation area on the basis of the shadow of the image data but also to measure the particle size (dimension) of individual gravel to analyze distribution. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a gravel measurement device in a river or the like, and a measurement method using the device. The sediment gravel is photographed with a camera in a river, and the size (particle size), particle size distribution, etc. of each photographed sediment gravel are photographed. An image is measured by image analysis using an image processing means.

In order to predict the risk of the occurrence of large-scale landslides in rivers, the Ministry of Construction's River Bureau, based on the river sabo technical standards, focused on areas where landslides exist, Check the landforms with geological conditions and geological conditions by aerial photographs, field surveys, etc., find the survey area, specify a fixed viewpoint within the survey area, and regularly (every 3 years, large river flow It is stipulated that a regular survey will be conducted once a year).
Within the survey area, the fixed viewpoint is determined at intervals of 1 km along the river basin, and in each fixed viewpoint, the gravel accumulation state in the two cross-sectional layers at a predetermined depth from the surface layer and the surface, in detail Gravel particle size, particle size array, distribution state, etc. shall be measured.

More specifically, as shown in FIG. 10, the gorge deposit area usually has a layered structure from the surface layer S 1 to the layer S 2 → the layer S 3 → the layer S 4 → the layer S 5. The surface layer S1 mainly includes gravel B having a large particle size, and the layer S2 having a depth of 30 cm from the surface layer S1 is a zone in which clay adheres to the gravel B. The gravel B, sandy clay CS, sand S, Fine sand SS and silty sand MS are mixed. The subsequent layer S3 is a peculiar layer of scavenging deposits, the layer S4 is a layer of silty sand MS and sandy clay CS, and the layer S5 is a large gravel B layer.
When the lamellar structure shown in FIG. 10 is recognized by the gravel measurement, it is recognized as a scavenging transport area, but when the layered structure is not the lamellar structure shown in FIG. Will be accepted.
From such a viewpoint, it is necessary to measure the gravel in the surface layer and two layers having a predetermined depth from the surface layer as described above.

  Conventionally, the measurement of the gravel from the fixed viewpoint is performed by a method in which the gravel is collected at the site and sifted to measure the size of the gravel. In that case, it is necessary to carry out the surface layer and two layers digging a predetermined depth from the surface layer, and three times of gravel measurement work are necessary from a fixed point of view. In this way, taking a method that relies on human labor in the field is a very time-consuming work, and because the fixing viewpoint is along the mountain stream, there are also problems such as having labor for transporting measuring instruments etc. There are many. Therefore, there is a need for a method that can easily and accurately measure sedimentary gravel in rivers.

  It is also necessary in other fields to measure the particle size of a granular material in a deposited state, and in various aspects, as a particle size measurement method for deposited particles, measure the “section length” by applying a “scale” to a particle photograph. Measurement is carried out by hand, and labor and labor saving of the work is required.

In order to solve the above problem, a method for photographing a deposit and measuring the particle size of the particle based on the photographed image is provided. The method is roughly classified into the following two methods.
1) Particle size measurement in separated state 2) Particle size measurement in accumulated state

As the particle size measurement method in the separated state of 1) above, particles falling from a belt compare, as disclosed in Japanese Patent Laid-Open No. 54-92389 (Patent Document 1), are photographed in silhouette, and this is imaged. There is known a method for obtaining the particle size by processing.
With this method, the particles to be measured can be separated from the background and imaged with good contrast, and the overlap between the particles is also small, so it is possible to achieve highly reliable measurement with simple image processing. is there. However, it is necessary to add or remodel equipment to separate the particles, especially when applied to sedimentary sediments in the river because there are large gravel and clay attached to the gravel 1) This method is not applicable.

  On the other hand, as the particle size measuring method in the above-mentioned 2) in the deposited state, Japanese Patent Publication No. 6-0775030 (Patent Document 2), Japanese Patent Application Laid-Open No. 11-63935 (Patent Document 3), Japanese Patent Application Laid-Open No. 2003-83868. A method disclosed in a gazette (Patent Document 4) is known.

