JP2000180369A - Method and apparatus for measurement of appearance quality of grain - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a method and an apparatus in which the appearance quality of a grain can be measured precisely without relying on the man power of an operator or his experience, by a method wherein, in addition to a process to acquire information on a plane, a process to acquire information on a side face is provided, and the measuring element of the thickness of the grain is added. SOLUTION: The grain-appearance quality measuring apparatus 1 which evaluates a grain such as a grain or rice or the like is provided with a light receiving sensor 11 and a light receiving sensor 12 such as CCD's or the like which are installed, e.g. at a disk 4 used to transfer the grain while being turned and which are arranged on the upper side and the side face while an observer's view point recessed part as one recessed part used to arrange the grain is used as an observer's view point. Then, light from a light source 13 is first irradiated, transmitted light is received by the light receiving sensor 11. Then, light from a light source 14a and light from a light source 14b are changed over, and reflected light is received by the light receiving sensors 11, 12. Obtained pixel information on a plane and that on a side face are processed by an image processing part 18. On the basis of the transmitted light and the reflected light of the plane, information on the width and that on the length of grain, information on the value of a quantity of transmitted light and information on the color on the surface of the grain are obtained. On the basis of the reflected light of the side face, information on the thickness of the grain is obtained. On the basis of these pieces of information, the appearance quality and the shape quality of the grain are discriminated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION In measuring the quality of a grain, not only the outer color and simple shape of the grain but also shape information such as dimensions, area, and volume that are very similar to the shape of the grain are added. The present invention relates to a method and apparatus for measuring the quality of a grain, which can comprehensively determine the quality and characteristics of the grain from information on the inside and outside of the grain.

[0002]

2. Description of the Related Art Conventional grain quality discriminating methods and apparatuses have discriminated only grain quality based on transmitted light and reflected light obtained from a grain. That is, for example, in the determination of the quality of rice grains, judging of immature sized, dead rice, colored rice, cracked rice, etc. is performed only by reflected light and transmitted light obtained from the rice grains. The sizing, immature grains, mold grains, foreign matter, and the like were discriminated only by the obtained reflected light.

[0003] Japanese Unexamined Patent Publication No. Hei 8-501386 and Japanese Unexamined Patent Publication No. Hei 10-500 employ a method in which the shape of a grain is adopted as a discriminating element.
No. 770. This technology receives the reflected light from the grain plane with a digital image, determines the size, shape, or volume of the grain, and further converts the digital image pixels into RGB.
This is an automatic evaluation method and apparatus that obtains conventional color information as a display and enables fine grain discrimination based on the size, shape or volume of these grains and color information.

[0004]

The information obtained from the plane of the kernel based on the conventional pixel information is exactly the plane shape of the kernel and only the area of the kernel as viewed from the plane. The information on the area of the grain thus obtained is merely an area, and it is extremely impossible to convert it into the volume of the grain with this information, and it is not possible to determine the exact thickness of the grain. Until now, the thickness of a grain has not been added to the discrimination factor in discriminating the grade of a grain, which is considered to be sufficient in conventional grain grade discrimination using only surface information from a flat surface. Because it has been

However, the planar shape of a grain is not always related to the thickness of the grain. In particular, even if the planar shape and color are the same, the thickness may be different. There is a case where an object must be determined so as to be classified into the same classification. This sometimes hindered the comparison and tabulation when comparing not only the thickness but also the volume.

[0006] The above relates to the item of traits that have been determined based on the experience of the inspector who visually inspects the grain, and has been determined only by inspection by the inspector so far. However, discrimination relying solely on the experience of the inspector is such that the classification of what is at the boundary of the distinction is slightly different depending on the inspector, and individual differences and regional differences slightly affect the grain quality judgment. The limits of discrimination depending on experience can be seen around. In addition, this item is not considered to be measurable with conventional testing equipment, and at present, there is no method or apparatus that can mechanically determine this trait.

[0007] From the above, a quality measurement method and a quality measurement method capable of mechanically enabling the discrimination based on the characteristics of the grain, which has been performed based on the experience of the inspector, and improving the accuracy of the conventional discrimination of the quality. It is an object of the present invention to provide a device.

