JP2009294178A - Apparatus and method for measuring sand - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sand measuring apparatus and a sand measuring method capable of measuring a ratio of labelled sand particles slightly included in a large amount of earth and sand. <P>SOLUTION: Sand is collected at a measuring site, and labelled sand is arranged at an initial location in this method. Sand at a location separated from the initial location is collected after a prescribed period has passed, and collected sand is transferred and imaged during its transfer. The labeled sand is counted on the basis of images obtained by imaging. Imaging is performing by simultaneously measuring sand images each transmitted through two filters having different transmission wavelength bands. When images transmitted through each filter are obtained as a first image and a second image, it is determined that pixels have been receiving light from labelled sand particles when a ratio between the intensity of pixels in the first image and the intensity of pixels in the second image is equal to a prescribed value or greater. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a sand measuring device and a sand measuring method.

  In various civil engineering works, the movement of earth and sand is precisely measured, and the optimum civil engineering design is being performed in consideration of the measurement results. In addition, the movement of earth and sand can be measured after civil engineering work, which can be used for verification of civil engineering works. Measuring the movement of sediment for purely scientific investigations can also help predict aging of coastlines, riverlines or dunes. Civil engineering technology in Japan is advancing worldwide, but if it can accurately measure the movement of earth and sand, it will be more advanced not only in Japan but also in other countries where civil engineering works are required. It is expected to be able to conduct efficient construction and research. In order to grasp the erosion state of the coast and the runoff state of sand and sand after beach nourishment, it is expected to utilize the sand movement measurement method (monitoring). As such a measuring method, (1) sand movement measurement by physical / chemical analysis of local sand, (2) sand movement measurement using color sand or fluorescent sand as a tracer can be considered.

  To measure the movement of earth and sand, place the marked earth and sand at a specific point, collect the earth and sand at multiple measurement points away from this point, and analyze the earth and sand at each measurement point. Then, the number of labeled sand particles contained in the earth and sand is counted. If the count number and the size of the sand particles at each measurement point are known, it should be understood what size of the sand particles have been diffused and moved from the specific initial point.

  However, the number of sand grains present on the beach etc. is enormous, and it is impossible or even possible to accurately count the labeled sand that is slightly present in it. Need. Conventionally, precise measurement of sand particles contained in a large amount of earth and sand has not been performed. The present invention has been made in view of such problems, and provides a sand measuring device and a sand measuring method capable of measuring the proportion of labeled sand particles slightly contained in a large amount of earth and sand. The purpose is to provide.

  In order to solve the above-described problems, a sand measuring device according to the present invention is a sand measuring device containing labeled sand particles. And a calculating means for discriminating and counting the labeled sand particles from the sand image picked up by the image pickup means. “Sand” means an aggregate of “sand grains”. As a labeling method, there is a method of painting the surface of sand particles with a pigment or a fluorescent paint.

  The labeled sand grains have a diffuse reflectance spectrum that is different from the surrounding sand grains that are not. Therefore, in the image picked up by the image pickup means, the pixel group indicating the labeled sand particles can be distinguished from the pixel group indicating the surrounding sand. Therefore, the computing means discriminates the labeled sand particles and counts them.

  Since the total number of all sand grains contained in one frame image is the number of pixels in the image when these are converted into pixels, the proportion of the number of labeled sand grains in the total number of pixels in one frame Becomes clear. This indirectly indicates the ratio of labeled sand particles contained in the sand to be measured. Also, if the sand grains in the image are indicated by a plurality of pixels, the number of sand grains composed of a plurality of pixels in one frame occupies a certain ratio to the total number of sand grains in one frame. Yes. That is, the ratio of the number of labeled sand grains to the total number of sand grains is determined. Note that the total number of sand grains can be a value obtained by dividing the total number of pixels in one frame by the number of pixels constituting one sand grain. As the size of the sand particles (converted value of the number of pixels), an average value of the size of the sand particles can be used.

  This ratio can also be handled as a value obtained by converting the total number N of sand grains detected in a plurality of frames F into a number per unit frame, that is, N / F (pieces / frame), and is included in one frame. The number of sand grains can be replaced with the value of N as required.

  Thus, in the present invention, it is possible to roughly measure the proportion of sand particles contained in the sand to be measured.

  In addition, the imaging means includes a first filter having a first transmission wavelength band, a second filter having a second transmission wavelength band different from the first filter, and a specific region being conveyed that has passed through the first filter. A first camera that picks up an image of sand, and a second camera that picks up an image of sand in the specific area being conveyed that has passed through the second filter at the same time as the first camera. When the ratio of the intensity of the pixel in the first image output from the first camera and the intensity of the pixel in the second image output from the second camera is equal to or greater than a predetermined value, the sand particle labeled with the pixel It is preferable to determine that the light from the light is received. In the calculation of the pixel intensity ratio, the lower intensity is used as the denominator.

  Labeled sand grains have a diffuse reflectance spectrum different from that of unlabeled sand grains. Therefore, when the sand is imaged in different wavelength bands, and the ratio of the intensity of each corresponding pixel constituting these images is equal to or greater than a predetermined value, this pixel indicates a labeled sand particle, It can be determined accurately.

  In addition, in the case where pixels having a predetermined value or more are consecutive in one frame image composed of a group of pixels that are imaged by the imaging unit and the intensity ratio is calculated, the calculation unit calculates the region where the continuous pixels are distributed. The number of pixels is measured as an index of the size of the labeled sand particles, the number of sand particles for each size of the sand particles is counted, and further, for each size of the sand particles included in the multi-frame image captured by the imaging means. It is preferable to integrate and output the number of sand grains.

  That is, the number of labeled sand grains contained in a large amount of sand is often very small. The number can be one per million. Also, if the size of the sand particles labeled in such a state is found, it is possible to analyze the sand diffusion distribution according to the size of the sand particles. Therefore, as described above, sand particles obtained from a plurality of frames of images are integrated for each size and output. Since the measurement data is not data obtained by measuring all the sand grains, it has a variation with respect to the true value. However, as the integration number is increased, the data approaches the true value.

  In addition, the measurement data includes errors caused by misjudging as labeled sand particles, but by counting the number of sand images that do not include labeled sand particles in the same way, the count number of error components is Since the calculation can be performed in advance, the data can be brought close to a true value by subtracting the error component count from the obtained count.

  The conveying means includes a belt conveyor having a pair of side walls extending in parallel with the sand conveying direction and conveying sand, and a sand supplying means for flowing sand onto the belt conveyor, and the sand supplying means on the belt conveyor. The apparent volume V of sand that flows in from unit time to time, the average particle size D of sand grains, the amount of movement L that the belt conveyor advances in unit time, and the distance W between the pair of side walls are W × L × D <V It is preferable that the formula is satisfied. It should be noted that the apparent volume (the bulk volume) is a volume of a container having a size that is completely filled with sand, and is an apparent volume including a gap between the sands.