The method disclosed in Patent Document 2 is a method for obtaining an average particle size by statistical spatial frequency analysis, focusing on a phenomenon in which the spatial period of the shadow of particles depends on the particle size.
Since this method does not require the extraction of individual particles, high reliability can be expected when the target particles have an isotropic shape (eg, “sphere”) and a small particle size distribution. .
However, since the gravels of rivers have different shapes and large variations in particle size, the above method cannot be adopted, and it is necessary to rely on a method of obtaining the particle size by separately separating and identifying individual particles.

The methods disclosed in Patent Document 3 and Patent Document 4 perform spatial differential processing on the image of the deposited particles to emphasize and extract the grain boundaries, and obtain the measured particle size of each particle.
In order for the above-mentioned method to be stably established, it is a necessary condition that the contrast of the grain boundary is conspicuous, but in general, it is difficult to ensure the contrast over the entire grain boundary by a normal illumination method, Realizing reliable particle size measurement is extremely difficult.
JP 54-92389 A Japanese Patent Publication No. 61-075030 Japanese Patent Laid-Open No. 11-63935 Japanese Patent Laid-Open No. 2003-83868

  The present invention has been made in view of the above-mentioned problems, and adopts a method of measuring sedimentary gravel from the viewpoint of fixing a river basin, photographing sedimentary gravel with a camera, and performing the image by image processing with a computer. In the image processing, an object is to provide an apparatus and a method capable of distinguishing individual gravels in a deposited state and measuring the particle size thereof with high accuracy and simply.

In order to solve the above problems, the present invention provides a photographing means installed in an observation area where gravel is accumulated,
Illuminating means for illuminating the observation area from different directions while leaving a required dimension from the photographing means, and to change the position of the shadow generated in the photographed image,
A plurality of pieces of image data photographed from the same direction by the photographing means are recorded under irradiation from different directions by the illumination means, and the gravel in the observation area is separated and identified based on the shading of the image data. And a gravel measurement device comprising an image processing means for measuring the particle size (radius) of each gravel and analyzing the distribution.

As described above, in the gravel measuring apparatus of the present invention, it is only necessary to photograph gravel from a fixed point of view in the river survey area, and the photographed image is image-analyzed by a computer. By extracting shadows along the entire circumference of the gravel, the gravel can be separated and identified, and the size and distribution of the gravel can be measured.
As described above, since it is only necessary to photograph the observation area from the fixed viewpoint, it is possible to eliminate the conventional gravel sieving operation by the worker from the fixed viewpoint and to save labor. In particular, since it is only necessary to shoot the surface layer at a point that has been dug a predetermined depth from the surface layer, the three measurement operations from a fixed point of view are more efficient than manual operations by conventional workers. Can be done well.
It should be noted that image processing means for captured images, specifically, a computer may be brought to the shooting site and image processing may be performed on-site, or the image data may be transferred and the images shot from each fixed point of view collectively. Image processing may be performed.

  The measuring apparatus according to the present invention sequentially captures images with different irradiation directions to obtain a plurality of images with different shadow formation positions, and a plurality of images with different shadow formation positions in simultaneous imaging. Any of the devices adapted to obtain the above may be used.

When the above observation area is a device for sequentially photographing from the outer peripheral position,
The photographing means comprises one camera arranged at a required height position from the ground surface at the central position of the observation area,
The illumination means includes a strobe or lamp that irradiates an observation area from at least three directions around the photographing means, and a plurality of units fixed at angular intervals around the photographing means, or Provided with one unit rotated around the photographing means and stopped at an angular interval,
Under the illumination by the illumination means, a plurality of image data photographed from the same direction by the photographing means is synthesized to extract the minimum luminance for each pixel, and a shadow along the entire circumference of each gravel is obtained. .

In addition, it is preferable that a synchronization signal generation unit (trigger generation unit) that synchronizes the shutter timing by the photographing unit and the irradiation timing by sequentially switching the irradiation direction by the illumination unit is provided.
By providing the synchronization signal generating means in this way, images irradiated from different directions can be obtained with high accuracy.