[0008]

According to the present invention, a step of irradiating a transferred grain with light to obtain plane information including reflected light and transmitted light by a plurality of pixels from the plane of the grain,
A step of obtaining side information including light reflected by a plurality of pixels from the side of the grain; a step of obtaining color information of the grain from reflected light in the plane information; and a length and width of the grain from transmitted light in the plane information Grain quality measurement comprising: a step of obtaining information and a transmitted light amount value; a step of obtaining grain thickness information from reflected light in the side information; and a step of discriminating a grain quality and a trait from each of the information. The problem is solved by a method. That is, by obtaining the side information obtained from the side surface of the grain in addition to the plane information obtained from the plane of the kernel, new information useful for the quality determination is added. That is, a combination of the color information of the plane information, the information of the plane shape, and the shape information of the side information can be selected according to the desired quality. The thickness is obtained from the side surface information, and may be the same measured value as the plane information so far. In other words, the volume obtained by the calculation from the X and Y dimensions of the plane obtained from the plane shape of the plane information, which may be referred to as the actually measured value, and the Z dimension of the side surface obtained from the side information, is obtained only from the conventional X and Y dimensions. The reliability of the numerical value is extremely high as compared with the volume that was predicted and obtained. The volume obtained here and the X, Y, and Z dimensions for obtaining this volume are effective in discriminating a trait that has not been provided as a device so far.

The step of judging the quality of the grain includes the length, width, and thickness of the grain, the area or volume of the grain obtained from each of these dimensions, the amount of green light and white light in color information, and the amount of transmitted light. And a grain quality measuring method for determining the grain quality from the combination of That is, the information for determining the quality may be selected in any combination from the plane information including the reflected light information and the transmitted light information and the respective dimensions, thickness, area, and volume. For example, immature or dead rice grains can be more accurately determined from plane information and dimensions, area, and volume.

The step of discriminating the trait of the grain includes the length, width, and thickness of the grain, the volume of the grain obtained from each of these dimensions, the color of the entire grain in the color information, and the transmitted light amount value. This is a grain quality measuring method for determining the character of the grain based on a combination of the occupancy by light amount classified by the threshold value. That is, the information for discriminating the trait may be selected in any combination from the plane information including the reflected light information and the transmitted light information and the respective dimensions, thickness, area, and volume. In particular, to determine the uniformity and particle size, the size, area,
Volume is used.

The reflected light information contains the color information of the grain, and by acquiring not only the light and shade of light but also the color information, for example, rice grains that are immature can be classified into blue immature and white powdery, Similarly, rice that is classified as dead rice can be classified into blue dead rice and white dead rice. In addition, by obtaining not only the plane information but also the side information as described above, the plane size and the thickness dimension of the rice grain are obtained, and it is possible to reliably determine dead rice or immature rice. Furthermore, since the transmitted light information includes the internal information of the kernel, the items that are determined to be white and white, which are items of the trait, are obtained by acquiring the dimensions, area, and volume of the transmitted light information of the plane information and the side information. Reliable discrimination becomes possible.

After the above-described discrimination, a step of selecting grains according to the discriminated quality is provided. Therefore, the results of the discrimination of the grain quality are used to sort the grains, and the discrimination results can be visually checked and the results can be stored. Becomes possible.

Although the terms “quality” and “character” are used separately in the above description, items that can be measured by the conventional quality discriminating apparatus are defined as “quality”. It is. More specifically, the traits are selected only from the raw material grains, and those that are close to the sized ones. This is a discrimination item that has been visually inspected by the inspector, and there has been no apparatus that can measure the traits so far. Further, in the present application, since the thickness information is added to the discrimination information of the conventional discriminating apparatus, the quality measurement performed so far can be measured more accurately.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a grain quality measuring apparatus suitable for the present invention will be described with reference to FIGS. FIG. 1 shows a side sectional view of the grain quality measuring device 1, and FIG. 2 shows a plan view of the disk 4 of the grain quality measuring device 1. The disc 4 is provided with a concave portion 2 in which grains can be arranged around the circumference, and is rotatable in the direction of arrow Q about the rotating shaft 3 to transfer the grains. The disk 4 is mounted on a transparent plate 5 having a larger diameter, and the rotating shaft 3 is rotationally driven by a driving motor 6 (deceleration motor). The disk 4 and the transparent plate 5 are fixed to the base 7 together with the drive motor 6 while being inclined. A detecting unit 8 is arranged at an upper part of the inclined disk 4, and a supply unit 9 is arranged at a lower part of the inclined disk 4. In addition, a partition R is provided around the transparent plate 5 in the rotation direction from below the inclination.
Is provided, and a storage section S that allows storage of grains is formed.