  When the above-mentioned conditions are satisfied, the sand is supplied so that one or more layers of sand particles are stacked in the region on the belt conveyor surrounded by the side walls. That is, the sand flowing from the sand supply means onto the belt conveyor is laminated in a single layer or more on the belt conveyor. In this case, since the sand particles are spread on the belt conveyor, the ratio of the labeled sand particles to the whole can be accurately measured.

  In order to spread sand grains as described above, it is preferable to have means for leveling the sand. That is, the conveying means includes a first flattening member having a planar bottom surface separated from the surface of the belt conveyor and a convex portion whose width becomes narrower in the direction opposite to the conveying direction of the belt conveyor, It is preferable that a gap exists between the flattening member and the side wall.

  In this case, the sand collides with the convex portion from the leading side (protruding tip end) of the first flattening member along with the conveyance, and a part of the sand is in contact with the bottom surface of the first flattening member and the belt. It sinks into the space between the surfaces of the conveyor, and the remainder is scraped in the width direction by the projections, but also sinks into the space between the bottom surface of the first flattening member and the surface of the belt conveyor and is flattened. Furthermore, since excess sand is conveyed through the gap between the side wall of the belt conveyor and the first flattening member, the sand does not overflow from the belt conveyor.

  Moreover, it is preferable that a conveyance means is further provided with the spatula-shaped 2nd planarization member which is arrange | positioned at the conveyance direction side of the belt conveyor of a 1st planarization member, and consists of an elastic body spaced apart from the surface of a belt conveyor.

The sand flattened by the first flattening member is further flattened by the second flattening member. That is, the shape of the second flattening member is a spatula shape made of an elastic body such as rubber, and the pressure applied by the first flattening member as the elastic body sweeps the surface of the sand as it is conveyed. The sand that has been slightly pushed in by is dispersed in a natural state, and the flatness of the sand is further increased.
The sand supply means further includes a container having a sand output port at the bottom, and a screw conveyor extending from the sand output port and discharging the sand onto the belt conveyor, and a brush is attached to a screw blade of the screw conveyor. Preferably it is.

  When sand is laminated on a belt conveyor, more accurate measurement cannot be performed unless the sand is sufficiently stirred. That is, if the probability that the labeled sand is always buried in a position where it cannot be imaged due to the layer of sand increases, the measurement data may not be accurate. Therefore, in the above description, sand is conveyed by a screw conveyor to agitate the sand grains so that more accurate data can be obtained. In addition, when the number of times of imaging reaches a plurality, the surface of the sand itself is shaved on the surface of the blade while being conveyed many times by a normal screw blade. That is, the state of the object to be measured, such as the size of sand, changes, and accurate measurement cannot be performed. Therefore, since the brush is attached to the blade so that the brush hits the sand, the wear of the sand particles is reduced, and accurate measurement is possible.

  Further, the transport means is a sand transporter for returning the sand transported by the belt conveyor into the sand supply means. It is preferable to provide a carrying device. If there is a sand transporting device, the sand transporting device can transport the sand once imaged into the sand supply means, so that the sand can be imaged many times and the measurement accuracy can be improved.

  Of course, the belt conveyor may extend annularly. When the belt conveyor is conveyed in an annular shape, the sand once imaged can be imaged again because it returns to the same position as the belt conveyor moves in an annular shape. This imaging can be repeated as many times as necessary.

  Moreover, the sand measuring method according to the present invention includes a first step of collecting sand at a measurement site, a second step of placing labeled sand at an initial point, and a distance from the initial point after a given period of time. 3 steps of collecting the sand at the point, a fourth step of imaging while conveying the sand collected in the third step, and a fifth step of counting the labeled sand from the captured image. And

  When measuring the behavior of sand for environmental assessment and civil engineering construction surveys, the sand at the measurement site is collected, marked, and placed at the initial reference point. For example, when a predetermined period of time such as one month later has elapsed, sand at a point away from the initial point is sampled and imaged, and the number of labeled sand is counted to determine the state of sand diffusion.

  The imaging in the fourth step is performed by simultaneously measuring sand images that have passed through two filters having different transmission wavelength bands, and the images that have passed through the filters are referred to as a first image and a second image. Then, in the fifth step, when the ratio of the intensity of the pixel in the first image and the intensity of the pixel in the second image is equal to or greater than a predetermined value, the light from the sand grain labeled with the pixel is received. It is preferable to discriminate them from those.

  In this case, since the labeled sand particles have a spectrum different from that of the unlabeled sand particles, when the intensity ratio of the corresponding pixels constituting the first image and the second image is equal to or greater than a predetermined value, this pixel Can indicate labeled sand particles and can be accurately determined.

  Further, in the sand measuring method according to the present invention, in the case where pixels that are equal to or greater than the predetermined value are consecutive in one frame image that is a group of pixels that are imaged and the ratio of the intensity is calculated, The number of pixels in the area where the particles are distributed is measured as an indicator of the size of the labeled sand particles, the number of sand particles for each size of the sand particles is counted, and the size of the sand particles included in the captured multiple frame images It is preferable to integrate and output the number of sand grains for each thickness.

  If the size of the labeled sand particles is known, it is possible to analyze the sand diffusion distribution according to the size of the sand particles. Further, since the measurement data is not data obtained by measuring all sand grains, the measurement data has a variation with respect to the true value, but the data approaches the true value as the integration number is increased. In addition, in order to obtain labeled sand, the above-described measurement method may further include a step of marking sand.

  According to the sand measuring apparatus and measuring method of the present invention, it is possible to measure the ratio of labeled sand particles.

  Hereinafter, the sand measuring device and the measuring method according to the embodiment will be described. The same reference numerals are used for the same elements, and redundant description is omitted.

  FIG. 1 is a block diagram of a sand measuring device.

  The sand 3 measured by this measuring apparatus contains labeled sand particles (hereinafter referred to as “labeled sand particles”). “Sand” means an aggregate of “sand grains”. As a method of labeling, there is a method of painting the surface of sand particles with a pigment or a fluorescent paint. Hereinafter, an example of colored sand coated with a red or blue pigment will be described.

  The sand 3 falls from the sand hopper (sand supply means) 1 and is discharged onto the belt conveyor 2. The belt conveyor 2 moves by rotating the motor 20 engaged with the belt of the belt conveyor 2. The belt of the belt conveyor 2 moves on the outer surface of one or more transport rollers 21. The belt conveyor 2 is a part of a conveying device (conveying means) and conveys the sand 3. The sand 3 being conveyed on the belt conveyor 2 is imaged by an imaging device (imaging means) including cameras 5 and 6 formed of digital video cameras.