In the gravel measurement device, a photographing means such as a digital camera is installed in an observation area which is a fixed viewpoint of a river basin, and when photographing with the photographing means, the observation area is irradiated with illumination means from different directions for each photographing. In addition, “shade” in different directions is generated in a plurality of these captured images, and by synthesizing these images, shadows are generated on the entire circumference of each gravel to identify each gravel.
For example, four strobes are arranged at intervals of 90 degrees with a digital camera as the center, and four images are sequentially obtained by illuminating the observation area with the four strobes. In each image, shadows of individual gravels are formed on the opposite side of the irradiation direction. Therefore, when four images are combined, shadows are generated on the entire circumference of the individual gravels. Grain boundaries can be obtained.
When synthesizing one minimum luminance image from the four images, a minimum luminance value is extracted for each pixel of the four images, and one piece of pixels composed of the pixels having the extracted minimum luminance value is extracted. Images are synthesized and shaded around the entire periphery of each particle.
By synthesizing with the minimum luminance image, an outline line composed of black lines is obtained over the entire grain boundary without depending on the reflectance of the particles (pebbles) and the illumination illuminance, and extremely between the grain surface and the grain boundary. High contrast can be secured stably. Therefore, grain boundaries can be easily and reliably extracted by simple image processing, and individual particles (pebbles) can be reliably separated and identified from the deposited particles, and the shape and particle size of the gravel are measured. It becomes possible to do.

The image processing means provided in the measurement apparatus of the present invention is capable of synthesizing the images, specifically,
The image processing means includes a recording section for receiving an image from the photographing means and recording it as image data, an arithmetic processing section for the recorded image data, and a display for displaying an image of a measurement result calculated by the arithmetic processing section. The arithmetic processing unit comprises:
An image composition unit that creates one minimum brightness composite image from a plurality of recorded images, a boundary emphasized image creation unit that emphasizes the outline of the gravel from the minimum brightness composite image, and a boundary emphasized image creation unit A particle recognition unit that separates and identifies each particle based on a single image produced to produce a particle identification image and a particle boundary display image, and a particle size of each gravel based on the image produced by the particle recognition unit (Radius), the number of distributions in a predetermined particle size range, and the weight of gravel, and further, a gravel analysis unit that calculates the passing percentage from the weight. Specifically, the projected area of each gravel is regarded as a circle to determine the radius, and each gravel is regarded as a sphere having the radius. The weight is obtained by multiplying this by the specific gravity of the gravel obtained in advance, and the weight-converted passage percentage is calculated therefrom.

Instead of the above-described method of sequentially shooting by irradiation from different directions, an apparatus for obtaining a plurality of images having shadows at different positions by simultaneous shooting may be used.
In this case, the photographing means includes one digital camera having a plurality of photographing elements having different spectral spectral sensitivities and having the same optical axis as the optical axis, or a plurality of digital cameras having different spectral spectral sensitivities. Are arranged such that the optical axes are bundled so as to be in substantially the same direction, while a plurality of illumination means are arranged at an angular interval around the photographing means, and the plurality of illumination means arranged are respectively The light corresponding to the spectral spectral sensitivity of each imaging element of one digital camera of the imaging means or the spectral spectral sensitivities of a plurality of digital cameras is irradiated.

  In the case of the above configuration, each illumination means is equipped with a color filter of a different color, and a color filter of a different color corresponding to each photographing element of the one digital camera or a plurality of digital cameras. It is preferable.

For example, with a digital camera as a center, strobes with four color filters having different spectral spectra are arranged at intervals of 90 degrees, and the lighting signals output from the lighting controller controlled by the sequence control circuit Lights up. An image sensor having a color filter corresponding to the color filter mounted on the strobe is also provided on the camera side, and an image at the time of each strobe lighting is simultaneously taken into the memory of the image processing means in synchronization with the lighting signal by the sequence control circuit. In the image processing means, the same image processing as in the case of sequential shooting, that is, the minimum luminance is extracted for each pixel, and one composite image having the minimum luminance is obtained. (Outline) is extracted.
In this way, if the color filters attached to the illumination means and the image sensor of the camera are separated from each other as the spectral transmission characteristic as a spectrum, the image obtained by each image sensor is attached with the same color filter as the image sensor. Since the image is illuminated by a strobe or lamp, the four-direction illumination image is simultaneously captured in the image memory.

