Method for calculating the weight ratio of grains that are in good condition using a grain quality discriminator

Abstract

To provide a method capable of highly accurately measuring a grain sizing weight ratio (sizing ratio) by a grain grade discrimination device.SOLUTION: Disclosed is a method for calculating a grain sizing weight ratio using a grain grade discrimination device, which includes: a volume weight measurement step of measuring the volume weight of the grain to be inspected; an imaging step of imaging the grain by the grain grade discrimination device and acquiring the image information for each piece of the grain; and an arithmetic step of determining the grain sizing weight ratio of the grain based on a value of the volume weight measured in the volume weight measurement step and the image information acquired in the imaging step.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a grain quality discrimination device that captures images of grains to be inspected and acquires images of each grain, and more specifically, to a calculation method that improves the accuracy of the grain size weight ratio determined by the grain quality discrimination device. [Background technology]

[0002] Conventionally, the sample rice to be inspected is optically photographed, and the appearance quality of each grain of the sample rice is mechanically judged (hereinafter referred to as "mechanical judgment") based on the image information obtained from the photograph. This mechanical judgment can judge the appearance quality of dead grains, discolored grains, cracked grains, broken grains, etc. in addition to regular grains.

[0003] The mechanical judgment is carried out by a grain quality judgment device as described in Patent Document 1. The judgment results of the appearance quality obtained by the grain quality judgment device are based on the Agricultural Products Inspection Act (Ministry of Agriculture, Forestry and Fisheries Notification No. 333 of 2001) (Internet<URL:https: / / www.maff.go.jp / j / seisan / syoryu / kensa / kokuryu / attach / pdf / index-3.pdf> ) can be used to assess the percentage of "dead grains" and "colored grains" in domestically produced non-glutinous rice. In addition, the percentage of "cracked grains" and "broken grains" can be assessed by the "Measurement of the percentage of cracked grains and broken grains using a grain classifier (Director of the Grain Division, Director-General for Policy Coordination, Ministry of Agriculture, Forestry and Fisheries)" (Internet<URL:https: / / www.maff.go.jp / j / seisan / syoryu / kensa / kokuryu / attach / pdf / index-5.pdf> ) notification, it can be used when assessing the "total amount of damaged grains, etc."

[0004] Furthermore, since the grain quality discrimination device can discriminate the appearance quality of each grain, it can count the number of grains of each quality contained in a sample (sample rice) of a predetermined number of grains (e.g., 1,000 grains). For example, as disclosed in Patent Document 2, a weight conversion coefficient is set for each grade of grain, and the weight ratio of sized grains (sized grain rate) is calculated using this weight conversion coefficient and the number of grains of each grade obtained by the above-mentioned discrimination by a grain appearance discrimination device. This calculation method has the advantage that the sized grain weight ratio can be easily determined without actually measuring the weight of the grains.

[0005] The weight ratio of whole grains is the ratio of the weight of whole grains contained in the whole grains to be inspected, and is therefore an index showing the quality of the grains. For this reason, it is desirable to be able to use the whole grain weight ratio calculated by a grain quality discrimination device in appraisal methods based on the Agricultural Products Inspection Act. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 6687826 [Patent Document 2] Patent No. 5533245 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the grain quality discrimination device described in Patent Document 2 determines the size-reduced weight ratio based on image information of captured grains, and does not actually measure the weight of the grains. As a result, there is a risk of a discrepancy (deviation) occurring between the measurement result of the size-reduced weight ratio calculated by the grain quality discrimination device and the size-reduced weight ratio (true value) determined by actually measuring the weight of each grain.