  The cameras 5 and 6 simultaneously image the sand 3 in the same imaging area on the belt conveyor 2. The camera 5 acquires an image (first image) of sand that has passed through the filter FL1 having the first transmission wavelength band λ1, and the camera 6 has an image of sand that has passed through the filter FL2 having the second transmission wavelength band λ2 ( 2nd image) is acquired. The first and second images of the sand 3 in the same imaging area captured by this imaging device are input to the image processing device 9. The image processing device 9 calculates a luminance ratio for each corresponding pixel, and outputs a composite image (each pixel has a value indicating the luminance ratio) as a calculation result to the computer 11. The image processing circuit 9 may be built in the computer 11.

  Moreover, the imaged sand 3 falls on a sand receiving device 4 such as a sand receiving cup.

  The computer (arithmetic unit) 11 determines that a pixel having a value equal to or greater than a predetermined value in the composite image is an image of the labeled sand particle, discriminates the labeled sand particle in this way, counts the number, The results are displayed on the monitor 10 for each size of the marked sand particles.

  The imaging device will be described in detail. The optical axis of the photographing lens LS is positioned on a straight line perpendicular to the surface of the belt conveyor 2, and a half mirror HM made of a prism or the like is disposed behind the photographing lens LS. The transmitted light of HM enters the camera 5 through the filter FL1, and the reflected light of the half mirror HM enters the camera 6 through the filter FL2. The cameras 5 and 6 have the same structure, but the distance from the taking lens LS is different. In other words, since the transmission wavelength bands of the filters FL1 and FL2 are different, the imaging positions of the transmitted lights are different, and the positions of the cameras 5 and 6 are adjusted to the imaging positions of the transmitted lights.

  The optical axes of the cameras 5 and 6 form an angle of 45 degrees with respect to the surface of the half mirror HM and are orthogonal to each other. The camera 6 is fixed to a three-dimensional moving stage, and this stage can also be moved in the rotation direction of the optical axis. The imaging lens LS is FA645, 45 mm, F2.8 manufactured by Pentax Corporation, and the optical bandpass filters FL1, FL2 are filters manufactured by Asahi Spectroscopy (transmission wavelength band λ1 = 400-540 nm, λ2 = 580-700 nm. The progressive VGA and XC-HR57 manufactured by Sony Corporation can be used as the cameras 5 and 6 for measuring sand as a moving object. The distance from the imaging lens LS to the surface of the belt conveyor 2 is about 30 cm.

  The light sources LP1 and LP2 emit light in synchronization with the timing of opening the shutters of the cameras 5 and 6 incorporating the CCD. The light sources LP1 and LP2 in this example are xenon flash lamps, and diffusion plates DF1 and DF2 are interposed between the light sources LP1 and LP2 and the imaging area of the belt conveyor 2, respectively. The diffusion plates DF1 and DF2 are preferably light emission plates of a panel strobe with built-in light sources LP1 and LP2. In this apparatus, the shutter speed of the cameras 5 and 6 is set to 1/4000 (seconds), for example, because the belt conveyor 2 performs imaging while continuously moving at a constant speed so as not to disturb the sand state.

  When an imaging trigger signal is input from the timing circuit 7 to the cameras 5 and 6 and the lamp driving circuit 8, the lamp driving circuit 8 outputs a strobe driving signal, and the cameras 5 and 6 open the shutter. After a lapse of time slightly delayed from the image acquisition timing at this time, the acquired images of the cameras 5 and 6 are transferred to the image processing circuit 9 and a composite image is calculated. The calculated composite image and / or its binarized image is sequentially stored in the image processing circuit 9 or a storage device (not shown) in the computer 9. The timing circuit 7 is controlled from the computer 11.

Note that the imaging timing interval t _INTERVAL is set longer than the period until all the sand in the imaging area moves outside the imaging area, so that the previous sand image and the current sand image do not overlap. Has been. That is, the distance in the conveying direction of the imaging area set on the belt conveyor 2 and _{L IM,} when the conveying speed of the belt conveyor 2 and _{S BELT,} meets _{t INTERVAL> L IM / S BELT} .

  Specifically, it is assumed that the size of the imaging area on the belt conveyor 2 is 3 cm × 4 cm, and the range in the imaging area is continuously photographed. The number of sand grains in the area of the imaging region is approximately 10,000 grains (however, only sand grains on the surface layer) assuming that the grain size is 0.04 cm. At that time, in order to prevent the field of view from overlapping in the direction in which the belt conveyor moves, the imaging by the cameras 5 and 6 is performed at 4 frames / second, and the conveyance speed of the belt conveyor 2 is set to 16 cm / second or more. .

  In addition, the sand before measurement is dried. Below, the sand drying method will be described briefly. 0.5 liters of water was mixed well with 3 liters of sand from the Nakatajima dunes in Hamamatsu City, Shizuoka Prefecture, and the mixture was introduced into a drying drum and rotated at a rotational speed of once per 5 seconds. The hot air temperature at the exhaust port of the sand dryer was 120 ° C. After 30 minutes from the start of operation, the sand of the Nakatajima dune was evaluated, and this sand was completely dry. When the amount of water was increased to 0.6 liter, the sand was not completely dried after drying for 30 minutes, but was completely dried after 40 minutes. In addition, the specific gravity of the sand of the Nakatajima dune is about 1.44, and the weight of 3 liters is 4.32 kg. The sand supplied initially is such dry sand.

  Next, generation of the composite image and counting of the marked sand particles will be described.

  FIG. 2 is a graph showing reflection spectra from various types of sand. The horizontal axis represents wavelength λ (nm), and the vertical axis represents reflected light intensity R (arbitrary constant).

  D0 indicates the reflection spectrum of sand collected from Nakatajima dune in Hamamatsu City, Shizuoka Prefecture, D1 is the reflection spectrum of red sand painted in red, and D2 is yellow sand coated in yellow. , D3 represents the reflection spectrum of green sand obtained by painting this sand in green, and D4 represents the reflection spectrum of blue sand obtained by painting this sand in blue. Moreover, D5 shows the reflection spectrum of sand of Izu scoria (lava stone), and D6 shows the reflection spectrum of sand of diorite.

  Thus, the labeled sand grains indicated by D1 to D4 have a reflection spectrum different from that of the sand of the Nakatajima dune before painting. On the belt conveyor, marked sand particles will be mixed with the sand of the unpainted Nakatajima dune. That is, the reflection spectrum of the sand particles of the painted Nakatajima dune has a reflected light spectrum different from the sand particles around the sand particles. Accordingly, in the image captured by the imaging device, the pixel group indicating the marked sand particles can be distinguished from the pixel group indicating the surrounding sand. Therefore, the computer 11 discriminates the count sand particles and counts them.