The plurality of imaging elements having the same optical axis direction can be formed by dividing the optical axis using an optical element such as a prism or a half mirror built in the camera. Alternatively, a plurality of cameras that are sufficiently small with respect to the visual field size may be bundled.
In the case where a visual field shift becomes a problem between a plurality of images with substantially the same optical axis direction, it is preferable to provide a visual field shift correction circuit that shifts each image by the visual field shift before the granularity calculation circuit.

In addition, the measurement apparatus includes an outer frame member that is arranged on the ground of the observation region and surrounds the outer edge of the observation region, and the outer frame member is a frame having sides forming a square or a rectangular right angle, and has an actual size. It is preferable to write a scale.
When photographing is performed including the outer frame material by the photographing means, the actual size of each gravel can be measured from the dimensions of the outer frame material. In addition, distortion occurs in the image when the imaging surface of the camera and the observation target are not kept parallel. In this case, the distortion is corrected by using the image of the outer frame material having a known right angle. Tilt correction can be performed.
By arranging the outer frame material as a scale as described above, the actual size of each gravel during image processing can be easily obtained.

  Secondly, the present invention provides a method for measuring gravel in a river, characterized by observing the gravel in the river using the above-described gravel measuring apparatus.

In the river gravel measurement method of the present invention, after observing the gravel of the surface layer of the observation area with the gravel observation device, the surface of the gravel at a predetermined depth is surfaced by excavating from the surface layer to a predetermined depth. The exposed gravel is photographed and measured by the image processing apparatus, and this is repeated for each set depth position.
In other words, from the viewpoint of fixing the river, the surface layer and the measurement at two positions at a predetermined depth from the surface layer are measured three times in total. The measurement is performed by the measurement device, and the captured image is measured by image processing by a computer. ing.

The effective particle size measured by the measuring device is about 2 mm to 200 mm. In the river sabo technical standards, 200 mm or more recommends manual measurement with a scale at the site, but a method based on binocular stereoscopic image analysis is also effective. On the other hand, sand gravel of 2 mm or less is preferably taken back to the laboratory and analyzed by the conventional method.
That is, the gravel in the range of 2 mm to 200 mm is measured by this measurement method, and the outside of the range is analyzed by the above-mentioned attacking rod, and the range requiring the most labor is automated.

  As described above, according to the gravel measuring apparatus and the measuring method of the present invention, it is only necessary to photograph gravel from a fixed viewpoint in the survey range of the river, and the photographed image is image-analyzed by a computer, and in that case, sedimentation is performed. By extracting shadows along the entire circumference of individual gravel from the collected gravel, the gravel can be separated and identified, and the size and distribution of the gravel can be measured easily and accurately.

  As described above, since it is only necessary to photograph the observation area from the fixed viewpoint, it is possible to eliminate conventional gravel sieving work by the worker from the fixed viewpoint, and to save labor. In particular, since it is only necessary to shoot the surface layer at a point that has been dug a predetermined depth from the surface layer, the three measurement operations from a fixed point of view are more efficient than manual operations by conventional workers. Can be done well.

Hereinafter, the gravel measuring apparatus of this invention is explained in full detail with reference to drawings.
1 to 4 show a first embodiment. As shown in FIG. 1, in a river basin fixing viewpoint R, a digital camera 1 as a photographing means is fixed at a central position P, and the digital camera 1 is used as a center. The lighting devices including the lamps 4 (4A to 4D) are arranged in four directions, front, rear, left, and right at intervals of 90 degrees with substantially the same radius.

The digital camera 1 is fixed to a cross-shaped arm 8 attached to a lower end of a support shaft 7 supported so as to be movable up and down by a bearing 6 supported on the upper end of a tripod 5 and arranged at a central position P of a fixing viewpoint R. , The distance to the ground is adjustable and can be focused.
The lamps 4 (4A to 4D) are fixed to the respective tip positions of the arm 8. In addition, a light emission amount control unit 9 for the lamp 4 is attached to the arm 8. In addition, each lamp 4 is attached to the arm 8 at an angle that irradiates obliquely toward the center position P.

Furthermore, a square-shaped outer frame member 10 is installed on the ground of the fixed viewpoint P with the center position P as the center. On the surface of the outer frame member 10, an actual scale 10a is attached to each side, and the outer frame member 10 is set to an actual scale.
The photographing range by the digital camera 1 is configured to photograph the gravel accumulated on the ground surface of the portion surrounded by the outer frame material 10 including the outer frame material 10.