[0008] Therefore, in consideration of the above problems, the present invention has as its technical object to provide a method that can calculate the sized grain weight ratio with high accuracy using a grain quality discrimination device so that there is no extreme difference between the calculated value and the true value. [Means for solving the problem]

[0009] In order to solve the above problem, the present invention provides a method for calculating a sized grain weight ratio using a grain quality discriminator, The method comprises a volume weight measuring step of measuring the volume weight of the grains to be inspected, an imaging step of imaging the grains using the grain quality discrimination device and acquiring image information for each of the grains, and a calculation step of determining the size-reduced weight ratio of the grains based on the volume weight value measured in the volume weight measuring step and the image information acquired in the imaging step. 、 In the calculation step, when determining the size-reduced weight ratio of the grains, the value of the volume weight measured in the volume weight measurement step and the value of measurement quality information related to the quality determined from the image information acquired in the imaging step are substituted into a calibration curve, The calibration curve was created using the volume weight value and the measured quality information value as explanatory variables and the granulation weight ratio as a response variable. It is characterized by:

[0010] Furthermore, the image information The measured quality information value for quality obtained from , the average length, average width, average thickness, average surface area, average lateral area, average volume and average length / width ratio of the kernel and these average values The method is characterized in that one or more average values ​​are selected from the above.

[0011] Furthermore, the image information The measured quality information value for quality obtained from , the value of the standard deviation of the length of the kernel, the value of the standard deviation of the width, the value of the standard deviation of the thickness of value, standard deviation value of surface area, standard deviation value of lateral area, standard deviation value of volume and standard deviation value of length / width ratio and the values ​​of these standard deviations One or more standard deviation values ​​are selected from the above. [Effects of the Invention]

[0012] According to the present invention, the sized grain weight ratio is calculated using grain information obtained by adding the actual measured value of the volume weight of the grain to image information of the grain to be inspected taken by a grain quality discrimination device, thereby making it possible to calculate the sized grain weight ratio with high accuracy. Furthermore, since this whole grain weight ratio is also measured in visual inspection based on the Agricultural Products Inspection Act, it is thought that this will expand the possibilities for utilizing the grain quality discrimination device as an alternative to visual inspection.

[0013] Furthermore, the image information used includes the average length, average width, average thickness, average surface area, average side area, average volume, and average length / width ratio of 100 or more grains, so that the characteristics of each grain being inspected can be appropriately processed, improving the accuracy of calculating the whole grain weight ratio.

[0014] Furthermore, the image information used includes the standard deviation of the length of the grains, the standard deviation of the width, the standard deviation of the thickness, the standard deviation of the surface area, the standard deviation of the lateral area, the standard deviation of the volume, and the standard deviation of the length / width ratio. Since the sample (sample rice) used for appraisal contains multiple grains, using the standard deviation values ​​makes it possible to appropriately process the characteristics of variation in the length, width, thickness, surface area, lateral area, volume, length / width ratio, etc. of the multiple grains contained in one sample, thereby improving the accuracy of calculating the grain size weight ratio. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a control block diagram of a grain quality discriminating device according to an embodiment of the present invention. [Figure 2] 1 is a vertical cross-sectional view of a grain quality discriminating device according to the present invention; [Figure 3] 1 is a plan view of a grain quality discriminating device according to an embodiment of the present invention. [Figure 4] FIG. 2 is a vertical cross-sectional side view of a first optical detection unit according to the embodiment of the present invention. [Figure 5] FIG. 3 is a vertical cross-sectional side view of a second optical detection unit according to the embodiment of the present invention. [Figure 6] 1 shows the procedure of a method for calculating a weight ratio of sized grains using a grain quality discriminating device according to an embodiment of the present invention. [Figure 7] This shows an image of a grain of rice. [Figure 8] FIG. 1 is a diagram showing the measurement results of the particle size distribution weight ratio obtained by the method of the present invention. [Figure 9] FIG. 1 is a diagram showing the measurement results of the particle size distribution weight ratio obtained by the method of the present invention. [Figure 10] FIG. 1 is a diagram showing the measurement results of the particle size distribution weight ratio obtained by the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] An embodiment of the present invention will be described below. As shown in Figure 1, the apparatus for carrying out this method comprises a Blawell grain meter 30 capable of measuring the volume weight of sample rice (specimen), and a grain quality discrimination device 1 equipped with imaging means for imaging the sample rice and obtaining image information for each grain, and also equipped with calculation means for calculating the whole grain weight ratio of the sample rice based on the volume weight and the image information.