  In the present invention, it is important to compare the marked sand grains with the surrounding sand grains. Therefore, in the graph shown in FIG. 2, the value obtained by dividing the intensity at each wavelength of the sand indicated by D1 to D6 by the intensity of the data D0 of the Nakatajima dune at this wavelength is relative to the reflection spectrum indicating the reference data D0. Obtained as a reflection spectrum.

  3 (a) and 3 (b) are graphs showing relative reflection spectra, and FIG. 4 is a wavelength (maximum wavelength) that gives the maximum value of the reflection spectrum of each sand, and a wavelength (minimum wavelength) that gives the minimum value. ). The maximum value and the minimum value can be obtained when a red filter, a green filter, or a blue filter is used, as shown in the table. FIG. 3A is a graph when the determination is performed using the blue filter and the red filter, and FIG. 3B is a graph when the determination is performed using the green filter and the red filter. In FIG. 3A, the transmission wavelength band λ1 (RED) of the red filter is 560 to 700 nm, and the transmission wavelength band λ2 (BLUE) of the blue filter is 400 to 540 nm. In FIG. 3B, the transmission wavelength band λ1 (RED) of the red filter is 580 nm to 700 nm, and the transmission wavelength band λ2 (GREEN) of the green filter is 440 nm to 560 nm.

  In addition, let the intensity | strength R of the reflection spectrum of the reference | standard Nakatajima dune be 100 (arbitrary constant).

  The ratio between the maximum and minimum values is shown at the right end of the table.

  When the reflection spectrum of red sand (D1) is measured, these ratios (maximum value / minimum value) = 198.4 / 49.1 = 4.04 are obtained. When the reflection spectrum of yellow sand (D2) is measured, these ratios (maximum value / minimum value) = 151.3 / 45.6 = 3.32. When the reflection spectrum of green sand (D3) is measured, these ratios (maximum value / minimum value) = 98.2 / 36.8 = 2.67 are obtained. When the reflection spectrum of blue sand (D4) is measured, these ratios (maximum value / minimum value) = 226.3 / 49.1 = 4.61 are obtained. When the reflection spectrum of Izu scoria (D5) is measured, the ratio (maximum value / minimum value) = 68.6 / 31.5 = 2.18 is obtained. When the reflection spectrum of bright green sand (D6) is measured, these ratios (maximum value / minimum value) = 124.2 / 99.6 = 1.25 are obtained. Since the reflection spectrum intensity of the Nakatajima dunes is constant, the maximum value / minimum value is 1.00.

  Thus, by taking the ratio between the maximum value and the minimum value of the relative reflection spectrum, it is possible to determine the characteristics of each type of sand. For example, when blue sand (D4) is used and blue sand is mixed with sand of the Nakatajima dune, the threshold TH is set to 4 and sand particles having the above ratio of 4 or more are recognized as blue sand. It is also possible to certify sand less than 4 as sand (D0) of the Nakatajima dune. Note that the maximum value and the minimum value are the integral value of the spectrum intensity in the transmission wavelength band of the filter including the maximum value and the integral value of the spectrum intensity in the transmission wavelength band of the filter including the minimum value, respectively. it can. Thus, by setting a wide wavelength band, the amount of received light can be increased and the measurement accuracy can be improved.

As shown in the table, when blue sand is used, a high ratio (discrimination rate) with respect to sand of Nakatajima dune can be obtained. Subsequently, it has a high ratio in the order of red, yellow, and green.
If the intensity of each pixel at the coordinates (x, y) in the first image is I1 (x, y) and the intensity of each pixel at the coordinates (x, y) in the second image is I2 (x, y) The value C (x, y) of each pixel at the coordinates (x, y) of the image = I1 / I2 (x, y), and I1> I2. When C (x, y) ≧ TH is satisfied, it is recognized that the light is from the labeled sand particles, and when C (x, y) <TH is satisfied, the light is from surrounding sand particles that are not labeled. Certify. In other words, in the present embodiment, the image processing circuit or the computer performs binarization processing on the composite image having the ratio value with reference to the discrimination threshold TH.

  The binarized image specifically shows light from only sand particles. The total number of all sand particles contained in one frame image is the total number of pixels in the image when these are converted into pixels. Therefore, when the number of pixels occupied by the marked sand particles in one frame is divided by the total number of pixels in one frame, the proportion of the number of pixels of the marked sand particles is found.

  This indirectly indicates the ratio of labeled sand particles contained in the sand to be measured. The sand grains in the image are indicated by a plurality of pixels, and the number of sand grains composed of the plurality of pixels in one frame occupies a certain ratio to the total number of sand grains in one frame. . That is, the ratio of the number of labeled sand grains to the total number of sand grains is determined. Note that the total number of sand grains can be a value obtained by dividing the total number of pixels in one frame by the number of pixels constituting one sand grain. As the size of the sand particles (converted value of the number of pixels), an average value of the size of the sand particles can be used. This average value may be obtained in advance, but can also be obtained by taking the average of the size of the marked sand particles finally measured.

  This ratio can also be handled as a value obtained by converting the total number N of sand grains detected in a plurality of frames F into a number per unit frame, that is, N / F (pieces / frame), and is included in one frame. The number of sand grains can be replaced with the value of N as required. Thus, according to the measurement method described above, it is possible to roughly measure the proportion of sand particles contained in the sand to be measured.

  In addition, the imaging apparatus includes a first filter FL1 having a first transmission wavelength band, a second filter FL2 having a second (or third) transmission wavelength band different from the first filter FL1, and a first filter FL1. A first camera 5 that captures the image of the sand in the specific area that is being transmitted and the sand camera that is in the specific area that is being transmitted that has passed through the second filter FL2 simultaneously with the first camera 5. And the computer 11 includes a pixel intensity I1 (x, y) in the first image output from the first camera 5 and a second image in the second image output from the second camera 6. When the ratio C (x, y) of the intensity I2 (x, y) of the pixel is equal to or greater than a predetermined value (C (x, y) ≧ TH), the pixel receives light from the marked sand particles. Is determined. In the calculation of the pixel intensity ratio, the lower intensity I2 is used as the denominator.

The intensities I1 and I2 are relative reflection spectrum intensities I1 (= i1 (λ _MAX ) / i0 (λ _MAX )) and I2 (= i2 (λ _MIN ) / i0 (λ _MIN )). i1 (λ _MAX ) is the intensity of the reflected spectrum from the labeled sand grains at the maximum wavelength λ _MAX (raw data before conversion to relative values), i0 (λ _MAX ) is the surrounding sand at the maximum wavelength λ _MAX (Nakatajima dune Reflection spectrum intensity from raw sand) (raw data before conversion to relative value), i2 (λ _MIN ) is reflection spectral intensity from labeled sand grains at minimum wavelength λ _MIN (raw data before conversion to relative value) , I0 (λ _MIN ) is a reflection spectrum intensity (raw data before conversion to a relative value) from surrounding sand (sand of Nakatajima dune) at the minimum wavelength λ _MIN . The raw data includes 16-channel linear photomultiplier tubes (a strip-shaped cathode is linearly arranged for 1 to 16 channels, and an interference filter every 20 nm from 400 to 700 nm is attached to each channel. It was obtained using a photomultiplier tube.