  The photographing of the fixed viewpoint P by the digital camera 1 is performed by sequentially lighting the lamps 4A → 4B → 4C → 4D with a time interval, and synchronizing the lighting timing of each lamp 4 with the shooting timing of the digital camera 1, Four images are obtained sequentially. These four images enable the formation of shadows in the opposite direction to the irradiation direction by obliquely irradiating the sedimentary gravel from the fixed point of view P with different directions spaced 90 degrees apart from the front, back, left and right. ing.

  Images taken by the digital camera 1 are stored in an image memory in the digital camera 1. The image data stored in the image memory of the digital camera 1 is processed by the image processing means 11 of the computer shown in FIG. 2 according to the procedure shown in the figure after the photographing in the river basin is completed.

The image processing means 11 includes a recording unit 21 that receives an image memory stored in the digital camera 1 and records image data, an arithmetic processing unit 22, and a display unit 23.
The arithmetic processing unit 22 creates an image synthesis unit 24 that creates one minimum luminance composite image from the recorded four pieces of image data, and one minimum luminance image produced by the image synthesis unit 24. A boundary emphasis processing unit 25 that performs a process of emphasizing a boundary line composed of shadows along the outer periphery of each gravel (hereinafter referred to as “gravel” and “particle” in the description of the image processing unit 11), the boundary is emphasized Based on the image, each particle size (granularity) based on an image created by the particle recognition unit 24, which produces particle identification images and particle boundary display images by performing particle separation identification, and the particle recognition unit 26, A gravel analysis unit 27 is provided that calculates the number of distributions in a predetermined particle size range and the weight of gravel, and further calculates the percentage of passage from the weight.

Next, measurement work by the measuring apparatus having the above-described configuration will be described.
In the fixed point P of the river, the four captured images # 1 to # 4 sequentially captured by the digital camera 1 with the lamps 4 from the four directions turned on are in the state shown in FIG. These captured images are taken into the recording unit 21 of the image processing means 11, and the captured four captured images are obtained by the minimum luminance image synthesis unit 24 to obtain the minimum luminance for each pixel. In the step shown, one minimum luminance image # 5 shown in FIG.

The minimum brightness image # 5 is processed by the boundary enhancement processing unit 25 to clarify the contrast between the grain boundary and the part other than the grain boundary.
As this grain boundary emphasis processing, image # 6 shown in FIG. 3C in which a wide boundary line is generated at a boundary part between grains (grain boundary) by spatial differentiation processing is created, The image # 7 shown in FIG. 3D that has been binarized is created, grain boundaries are clearly extracted, and a base of particle recognition with high “robustness” is secured.
Furthermore, in order to further improve the “robustness” of “particle recognition”, labeling processing, expansion processing, and contraction processing are performed to remove noise other than the grain boundary portion.
In the labeling process, the interior of the particle is defined as a “black” region, and the area (number of pixels) of each cluster of “white” therein is obtained, and white clusters having a small area inside the particle are removed as independent noise.
In the expansion process and the contraction process, whisker-like noise near the grain boundary is removed by the expansion and contraction of the grain boundary part.

The image created by the enhancement emphasis processing unit is subjected to processing for recognizing individual particles by separating a pseudo combination of a plurality of particles in a particle recognition unit 26.
First, a potential image is created by calculating the distance from the cluster periphery for each internal pixel surrounded by a black cluster of grain boundaries. In the case of a class in which multiple particles are combined, multiple “mountains” are generated. Next, assuming that the summit of each cluster corresponds to one particle, a particle separation process for separating “mountains”, that is, “particles”, is performed, and then the particles excessively separated in the particle separation process are re-evaluated. The particle recognition image # 8 shown in FIG.
Further, a particle boundary display image # 9 shown in FIG. 3F is generated from the particle recognition image # 6.