The detecting unit 8 is provided at the upper part of the
The concave portion 2 with respect to the disk 4 with the two viewpoint concave portions 10 as viewpoints.
A light-receiving sensor 11 as a plane light-receiving means, preferably a CCD sensor, and a light-receiving sensor 12 as a side light-receiving means as viewed from the side surface of the viewpoint recess 10, preferably a CCD sensor disposed above the vertical line. , Viewpoint recess 10
The light source 13 is provided as an irradiating unit for irradiating light from below the transparent plate 5, and the light sources 14 a and 14 b are provided as irradiating units for irradiating the viewpoint concave portion 10 with light from above. Further, a light receiving sensor-1 is provided between the light source 14 and the light receiving sensor-11.
A light receiving sensor is provided so that the light receiving device 1 does not directly receive the light from the light source 14.
A shielding plate 15a having a line-of-sight slit connecting the lens 11 and the viewpoint recess 10 and a light receiving sensor 12 are provided between the side viewpoint of the viewpoint recess 10 and the light receiving sensor 12 so that the light receiving sensor 11 does not directly receive the light of the light source 14. Sensor 11 and recess 1
A shielding plate 15b having a line-of-sight slit that connects to the line 0 is provided.

The above components are controlled by the following control unit 16. That is, an arithmetic control unit 17 including a CPU for controlling the entire apparatus is provided. The arithmetic control unit 17 includes an image processing unit 18 as a signal processing unit and a storage unit that stores necessary information including criteria for determining quality and traits. 19 and a signal input / output unit 20 for connecting to an external device. The image processing unit 18 is connected to the optical sensors 11 and 12, and the signal input / output unit 20 is connected to an input unit 21 such as a keyboard and the display unit 2.
2 and the mechanical control unit 23 are connected. The mechanical control unit 23 is connected to the drive motor 6, the light sources 13, 14 and the supply unit 9 and, after measuring the quality and traits, sorts out the grains according to the discrimination. Is connected.

FIG. 3 shows the measuring operation of the apparatus 1 having the above configuration.
This will be described below. The following measurement operation is controlled by the arithmetic control unit 17 and the mechanical control unit 23. First, after inputting the kind and moisture value of the grain to be measured from the input unit 21 (3-1) and then giving a start instruction, the driving motor 6
Rotation (3-2) causes the disk 4 to start rotating. Further, supply of grains from the supply unit 9 to the storage unit S is started (3-3).
Then, the grains input into the storage section S are captured by the concave portions 2 of the rotating disk 4 and are conveyed to the inclined upper portion of the disk 4 which is inclined one by one with the rotation of the disk 4. In the upper part of the slope, the detection unit 8 measures the quality and character of the grain.

In the detecting unit 8, first, the light transmitted from the grain is received by the light receiving sensor 11 (3-5) at the same time as the irradiation of light only by the light source 13 (3-4). Next, the light source is switched from the light source 13 to the light sources 14a and 14b, and at the same time as the light irradiation by the light source 14 alone (3-6), the reflected light from the grain is received by the light receiving sensor 11 (3-7). As a result, plane information of the grain can be obtained. During this time, the light receiving sensor 12 receives the reflected light of the kernel from the side surface of the kernel (3-8). That is, it is possible to obtain the side information of the grain. In order to continuously obtain information from grains, the above (3-4) to (3-8) are repeatedly executed. Upon completion, the light source is turned off, the supply unit 9 is stopped, and the drive motor 6 is stopped (3-9). The information here may be obtained by decomposing one grain into a plurality of pixels, as shown as an example of plane information in FIG. 4 and an example of side information in FIG.
It is preferable to obtain it as information decomposed into 00 pixels or more. For this purpose, an area image sensor or a linear image sensor may be used as the light receiving sensor. The information for each grain is obtained by repeating the above operation.

The plane and side pixel information obtained from one grain is processed by the image processing unit 18 of the control unit 16 as follows. Basically, in the present invention, the plane information of FIG. 4 and the side information of FIG. 5 are obtained, and the plane information further obtains information of transmitted light and information of reflected light. From the transmission information among these pieces of information, it is possible to obtain the internal information of the grain and the information associated with the grain surface properties by image processing. From the reflection information, it is possible to obtain information relating to properties including the color of the grain surface. Then, the X and Y dimensions of FIG. 4 are obtained from the entire plane information. Further, the Z dimension shown in FIG. 5 can be obtained from the side information. The information is processed into information that can be used for quality measurement by the image processing unit 18, and the information is analyzed and determined by the next arithmetic control unit 17.