That is, C (x, y) = I1 / I2 (x, y) = [(i1 (λ _MAX ) / i0 (λ _MAX )) / (i2 (λ _MIN ) / i0 (λ _MIN ))] (x, y).

The maximum wavelength λ _MAX is a wavelength band whose central wavelength is the wavelength that gives the maximum value of the spectrum transmitted through the corresponding filter, and the maximum wavelength λ _MIN is the wavelength that gives the minimum value of the spectrum that has passed through the corresponding filter. The wavelength band is the center wavelength. That is, the intensities I1, I2, i1 (λ _MAX ), i0 (λ _MAX ), i2 (λ _MIN ), and i0 (λ _MIN ) mean integrated values of spectrum intensities in each wavelength band.

  As described above, the labeled sand grains have a spectrum different from that of the unlabeled sand grains. Therefore, the sand is imaged in different wavelength bands, and the intensity ratio of each corresponding pixel constituting these images is equal to or greater than a predetermined value. In this case, if this pixel indicates a marked sand particle, the marked sand particle can be accurately identified.

  Next, the detailed structure of the above-described measuring apparatus will be described.

  FIG. 5 is a perspective view of the sand measuring device.

  The light sources LP1 and LP2 constituting the panel strobe are arranged so as to face the imaging region IM. These panel strobes are fixed to arms 101 and 102 extending from the side surface of a support frame (rack) 100. A second belt conveyor as a sand receiving device 4 is disposed below the end point of the belt conveyor 2, and the second belt conveyor has a conveying blade 4 a erected from the surface. The sand that has fallen on the second belt conveyor is transported diagonally upward while being supported by the surface and the side surface of the transport blade 4a, and on the surface of the third belt conveyor 103 at the end point of transport. Drop the sand.

  The third belt conveyor 103 is provided with a conveying blade 103a erected from the surface thereof, and the sand dropped on the third belt conveyor 103 is separated from the surface of the third belt conveyor 103 and the conveying blade 103a. It is conveyed obliquely upward while being supported on the side. A fourth belt conveyor 104 is arranged below the end point of conveyance of the third belt conveyor 103a, and the sand that has fallen from the third belt conveyor 103 falls on the surface of the fourth belt conveyor 104. To do.

  The fourth belt conveyor 104 conveys the sand on it in the horizontal direction, and the sand hopper 1 as sand supply means is disposed below the end point of the sand, and the sand that has fallen from the fourth belt conveyor 104 is disposed. Enters into the opening of the sand hopper 1. A sand agitator SDF is attached to the lower part of the sand hopper 1, and the sand agitator SDF agitates the sand discharged from the sand output port located at the lower part of the sand hopper 1 and conveys it in the horizontal direction. It discharges from the sand discharge port located in the lower part on the surface of belt conveyor 2 so that the amount of discharge may change periodically.

  As described above, the transport means includes the sand transport devices 4, 103, and 104 that return the sand 3 transported by the belt conveyor 2 into the sand hopper 1. If there is a sand transporting device, the sand transporting device can transport the sand 3 once imaged into the sand hopper 1, so that the sand 3 can be imaged many times and the measurement accuracy can be improved.

  In addition, the shaping board 105 etc. for suppressing scattering of sand etc. are provided in the end point vicinity of each belt conveyor, and each belt conveyor is surrounded.

  In the first belt conveyor 2, a first flattening member 110 and a second flattening member 111 are provided between the sand hopper 1 and the imaging region IM. An imaging lens LS is positioned vertically above the imaging area IM. The half mirror HM, the filters FL1 and FL2, the cameras 5 and 6, the timing circuit 7, the lamp driving circuit 8, and the image processing circuit 9 shown in FIG. The built-in imaging device IPD is fixed on the central shelf of the support frame 100. Various belt conveyors are supported by a plurality of support legs 200.

  A computer 11 to which a signal from the image processing circuit 9 shown in FIG. 1 is input is arranged in a control tower (rack) 300. In the control tower 300, in addition to the monitor 10, a control panel 12 having control buttons for a belt conveyor and an imaging device, and an input device 13 such as a keyboard are arranged. On the control panel 12, buttons for controlling the operation, stop and speed of the hopper and the belt conveyor for measurement are arranged. That is, the control panel 12 includes an operation mode changeover switch (automatic operation, test operation), an operation and stop in the automatic operation mode, a sand supply and recovery switch, and a hopper and belt conveyor operation and stop changeover switch in the test operation mode. These speed adjusting knobs can be provided.

  FIG. 6 is a perspective view of the vicinity of the sand agitating device, with the sand agitating device SDF partially broken away.

  A sand output port 1E having a side surface that intersects in the horizontal direction is provided at the bottom of a container having a trapezoidal vertical cross section constituting the sand hopper 1 constituting the sand supply means. A cylinder 1A of a screw conveyor constituting the stirring device SDF is fixed continuously. The cylindrical body 1A extends along the conveying direction TD of the belt conveyor 2, and the screw shaft 1S extends along the central axis X thereof. A screw blade 1B is spirally provided around the screw shaft 1S, and a brush 1F is attached to the tip of the blade 1B.

  Thus, the sand supply means includes a sand hopper (container) 1 having a sand output port 1E at the bottom, and a screw conveyor SDF that extends from the sand output port 1E and discharges sand onto the belt conveyor 2. The brush 1F is attached to the blade 1B of the screw of the screw conveyor SDF.

  When the sand 3 is laminated on the belt conveyor 2, more accurate measurement cannot be performed unless the sand 3 is sufficiently stirred. That is, if the probability that the labeled sand is always buried in a position where it cannot be imaged due to the layer of sand increases, the measurement data may not be accurate. Therefore, in this example, the sand 3 is conveyed by the screw conveyor SDF, whereby the sand particles are agitated so that more accurate data can be obtained. In addition, when the number of times of imaging reaches a plurality, while sand is conveyed many times by the normal screw blade 1B, the surface of the sand itself is shaved on the surface of the blade 1B. That is, the state of the object to be measured, such as the size of sand, changes, and accurate measurement cannot be performed. Therefore, since the brush 1F is attached to the blade 1B and the brush 1F hits the sand, the wear of the sand particles is reduced, and accurate measurement is possible.

  The terminal portion of the cylinder 1A is open, and the sand 3 is intermittently discharged onto the surface of the belt conveyor 2 by the rotation of the blade 1B.