Based on the images # 8 and # 9 generated by the particle recognition unit 26, the gravel analysis unit 27 uses a particle size analysis software having a river gravel particle size analysis system to distribute each particle size (particle size) and a predetermined particle size range. The number is measured and the volume weight of the particles is calculated as a percentage of passage.
Specifically, the particle size measurement of the identified particles counts the number of pixels to obtain particle size parameters such as area, circle equivalent diameter, ferret diameter, and the like. The total metric such as distribution, average value, standard deviation, etc. of the particle size parameter is calculated and output. At the time of this calculation, since the outer frame member 10 having an actual scale is also shown in the image, the actual dimensions of each particle can be easily derived.

The image shown in FIG. 3 is obtained by setting the photographing conditions and the image processing conditions as follows.
"Shooting conditions"
Camera: SLR digital camera Resolution: 6 million pixels (3000 x 2000)
Field size: 1000mm x 650mm
Resolution: 0.33mm
Focal length: f = 24mm
Lighting: Strobe 4-way lighting “Image processing”
Notebook PC: CPU Pentium (registered trademark) 4 (2GHz)
Processing method Grain boundary extraction by minimum luminance image synthesis Particle recognition / particle size measurement Percentage calculation Calculation time: Approximately 60 seconds / screen

In this way, the measurement of sedimentary gravel in rivers is performed by image processing by the image processing means 11 using a plurality of images with different shadow directions taken in the field, and the particle size, distribution, and volume weight of the gravel are measured.
Shooting performed by illuminating the illumination means is performed for the first time on the surface, the second time at a point that has been dug a predetermined depth from the surface, and the third time at a point that has been further dug from the second point. The image data recorded in the digital camera 1 by taking these three times is brought back, and the state of the accumulated gravel is measured by image processing by the image processing means.

  The gravel whose particle size is measured by the gravel analyzing unit 24 of the image processing means 11 is a medium size 2 mm to 200 mm sand gravel that can be efficiently measured by image processing. Large gravels of 200 mm or more are measured by binocular stereoscopic vision, and 2 mm or less are analyzed by a conventional method in a laboratory.

As described above, the lamps 4A to AD are sequentially turned on at the site of the river fixation point P, and the digital camera 1 takes images four times in synchronism with the lighting of each strobe. Compared to the case where conventional investigators measure sand and gravel using a sieve at the site, the on-site survey time can be greatly reduced, and analysis with a small amount of take-out sample is possible. It becomes possible.
In addition, since each gravel is separated and identified from the gravel deposited by the image processing means 11 and the particle size, particle size distribution, and volume weight are obtained based on this, accurate measurement can be obtained.

FIG. 5 shows a modification of the first embodiment. A strobe is used instead of the lamp, and the sequential switching illumination time of the strobes 4A ′ to 4B ′ and the photographing time by the digital camera 1 are controlled by the sequence control circuit. .
In this modification, the illumination / imaging / image recording timing and the granularity calculation sequence are all controlled by the sequence control circuit 30 driven by the synchronization signal generated by the synchronization signal generation circuit 33.
The strobes 4A ′ to 4D ′ are lit based on a lighting signal output from a lighting controller 31 controlled by the sequence control circuit 30.

  In measurement, the sequence control circuit 30 sequentially turns on the strobes 4A ′ to 4D ′, and records the accumulated gravel image captured by the digital camera 1 in synchronization with the lighting, via the camera controller 32. 21 as image memories # 1 to # 4. When the image input to the recording unit 21 is completed, the sequence control circuit 30 outputs a calculation command to the calculation processing unit 22. The configuration of the arithmetic processing unit 22 is the same as that of the first embodiment. The built-in minimum luminance image synthesis circuit 24 reads the images in the image memories # 1 to # 4, calculates the minimum luminance image, and stores the minimum luminance image in the memory. Write. Thereafter, the boundary emphasis processing unit 25 forms an image in which the shadow lines of the grain boundaries are emphasized, the particle recognition unit 26 separates and identifies the individual particles, and the analysis unit 27 calculates and outputs the particle size.

FIG. 6 shows a second embodiment of the present invention. In the second embodiment, the illumination means composed of a lamp is used as one unit to move and illuminate.
That is, in the present embodiment, one lamp 4 attached to the rotation unit 40 as an illumination system is connected to the rotation unit drive circuit via the lamp position control circuit 41 based on the control signal output from the sequence control circuit 30. While rotating by 42, images of sedimentary gravel are taken at positions in four directions.
Since other configurations are the same as those of the modification of the first embodiment, the same reference numerals are given and description thereof is omitted.