Specifically, in the image processing section 18, the sensor 1
The image signals, generally RGB signals, obtained from 1, 12 are converted to digital signals. From these signals, information on the luminance, color, and color difference of each pixel is obtained. For example, when the luminance signal is binarized at a predetermined threshold, the contour of the grain can be determined. Further, when the area of the grain is divided by a plurality of threshold values, the degree of the light amount unevenness can be calculated. As for the values processed in this way, for example, when a plurality of threshold values are set to three levels, the occupancy can be calculated for each level. From the calculated values, for example, if the occupancy ratio of each stage is small for each of the three sections, and if the occupancy rate varies among the three sections, it may be in the form of a powdery substance, and if the occupancy rate is confined to some sections different from the normal transmitted light amount. It becomes a judgment such as heart white. For the color based on the amount of reflected light, calculation of the ratio of the green portion to the whole grain, calculation of the ratio of the white portion to the whole grain, and discrimination between blue, white immature, blue, and dead rice, or calculation of the color ratio of the whole grain Thus, the gloss of the grain can be determined by comparison with the reference color ratio. Although a plurality of threshold values are provided in three levels, it goes without saying that more detailed analysis is possible by providing more categories.

To explain again, from the obtained transmitted light quantity, it is possible to detect the light quantity unevenness due to the state inside the grain and the light quantity unevenness due to light refraction related to the surface properties. Since the transmitted light amount is obtained in pixel units, the difference in the light amount in the plane is calculated as a ratio. For example, in the case of rice grains, the transmitted light amount of the sizing is large and the light amount becomes uniform as a whole. In other words, in the powdery substance, unevenness of light and shade of multiple gradations where the amount of light is scattered and dispersed as a whole is detected, the part of the heart white where the amount of transmitted light is reduced is detected, and the part of the light and dark is also a part of the gradation. There is a feature that is biased toward.

Referring to FIG. 6, the received light data of the transmitted light amount is divided by a predetermined threshold value and binarized.
By performing 1), the outline of the grain can be obtained. That is, the grain pixel is specified (6-2). When the kernel pixel is specified, the X dimension (6-3) of the kernel length and the Y dimension (6-4) of the kernel width can be specified as shown in FIG. Finally, the grain area is calculated (6-5). Further, the transmitted light amount in the grain outline is calculated from the light amount of each pixel (6-6). Set the threshold to 3
In step (6-7), pixels having a first threshold value and a predetermined light amount of 1 or more are specified. The second light amount smaller than the predetermined light amount 1
(6-8). The number of pixels with a predetermined light quantity of 1 or more is calculated (6-9), and the number of pixels with a predetermined light quantity of less than 1 and a light quantity of 2 or more is calculated (6-1).
0). Calculate the number of pixels less than the predetermined light amount of 2 (6-11)
I do. Based on the number of pixels for each of the three stages, the occupancy of each of the three stages with respect to the entire grain is calculated (6-12).

As described above, since the range of the amount of light different from the amount of transmitted light of the normal grain can be grasped for each pixel, the ratio can also be calculated. In addition, when there is a difference in brightness between the left and right in the length direction of the grain in the amount of transmitted light, the information can be used as information for determining a crack in a rice grain. In particular, the difference in the amount of transmitted light related to the trait is that the amount of transmitted light becomes uneven due to the thickness or thinness of the skin portion of the grain. This is due to the difference in the total amount of transmitted light and the unevenness caused by the difference in the degree of grain. Further, there are differences in the total amount of transmitted light and unevenness that occur with respect to the white heart described above. In order to discriminate a powdery substance or a heart white based on the occupation rate obtained in the above three stages, for example, using the total amount of transmitted light and the occupation rate obtained from an apparent powdery substance as an explanatory variable, this is discriminated as a powdery particle. It is preferable to create and store a calibration curve for discriminating this as a white heart grain, using the total amount and occupancy of the amount of transmitted light obtained from a clear white heart as an explanatory variable. Needless to say, such a calibration curve can be applied to determination based on other optical information.