  FIG. 7 is a plan view of the belt conveyor in the vicinity of the flattening members 110 and 111, and FIG. 8 is a cross-sectional view of the apparatus shown in FIG.

  The sand flow F3 is substantially the same as the transport direction TD, but when the sand collides with the first flattening member 110, the sand is flattened, and when the sand collides with the second flattening member 111, the sand is further flattened. The The first flattening member 110 is fixed to the support member 110A, and both ends in the width direction of the support member 110A are fixed to fixing members S1 and S2 provided at both ends in the width direction of the belt conveyor 2, respectively. . The second planarizing member 111 is fixed to the support member 111A, and both ends of the support member 111A in the width direction are fixed to the fixing members S1 and S2, respectively.

  The shape of the first flattening member 110 is a shape like the tip of a ship, and protrudes on the opposite side to the transport direction TD.

  That is, it is preferable to spread sand particles on the belt conveyor 2 from the viewpoint of measurement accuracy, but it is preferable to flatten the sand in order to perform such spreading. In other words, the transport means includes the first flattening member 110, and the first flattening member 110 includes a flat bottom surface 110 </ b> B that is separated from the surface of the belt conveyor 2 by a distance d <b> 1 and the transport direction of the belt conveyor 2. It has the convex part 110P which became narrow toward the opposite side to TD. Further, there is a gap between the first flattening member 110 and the side walls 2W located at both ends of the belt conveyor 2 in the width direction.

  In this case, the sand collides with the convex portion 110P from the leading side of the first flattening member 110 (the tip of the convex portion 110P) with the conveyance, but a part of the sand 3 is the first flattening. While sinking into the space between the bottom surface 110B of the member 110 and the surface of the belt conveyor 2 (see FIG. 9), the remainder is scraped in the width direction by the convex portion 110P, and again, the bottom surface 110B of the first flattening member 110 and the belt conveyor It sunk into the space between the two surfaces and flattened. Furthermore, since excess sand is conveyed through the gap G1 between the side wall 2W of the belt conveyor 2 and the first flattening member 110, the sand does not overflow from the belt conveyor 2.

  The conveying means is arranged on the conveying direction TD side of the belt conveyor 2 of the first flattening member 110 and has a spatula-shaped second flattening member 111 made of an elastic body separated from the surface of the belt conveyor 2 by a distance d2. In addition. d1> d2, and d2 = 2 mm is set in this example. The sand 3 flattened by the first flattening member 110 is further flattened by the second flattening member 111. That is, the shape of the second flattening member 111 is a spatula shape made of an elastic body such as rubber, and the bottom surface 111B of the elastic body elastically sweeps the surface of the sand 3 as it is transported. The sand 3 slightly pushed in by the pressure applied by the flattening member 110 is dispersed in a natural state, and the flatness of the sand 3 is further increased. The shape of the second flattening member 111 is a rectangular plate shape, and one side surface thereof is arranged in parallel to the surface of the belt conveyor 2. Note that a gap G2 is also provided between the second planarizing member 111 and the side wall 2W, and finally the surplus sand flows through the gap G2 toward the imaging region IM.

  Next, the amount of sand introduced and the side wall structure of the belt conveyor 2 will be described.

  As shown in FIG. 6, the belt conveyor 2 as a conveying means has a pair of side walls 2W extending parallel to the sand conveying direction TD, and conveys sand. The sand hopper 1 as the sand supply means allows sand to flow into the belt conveyor 2, but the apparent volume V of sand flowing into the belt conveyor 2 from the sand hopper 1 (see FIG. 6) per unit time, and the average particle diameter of the sand particles D, the amount L of movement of the belt conveyor 2 per unit time, and the distance W between the pair of side walls satisfy the relational expression W × L × D <V. It should be noted that the apparent volume (the bulk volume) is a volume of a container having a size that is completely filled with sand, and is an apparent volume including a gap between the sands.

  When the above-described conditions are satisfied, the sand 3 is supplied in such a manner that one or more layers of sand particles are stacked on the region on the belt conveyor 2 surrounded by the side walls 2W. That is, the sand 3 flowing from the sand hopper 1 (screw conveyor SDF) onto the belt conveyor 2 is laminated on the belt conveyor 2 in a single layer or more. In this case, since the sand particles are spread on the belt conveyor 2, the ratio of the marked sand particles to the whole can be accurately measured.

  Here, integration of the number of marker sand particles (or pixels thereof) included in the composite image will be described.

  Although the number of labeled sand grains contained in a large amount of sand is very small, if the size of the marked sand grains is known, it is possible to analyze the sand diffusion distribution according to the size of the sand grains. Therefore, in the imaging device, images of a plurality of frames are captured, and the number of pixels (or marker sand particles composed of a plurality of pixels) having a value equal to or greater than the threshold value included in the composite image is determined for each size of the sand particles obtained from these images. Integrated and output. Since the measurement data is not data obtained by measuring all the sand grains, it has a variation with respect to the true value. However, as the integration number is increased, the data approaches the true value.

  In addition, the measurement data includes errors caused by misjudgment as marked sand particles, but the count number of error components can be calculated in advance by accumulating sand images that do not contain marked sand particles. By subtracting the error component count from the obtained count, the data can be brought closer to the true value. These calculations can be performed by the computer 11.

  In addition, the computer 11 is continuous when pixels of a predetermined value or more are continuous in a composite image of one frame including a pixel group that is imaged by the cameras 5 and 6 and that calculates the intensity ratio C (x, y). The number of pixels in the area where the pixels are distributed is measured as an index of the size of the marked sand particles, the number of sand particles for each size of the sand particles is counted, and further, an image of a plurality of frames captured by the imaging device is displayed. The number of sand particles for each size of the contained sand particles is integrated and output to the monitor 10.

  Here, I will describe the sand of the Nakatajima dunes. The particle size of Nakatajima sand dunes is said to be most often from 0.3 to 0.5 mm, and when the particle size distribution is examined by screening the sand collected near the shoreline of Nakatajima dunes, The sand particles of ˜0.5 mm account for nearly 90%. For example, when 1 liter of sand having an average particle diameter of 0.04 cm (assumed) is uniformly leveled to a width of 8 cm × 0.2 cm, the longitudinal direction of the conveyor L = 1.000 / 8 / 0.2 = 625 cm It becomes. However, since this value is the length of the state in which the sand is spread without any gaps, the gap between adjacent sand grains is actually generated. Therefore, the longitudinal direction L considering the gap between the sand grains is true specific gravity / mass specific gravity = 2.5 / Multiplying by 1.44 = 1.74 times, the longitudinal direction L ′ = 625 × 1.74 = 1,088 cm in consideration of the gap between the sand grains.