7 to 9 show a third embodiment of the present invention.
The third embodiment is an apparatus that can simultaneously illuminate images from four directions by performing simultaneous illumination with separated spectrum light.
That is, in this embodiment, four strobes 4A ′ to 4D ′ arranged at intervals of 90 degrees with the digital camera 1 as the center are provided with four color filters 50A to 50D having different spectral transmission characteristics as shown in FIG. Each is attached.

On the other hand, the digital camera has a structure shown in FIG. 8A or 8B.
In the digital camera 1 ′ shown in FIG. 8A, four image sensors 52 </ b> A to 52 </ b> D having different spectral spectral sensitivities and the same optical axis, and colors attached to the front surfaces of the image sensors 52 </ b> A to 52 </ b> D. Filters 50A ′ to 50D ′ and a half mirror 54 are incorporated. In the digital camera 1 ′, the optical axis 60 at the time of photographing is the same optical axis, and the optical axis 60 is divided into four half mirrors 54 so as to coincide with the optical axes of the four image sensors 52 </ b> A to 52 </ b> D.
The digital camera 1 ″ shown in FIG. 8B is composed of four small digital cameras 1a ″ to 1d ″ having different spectral spectral sensitivities, and a color filter is provided in front of each of the digital cameras 1a ″ to 1d ″. 50A ″ to 50D ″ (the color filter 50C ″ is not shown on the back side and is not shown) is attached, and the optical axes 60a are bundled so as to be in substantially the same direction.

  As shown in FIG. 9, the four strobes 4 </ b> A ′ to 4 </ b> D ′ are turned on all at once by the lighting signal output from the lighting controller 31 controlled by the sequence control circuit 30. The digital camera 1 ′ (1 ″) lights the strobes 4 A to 4 D by the lighting trigger signal output from the lighting controller 31 controlled by the sequence control circuit 30, and is also controlled by the camera controller 32 controlled by the sequence control circuit 30. The electronic shutter is configured to be capable of instantaneous imaging with the electronic shutter in synchronization with the lighting of the strobe, and the gravel gravel when the strobes 4A ′ to 4D ′ are turned on is photographed. The image memories # 1 to # 4 are simultaneously loaded into the recording unit 21. The process of processing the captured image by the arithmetic processing unit 22 and analyzing the particle size, distribution state, etc. of the gravel is the same as in the first embodiment. It is.

In this way, if the spectral transmission characteristics of the four color filters mounted on the strobe of the illumination means and the image sensor of the digital camera are separated from each other as a spectrum, the accumulated gravel image obtained by each image sensor is the image sensor. Since the image is illuminated by a flash equipped with the same color filter, a four-way illumination image is simultaneously captured in the image memory.
If the field of view between the four images is a problem when photographed with the digital camera 1 ″ shown in FIG. 8B, each image is shifted by the amount of field of view displacement before the arithmetic processing unit 22. A visual field shift correction circuit unit 36 may be provided.

  In the first to fourth embodiments, the case of four-way illumination has been described. However, the present invention is not limited to four directions, and it is not limited to four directions. As long as a shadow is obtained, there may be three directions or five or more directions.

  The present invention is a device for measuring gravel accumulated in rivers, but can also be used for measurement of accumulated granular material other than gravel, and in a state where particles are randomly deposited, the particle size of each particle, particle size distribution, It can be used effectively when the total weight is obtained from the volume occupancy.

It is a perspective view which shows the imaging | photography means and illumination means of 1st Embodiment of this invention. It is drawing which shows the image processing means of 1st Embodiment. (A)-(F) are the photographs which show each example of an image. It is drawing which shows the synthetic | combination procedure of the minimum brightness | luminance image from several illumination images. It is whole explanatory drawing which shows the modification of 1st Embodiment. It is whole explanatory drawing which shows 2nd Embodiment. It is whole explanatory drawing which shows 3rd Embodiment. (A) and (B) are drawings showing a configuration of a digital camera used in the third embodiment. It is a graph which shows the relationship between the color filter spectral transmission characteristic and spectral transmittance which are used in 3rd Embodiment. It is drawing which shows the tomographic plane of a river.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Digital camera 4 (4A-4D) Lamp 4 '(4A'-4D') Strobe 10 Outer frame material 11 Image processing means 21 Recording part 22 Calculation part 23 Display part 24 Image composition part 25 Boundary emphasis processing part 26 Particle recognition part 27 Gravel analysis section 30 Sequence control circuit 33 Synchronization signal generation circuit