The reflection information is information corresponding to the properties including the color of the grain surface, and mainly the information on the color similar to the conventional one. However, the difference from the original sizing color can be determined. That is, the difference in the gloss of the grain is determined by comparing the color of the reflected light with the reference color. Referring to FIG. 7, the contour of the grain can be obtained by binarizing the received light data of the reflected light amount by a predetermined threshold value (7-1). That is, the grain pixel is specified (7-2). When the kernel pixel is specified, as shown in FIG.
The dimension (7-3) and the Y dimension (6-4) of the grain width can be specified. Finally, the grain area is calculated (7-5). The reflected light amount in the grain outline is calculated (7-6) from the light amount of each pixel. Using the RGB signals, the luminance of each R / G / B in pixel units is averaged to obtain an average value for each of the entire R / G / B signals, and the color and luminance of the entire grain are determined by the obtained average RGB signals. Is calculated, and the difference between the color and the brightness of the entire grain is calculated by comparing the color with a predetermined color reference signal (7-7).
I do. Further, the color of each pixel is specified from the RGB signal of each pixel, and the total amount of light of the green signal necessary for determination is calculated (7-8), and the total amount of light of the white signal is calculated (7-9). ).

Further, according to the above transmission information and reflection information,
As shown in FIG. 4, the X and Y dimensions of the grain can be measured, which can be replaced with conventional surface area. Further, crushed granules can be specified depending on the X dimension and the Y dimension. In the transmission information, the ratio between the different color portion and the entire area can be calculated, so that the degree of coloring can be calculated, and information over the details of the sample grains can be obtained. By calculating the ratio of a specific color, for example, the ratio of a specific transmitted light amount portion such as green or white, it can be used for determining dead rice and colored grains.

In the aspect information according to the present invention, the Z shown in FIG.
A major feature is obtaining dimensions. The side information is so-called grain thickness information, and the X dimension, Y dimension,
In addition to the grain size, immature, dead rice, and crushed grains other than the sized grains can be clearly distinguished. In the trait section,
It is now possible to numerically compare the uniformity and particle size. Further, by obtaining the X, Y, and Z dimensions, the volume can be obtained more accurately. The method of obtaining the volume in the prior art is a predicted value based on the X dimension and the Y dimension, and naturally does not incorporate information on the thickness. In particular, uniformity is an important item in the inspection of traits performed by the inspector, but in the present invention, since the X, Y, and Z dimensions are uniform, the volume can be determined more accurately than in the past, so the traits Items, the degree of fulfillment, the uniformity of grains, and the particle size can be determined based on the measured dimensions, area, and volume.

Referring to FIG. 8, the received light data of the reflected light amount is divided by a predetermined threshold value and binarized (8-
By performing 1), the outline of the grain can be obtained. That is, the grain pixel is specified (8-2). When the kernel pixel is specified, the X dimension (8-3) of the kernel length and the Z dimension (8-4) of the kernel thickness can be specified as shown in FIG. Finally, the grain area is calculated (8-5). The reflected light amount in the grain outline is calculated (8-6) from the light amount of each pixel. Next, similarly to the reflected light amount of the plane information, the color of each pixel is specified from the RGB signals of each pixel, and the total of the light amounts of the green signals necessary for the discrimination is calculated from the colors (8-7). Is calculated (8-8).

Based on the information obtained by processing each piece of information in the image processing unit 8 as described above, the judgment table illustrated in Table 1 stored in the storage unit 19, and the predetermined criteria, the arithmetic control unit 17 performs quality control. And the trait. Here, an example of rice grains will be described with reference to FIG.

[Table 1] That is, in the quality determination, if the X dimension is equal to or less than the reference, the grain is determined to be crushed. Among the non-crushed grains, the sized grains can be discriminated on condition that the size (X, Y, Z dimensions, area or volume) and the amount of transmitted light are above the reference (9-2). Among the rice grains that were not discriminated as sized, dead rice can be discriminated based on the fact that the thickness (Z dimension) is smaller than the standard for immature grains or the volume is smaller than the standard for immature grains (9-3). Grains that are neither dead nor dead become immature. In addition, the blue immature has a large amount of reflected green light and the transmitted light amount is lower than the reference (9-4), and the white immature has a reflected light amount higher than the reference, a white light amount is large and the transmitted light amount is lower than the reference. It can be determined from the presence of a low portion (9-5). Subsequently, the dead rice is the one in which blue dead rice has a large amount of reflected green light and the amount of transmitted light is lower than the standard (9-6). The white dead rice has large amount of reflected light, large amount of white light and the lowest amount of transmitted light. (9-7). A body crack is determined to be a body crack if the left and right brightness in the length direction are different in sizing.