Assuming that the total number of sand grains in 1 liter of sand is a true sphere with a sand grain of 0.04 cm, the volume is 3rd power of v = 4/3 × π (0.02) = 0.000033 cm ^3. The total number of grains of sand N = 1,000 / 0.000033 = about 30 million grains, but here again considering sand specific gravity / true specific gravity = 1.44 / 2.5 = 0.58, The number of sand grains of 1 liter is N1 ′ = 30 million grains × 0.58 = 17.4 million grains, and the number of sand grains of 3 liters N3 ′ = 90 million grains × 0.58 = 52.2 million grains.

  As described above, while the number of sand particles is enormous, the amount of labeled sand particles is small.

  Hereinafter, data on the number of labeled sand grains actually obtained will be described.

  First, in the above-described measuring apparatus, the imaging area is set to 3 cm × 4 cm. Assuming that the grain size of the sand grains is 0.04 cm, about 10,000 grains of sand are present in the surface layer. 4 frames are captured per second. The conveying speed of the belt conveyor 2 is 16 cm / second or more.

The sands of Enshu Pass, including the Nakatajima dunes, are mainly formed from the earth and sand of the Tenryu River. Therefore, the sand of Enshu Pass includes minerals from the Ryoke and Shimanto belts, as well as minerals from the Chichibu and Sanbagawa belts, and contains various colored minerals.
(Measurement of blue sand)

  First, the sand of the Nakatajima dune before coloring was measured under the same conditions as those for measuring the blue sand of FIG. 4 (the sand of the Nakatajima dune colored with a blue pigment) (FIG. 10). FIG. 10 is a graph showing a group of pixels (= labeled sand particles) having a value equal to or larger than the threshold value included in the composite image and the number N of the marked sand particles having such an area S. Although the blue sand was not mixed in the sample, 10000 times of imaging were performed for 40 minutes, and two sand grains erroneously determined to be blue sand were confirmed as background noise. That is, the background noise is 0.01 cpf (count per frame: count / imaging field of view).

  First, the sand of the Nakatajima dune before coloring was measured under the same conditions as those for measuring the red sand of FIG. 4 (the sand of the Nakatajima dune colored with a red pigment) (FIG. 11). FIG. 11 is a graph showing a group of pixels (= labeled sand particles) having a value equal to or greater than the threshold value included in the composite image and the number N of labeled sand particles of such area S. Although red sand was not mixed in the sample, it was confirmed that four sand grains erroneously determined to be red sand were detected as background noise after 10,000 times of imaging for 40 minutes. That is, the background noise is 0.04 cpf.

  Next, the detection efficiency of blue colored sand particles is about 0.14 g (about 500 particles) of blue sand particles (particle diameter of 0.5 mm or less) with respect to 1 liter of natural sand (about 1,440 g). Confirmed by mixing. Strictly speaking, the added blue sand grains must be added to the weight of the natural sand, but it was neglected because it was as small as 0.42 g at the maximum.

  FIG. 12 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to blue sand is 10000: 1 (weight ratio). From this graph, it can be seen that several hundred sand having a size smaller than 50 pixels are distributed. The total number of sand grains measured is 1911, and the number of measurements is 10,000. In this case, a value of 0.19 cpf is obtained.

  FIG. 13 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to blue sand is 10000: 2 (weight ratio). The total number of sand grains measured is 3881, and the number of measurements is 10,000. In this case, a value of 0.39 cpf is obtained.

  FIG. 14 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to blue sand is 10000: 3 (weight ratio). The total number of sand grains measured is 5854, and the number of measurements is 10,000. In this case, a value of 0.59 cpf is obtained.

FIG. 15 is a graph showing the relationship between the blue sand concentration C _BLUE (ppm) and the average count value NAVE, which is the average of the count values N. It can be seen that the average count varies substantially linearly with the blue sand concentration. The concentration (weight ratio) of blue sand correlates with the average count value.
(Measurement of red sand)

  In the detection of red sand, the influence of background noise contained in natural sand cannot be ignored. This is because a relatively large amount of red chart having a diffuse reflection spectrum similar to that of red sand is contained in natural sand. In addition, feldspar that strongly reflects the illumination light in a wide band is also mixed.

  The detection efficiency of red sand was also confirmed by mixing about 0.14 g (about 120 grains) of red sand (particle diameter: 1 mm) with 1 liter of natural sand (about 1440 g). At this time, it was ignored without considering the weight of red sand.

  FIG. 16 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to red sand is 10000: 1 (weight ratio). The total number of sand grains measured is 546, and the number of measurements is 10,000. In this case, a value of 0.05 cpf is obtained.

  FIG. 17 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to red sand is 10000: 2 (weight ratio). The total number of sand grains measured is 788, and the number of measurements is 10,000. In this case, a value of 0.08 cpf is obtained.

  FIG. 18 is a graph showing the relationship between the sand grain area S (number of pixels) and the sand grain count value N. The ratio of natural sand to red sand is 10000: 3 (weight ratio). The total number of sand grains measured is 957, and the number of measurements is 10,000. In this case, a value of 0.10 cpf is obtained.

FIG. 19 is a graph showing the relationship between the red sand concentration C _RED (ppm) and the average count value NAVE, which is the average of the count values N. It can be seen that the average count varies substantially linearly with the red sand concentration. The concentration (weight ratio) of red sand correlates with the average count value.

  The light emission timing of the light source at the time of imaging will be described.

  FIG. 20 is a timing chart of various signals.

  A light source driving signal S1, a CCD activation trigger signal S2, a CCD shutter opening signal S3, and a light source emission timing S4 are shown.

  First, in the cameras 5 and 6 each incorporating a CCD, when the CCD activation trigger signal S2 rises (time t1), imaging in the CCD is started. The trigger falls quickly (time t2). After time t1, the CCD shutter is opened, and subsequently, when the light source drive signal S1 rises (time t3), the light source actually emits light in synchronization with the light source drive signal S1, as shown by S4, and imaging is performed. Is called. Thereafter, the light source drive signal S1 falls, the light source is turned off (time t4), and the CCD shutter is closed (time t5). Imaging is performed as described above. Although the CCD may have a shutter function, the light emission period of the light source limits the imaging period.

  Furthermore, the above-mentioned sand measuring method will be described.

  In addition, the sand measurement method includes a first step of collecting sand at a measurement site, a second step of placing labeled sand at an initial point, and sand at a point away from the initial point after a given period. 3 steps of sampling, a fourth step of imaging while conveying the sand collected in the third step, and a fifth step of counting the sand labeled from the captured image.

  When measuring the behavior of sand for environmental assessment and civil engineering construction surveys, the sand at the measurement site is collected, marked, and placed at the initial reference point. For example, when a predetermined period of time such as one month later has elapsed, sand at a point away from the initial point is sampled and imaged, and the number of labeled sand is counted to determine the state of sand diffusion.