Claims (9)

Photography means installed in the observation area where gravel is accumulated,
Illuminating means for illuminating the observation area from different directions while leaving a required dimension from the photographing means, and to change the position of the shadow generated in the photographed image,
A plurality of pieces of image data photographed from the same direction by the photographing means are recorded under irradiation from different directions by the illumination means, and the gravel in the observation area is separated and identified based on the shading of the image data. And a gravel measurement device comprising an image processing means for measuring the particle size (radius) of each gravel and analyzing the distribution. The image processing means includes a recording section for receiving an image from the photographing means and recording it as image data, an arithmetic processing section for the recorded image data, and a display for displaying an image of a measurement result calculated by the arithmetic processing section. The arithmetic processing unit comprises:
An image composition unit that creates one minimum brightness composite image from a plurality of recorded images, a boundary emphasized image creation unit that emphasizes the outline of the gravel from the minimum brightness composite image, and a boundary emphasized image creation unit A particle recognition unit that separates and identifies each particle based on the produced one image to produce a particle identification image and a particle boundary display image;
A gravel analysis unit that calculates the particle size (radius) of each gravel, the number of distributions in a predetermined particle size range, the gravel weight based on the image created by the particle recognition unit, and further calculates the passing percentage from the weight. The gravel measuring apparatus according to claim 1. The photographing means comprises one camera arranged at a required height position from the ground surface at the central position of the observation area,
The illumination unit includes a plurality of lamps or strobes that irradiate an observation area from at least three directions around the photographing unit, and fixed at an angular interval around the photographing unit, or Provided with one unit that is rotated about the photographing means and stopped at an angular interval,
Under the illumination by the illuminating means, a plurality of image data taken sequentially from the same direction by the photographing means are combined to extract the minimum luminance for each pixel, and a shadow along the entire circumference of each gravel is obtained. The gravel measuring apparatus according to claim 1 or 2.   The gravel measurement apparatus according to claim 3, further comprising a synchronization signal generating unit that synchronizes a shutter timing by the photographing unit and an irradiation timing performed by sequentially switching an irradiation direction by the illumination unit. The imaging means includes one digital camera having a plurality of imaging elements having different spectral spectral sensitivities and having optical axes in the same direction, or a plurality of digital cameras having different spectral spectral sensitivities. On the other hand, it is configured to be bundled so that they are in substantially the same direction,
The plurality of illuminating units arranged at angular intervals with the imaging unit as a center are respectively spectral spectral sensitivities of each imaging element of one digital camera of the imaging unit or spectral spectra of a plurality of digital cameras. It is configured to emit light corresponding to the sensitivity,
3. The gravel measuring apparatus according to claim 1, wherein illumination by the plurality of illumination units is simultaneously performed during imaging by the imaging unit.   The color filter of a different color is attached to each of the illumination means, and a color filter of a different color is attached to each imaging element of the one digital camera or a plurality of digital cameras. The gravel measuring apparatus described.   Provided with an outer frame material that is arranged on the ground of the observation area and surrounds the outer edge of the observation area, the actual size is written on the outer frame material as a scale, and the outer frame material is photographed by the photographing means, The gravel measurement apparatus according to any one of claims 1 to 6, wherein an actual size of each gravel can be measured from the dimensions of the outer frame member and the inclination can be corrected.   A method for measuring gravel in a river, wherein the gravel in the river is observed using the gravel measurement apparatus according to any one of claims 1 to 7.   After observing the gravel on the surface layer of the observation area with the gravel observation device, dig up to a predetermined depth from the surface layer to expose the gravel at a predetermined depth over the entire surface of the observation area, and photograph the exposed gravel. The river gravel measurement method according to claim 8, wherein measurement processing is performed by the image processing apparatus, and this is repeated for each set depth position.