Next, rice grains determined to be sized in quality are set as trait discrimination targets. This will be described with reference to FIG. First, based on at least one of the size, area, and volume of the sized particles, it is preferable to determine the uniformity of the particles based on the variation in the volume value (10-1). The area of the powdery material portion and the area of wrinkles (brightness unevenness) on the surface obtained from the transmitted light amount are compared with a reference. That is, the above-mentioned transmitted light amount is divided into three stages, and the occupation ratio is determined for each stage. The degree of fulfillment is determined based on the calculated value, for example, depending on whether the occupation ratio varies in a plurality of sections (10-2). Length (X dimension),
The grain size is determined from the average value based on the width (Y dimension) and the thickness (Z dimension) (10-3), and the gloss is determined from the difference between the surface color of the kernel due to the reflected light and the reference (10-4). ).
The transmitted light amount is divided into three stages, the occupancy is calculated for each stage, and the calculated value, that is, whether or not the occupation ratio is grouped into a part different from the normal transmitted light amount, is determined as heart white ( 10-5). In this way, the trait determination can be changed to an inspection using a device without relying on a visual inspection by an inspector, so the determination is invariable, and a stable trait measurement can be performed similarly to the conventional quality measurement.

The measurement of the quality of wheat is shown below.
When the plane information and the side information are obtained similarly to the rice grain, the storage unit 1
The quality and traits are determined in the arithmetic and control unit 17 based on the determination table exemplified in Table 2 stored in the storage unit 9 and a predetermined reference. The main determination items are shown.

[Table 2] In the case of wheat grains, the sizing in the quality determination is performed on the condition that the size of the reflected light is larger than the reference, the color of the reflected light is entirely uniform, and the amount of transmitted light is higher than the reference. Here, the size may be compared and determined by individual numerical values of X, Y, Z dimensions, area, and volume, and the side information occupies a large weight here. Damaged grains (germinated grains, diseased grains, combed grains, colored grains, insect damage grains, crushed grains, heat-damaged grains) are the colors of reflected light for wheat grains that were not discriminated as sized, that is, wheat grains less than the standard. Judge from information and size. The ergot grains can also be determined from the color information and the size of the reflected light. Since red mold grains and black mold grains have extreme colors, they can be determined from the color information of the reflected light. Heterogeneous grains (foreign matter) are different from wheat grains in both color and size, and thus grains having extremely different sizes and different grain colors are determined here. The glass ratio can be determined from the transmitted light amount unevenness and the size of the area.

Next, the determination of the trait will be described. Wheat grains determined to be sized in quality are targeted for trait determination.
Since the sizing trait has a large weight on the degree of grain size, the information on the thickness obtained as the side information is very useful. First, the uniformity of the granules is measured based on at least one of the size, area, and volume of the sized granules from the variation in the values. The degree of fulfillment is measured by comparing the area of the vitreous part and the area of wrinkles (brightness unevenness) on the surface obtained from the amount of transmitted light with a reference. The particle size is determined from the average value of the length (X dimension), width (Y dimension), and thickness (Z dimension), and the gloss is measured by comparing the surface color of the reflected light with a reference. The whiteness can be measured from the unevenness of the transmitted light amount and the area ratio of the unevenness. In this way, even in wheat grain, the trait discrimination can be changed to an inspection using a device without relying on visual inspection by an inspector, so the discrimination is invariable and stable trait measurement can be performed in the same way as conventional quality measurement Becomes

As described above, when the grains are measured by the image processing of the image processing section 18 and the arithmetic processing of the arithmetic control section 17 and are determined based on the standard, the sorting section 24 is operated based on the determination. The sorting unit 24 includes a plurality of sorting nozzles 25a to 25
d, each sorting nozzle corresponds to the quality item,
In addition, it is good to correspond to a trait item. In this embodiment, there are four sorting nozzles, but the number may be set appropriately according to the required number of items. When the target grain reaches the position of the corresponding sorting nozzle 25a due to the rotation of the disk 4, the sorting nozzle a is activated and the target grain is repelled by the jet wind. Finally, the display unit 22 may display the above-described processing progress or processing result, a totaling result of a plurality of data, and the like.