  The imaging in the fourth step is performed by simultaneously measuring the sand images transmitted through the two filters FL1 and FL2 having different transmission wavelength bands using the above-described measuring apparatus, and the respective filters FL1 and FL2 are measured. In the fifth step, when the ratio of the intensity of the pixel in the first image and the intensity of the pixel in the second image is a predetermined value or more, It is determined that the pixel receives light from the marker sand particles.

In this case, since the labeled sand particles have a spectrum different from that of the unlabeled sand particles, when the intensity ratio of the corresponding pixels constituting the first image and the second image is equal to or greater than a predetermined value, It can indicate the marked sand particles and can be accurately discriminated.
Further, in the sand measuring method of the present embodiment, in the case where pixels having a predetermined value or more are consecutive in one frame image composed of a group of pixels that are imaged and the intensity ratio is calculated, the distribution of the continuous pixels The number of pixels in the area to be measured is measured as an index of the size of the marked sand particles, the number of sand particles for each size of the sand particles is counted, and further, the sand particles for each size of the sand particles included in the captured multiple frame image Is accumulated and output.

  If the size of the marked sand particles is known, it is possible to analyze the sand diffusion distribution according to the size of the sand particles. Further, since the measurement data is not data obtained by measuring all sand grains, the measurement data has a variation with respect to the true value, but the data approaches the true value as the integration number is increased.

It is a block diagram of a sand measuring device. It is a graph which shows the reflection spectrum from various sand. It is a graph which shows a relative reflection spectrum. It is a table | surface which shows the wavelength (maximum wavelength) which gives the maximum value of the reflection spectrum of each sand, and the wavelength (minimum wavelength) which gives the minimum value. It is a perspective view of the sand measuring device. It is a perspective view of a sand agitator vicinity part showing a sand agitator SDF partially broken. It is a top view of the belt conveyor of the vicinity of the planarization members 110 and 111. FIG. It is VIII-VIII arrow sectional drawing of the apparatus shown in FIG. It is a figure which shows the belt conveyor which is conveying sand. 5 is a graph showing a group of pixels (= labeled sand particles) having a value equal to or greater than a threshold included in a composite image and the number N of labeled sand particles of area S 5 is a graph showing a group of pixels (= labeled sand particles) having a value equal to or greater than a threshold included in a composite image and the number N of labeled sand particles of area S It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. Blue sand concentration _C BLUE (ppm) and is a graph showing the relationship between the average count NAVE. It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. It is a graph which shows the relationship between the area S (number of pixels) of a sand grain, and the count value N of a sand grain. Is a graph showing the average count NAVE relationship between red sand concentration _C RED (ppm). It is a timing chart of various signals.

Explanation of symbols

  2 ... belt conveyor (conveying means), 5, 6 ... camera, FL1, FL2 ... filter, 11 ... computer (calculating means).

Claims (11)

In a sand measuring device containing labeled sand particles,
Conveying means for conveying sand;
Imaging means for imaging sand being conveyed by the conveying means;
From the sand image picked up by the image pickup means, computing means for discriminating and counting the labeled sand particles,
An apparatus for measuring sand, comprising: The imaging means includes
A first filter having a first transmission wavelength band;
A second filter having a second transmission wavelength band different from the first filter;
A first camera that captures an image of sand in the specific area being conveyed that has passed through the first filter;
A second camera that captures an image of sand in the specific area being conveyed that has passed through the second filter simultaneously with the first camera;
With
The computing means is
If the ratio of the intensity of the pixel in the first image output from the first camera and the intensity of the pixel in the second image output from the second camera is equal to or greater than a predetermined value, the pixel is marked. Discriminate that it is receiving light from the sand particles,
The sand measuring apparatus according to claim 1. The computing means is
In a single frame image composed of a group of pixels that are imaged by the imaging means and for which the intensity ratio has been calculated, if pixels of the predetermined value or more are consecutive, the number of pixels in the region where the continuous pixels are distributed is determined. Measured as an index of the size of the labeled sand particles, counted the number of sand particles for each size of the sand particles, and further sand particles for each size of the sand particles included in the image of a plurality of frames imaged by the imaging means The number of
The sand measuring apparatus according to claim 2. The conveying means is
A belt conveyor having a pair of side walls extending parallel to the sand conveying direction and conveying sand;
Sand supply means for flowing sand onto the belt conveyor;
With
The apparent volume V of sand flowing from the sand supply means on the belt conveyor per unit time, the average particle size D of sand grains, the amount of movement L that the belt conveyor advances per unit time, and the distance W between the pair of side walls 4 satisfies the relational expression W × L × D <V. 4. The sand measuring apparatus according to claim 1, wherein: The conveying means is
A first flattening member having a flat bottom surface separated from the surface of the belt conveyor and a convex portion having a width narrowed toward the opposite side to the conveying direction of the belt conveyor; The sand measuring device according to claim 4, wherein a gap exists between the member and the side wall. The conveying means is
The spatula-shaped 2nd planarization member which is arrange | positioned at the conveyance direction side of the said belt conveyor of the said 1st planarization member, and consists of an elastic body spaced apart from the surface of the said belt conveyor is further provided. 5. The sand measuring apparatus according to 5. The sand supply means
A container having a sand output port at the bottom;
A screw conveyor extending from the sand output port and discharging sand onto the belt conveyor;
With
The brush is attached to the blade of the screw of the screw conveyor,
The sand measuring device according to any one of claims 4 to 6, wherein The conveying means is
The sand measuring device according to any one of claims 4 to 7, further comprising a sand transporting device for returning the sand transported by the belt conveyor into the sand supply means. A first step of collecting sand at the measurement site;
A second step of placing the labeled sand at the initial point;
3 steps to collect sand at a point away from the initial point after a given period of time;
A fourth step of imaging while conveying the sand collected in the third step;
A fifth step of counting the labeled sand from the captured image;
A method for measuring sand, comprising: The imaging in the fourth step is performed by simultaneously measuring sand images respectively transmitted through two filters having different transmission wavelength bands,
Assuming that the images transmitted through the respective filters are the first image and the second image, in the sixth step, the ratio of the intensity of the pixel in the first image and the intensity of the pixel in the second image is a predetermined value. In the above case, it is determined that the pixel receives light from the labeled sand particles.
The method for measuring sand according to claim 9.   In a single frame image composed of a group of pixels that have been imaged and the ratio of the intensity calculated, when pixels of the predetermined value or more are consecutive, the number of pixels in the region where the continuous pixels are distributed is marked. Measured as sand particle size index, counts the number of sand particles for each sand particle size, and further adds and outputs the number of sand particles for each sand particle size included in the captured multiple frame images The method for measuring sand according to claim 10.

Images