[0033]

As described above, by adding a measuring element having a thickness which is not included in the conventional quality measuring apparatus, not only the discrimination by measurement can be performed more accurately, but also the inspection by the inspector can be performed. It was possible to mechanically determine the trait that had been performed. Therefore, a measuring method and a measuring apparatus capable of improving measurement accuracy and measuring and distinguishing quality traits based on an invariant criterion can be provided irrespective of manpower or experience.

[Brief description of the drawings]

FIG. 1 is a side sectional view of a grain quality measuring device according to the present invention.

FIG. 2 is a plan view of a disk of the grain quality measuring device of FIG.

FIG. 3 is an operation flowchart of the grain quality measuring device.

FIG. 4 is plane information of a grain obtained by the grain quality measuring device.

FIG. 5 is side information of a grain obtained by the grain quality measuring device.

FIG. 6 is a flowchart of signal processing of transmitted light.

FIG. 7 is a flowchart of signal processing of reflected light.

FIG. 8 is a flowchart of signal processing of reflected light from the side surface.

FIG. 9 is a flowchart of quality determination.

FIG. 10 is a flowchart of trait discrimination.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Grain quality measuring device 2 Concave part 3 Rotation axis 4 Disk 5 Transparent plate 6 Drive motor 7 Base 8 Detecting part 9 Supply part 9 10 Perspective concave part 11 Light receiving sensor 12 Light receiving sensor 13 Light source 14 Light source 15 Light shielding plate 16 Control unit 17 Operation control unit 18 Image processing unit 19 Storage unit 20 Signal input / output unit 21 Input unit 22 Display unit 23 Mechanical control unit 24 Sorting unit 25 Sorting nozzle

Claims (9)

[Claims] 1. A step of irradiating a grain to be transferred with light to obtain plane information including reflected light and transmitted light by a plurality of pixels from a plane of the grain, and reflection by a plurality of pixels from a side of the grain. A step of obtaining side information including light; a step of obtaining color information of the kernel from reflected light in the plane information; and a step of obtaining information on the length and width of the kernel and the transmitted light amount value from the transmitted light in the plane information And a step of obtaining information on the thickness of the grain from the reflected light in the side information; and a step of determining the grade and the trait of the grain from each piece of the information. 2. The step of judging the quality of a grain includes the length, width and thickness of the grain, the area or volume of the grain obtained from each of these dimensions, the amount of green and white light in color information, The grain quality measuring method according to claim 1, wherein the grade of the grain is determined from a combination of the light quantity value and the light quantity value. 3. The step of discriminating the trait of the grain includes the length, width and thickness of the grain, the volume of the grain obtained from each of these dimensions, the color of the entire grain of the color information, and the transmitted light amount value. 2. The grain quality measuring method according to claim 1, wherein the trait of the grain is determined from a combination of the occupancy by light amount classified by a plurality of threshold values. 4. The method for measuring grain quality according to claim 1, further comprising the step of selecting grains according to the determined quality. 5. A transferring means for transferring a grain, an irradiating means for irradiating the transferred grain with light, and a plane light receiving means comprising a plurality of pixels for obtaining plane information including reflected light and transmitted light of the grain. A side light receiving unit composed of a plurality of pixels for obtaining side information of a grain; a signal processing unit connected to each of the light receiving units and processing signals of plane information and side information obtained by each light receiving unit; Means are connected, and arithmetic control means for discriminating the grade and trait of the grain based on information relating to the grade and trait of the grain processed by the signal processing means and a predetermined discrimination criterion is provided. Grain quality measuring device. 6. The signal processing means calculates the plane size and the side size of the grain from the plane information and the side information by image processing to obtain the thickness, area and volume of the grain. Item 6. The grain quality measuring device according to Item 5. 7. The grain quality measuring apparatus according to claim 5, wherein the signal processing means obtains color information, green information and white information of the whole grain from the reflected light of the plane information by image processing. . 8. The grain quality according to claim 5, wherein the signal processing means obtains the transmitted light amount value of the plane information and the occupancy for each light amount obtained by dividing the transmitted light by a plurality of threshold values by image processing. measuring device. 9. The grain quality measuring device according to claim 5, wherein a sorting means for sorting the grains according to the grade by the discriminating means is connected to the arithmetic control means.