[0017] The Blawell grain meter 30 is described in the standard measurement method based on the Agricultural Products Inspection Act (Ministry of Agriculture, Forestry and Fisheries Notification No. 332 of 2001). The volume weight is the weight of brown rice per specified volume (unit: g / L).

[0018] The grain quality discrimination device 1 is composed of an input unit 26 into which the volume weight value measured by the Brawell grain meter 30 is input, two optical detection units (2a, 2b), a display unit 25, and a control unit 20 that controls these.

[0019] The grain quality discrimination device 1 of the present invention can be a known device such as that described in Japanese Patent No. 6805506. Fig. 2 is a vertical cross-sectional view of the grain quality discrimination device 1 of the present invention. This grain quality discrimination device 1 is provided with a conveying unit 3 that conveys grains to be measured (hereinafter referred to as "rice grains S") to an imaging unit 2. The conveying unit 3 is of a disk conveying type and has a disk 4 with a plurality of grooves 4a around its periphery, into which the rice grains S fit one by one.

[0020] The disk 4 is disposed on a base plate 6 supported at an incline by a stand 6b, and the output shaft 5a of a motor 5 fixed to the stand 6b is attached to the center of the disk 4 so that it can rotate freely (see Figures 2 and 3). In this embodiment, the disk 4 rotates clockwise (arrow Y). Each groove 4a has a bottom 4b made of a transparent material and an open edge 4c. A weir 6a is formed on the base plate 6 along the periphery of the disk 4 to prevent rice grains S in the groove 4a from being released through the open edge 4c.

[0021] In addition, a rice grain supply section 7 is provided at a position inclined downward of the disk 4. This rice grain supply section 7 has a wide flat surface connected to the weir section 6a so that the rice grains S can be retained above the peripheral edge of the disk 4. If this rice grain supply section 7 is the starting end of the conveyance of the disc 4, then the end of the conveyance has a notch 6c formed by cutting out the dam section 6a, and is configured so that the rice grains S that have been measured fall through the groove 4a.

[0022] The imaging unit 2 is composed of a first optical detection unit 2a and a second optical detection unit 2b, and is disposed at an inclined upper position of the disk 4 (see Figs. 2 and 3). The first optical detection unit 2a is configured to be able to optically detect (image) the top and side surfaces of the conveyed rice grains S (see Fig. 4). The first optical detection unit 2a comprises a condenser lens 8, a CCD linear sensor (light receiving sensor) 9, and red (R), green (G), and blue (B) light emitting diodes as an irradiation unit 10 above the groove 4a (rice grain S), and a condenser lens 11 and a CCD linear sensor (light receiving sensor) 12 on the side of the groove 4a.

[0023] Further, below the groove 4a, an irradiating section (light emitting diode) 13 for irradiating the rice grains S from below is formed in the internal space formed by making the base plate 6 concave. In the first optical detection unit 2a, the weir portion 6a on the side of the groove 4a is made of a transparent material so that light from the rice grains S can enter the CCD linear sensor 12. The CCD linear sensors 9 and 12 are capable of receiving red (R), green (G), and blue (B) light, and are arranged to scan in a direction perpendicular to the conveying direction of the rice grains S (the arrow Y).

[0024] Next, the second optical detection unit 2b is configured to optically detect (image) the underside (bottom) of the rice grain S (see Figure 5). The second optical detection unit 2b is configured with a condenser lens 14, a CCD linear sensor (light-receiving sensor) 15, and red (R), green (G), and blue (B) light-emitting diodes as an irradiation unit 16 in the internal space of the concave base plate 6, while an irradiation unit (light-emitting diode) 17 that irradiates the rice grain S from above is configured above the groove 4a. The CCD linear sensor 15 is also capable of receiving red (R), green (G), and blue (B) light, and is positioned so that it scans in a direction perpendicular to the conveying direction of the rice grain S (the arrow Y).

[0025] Next, an example of the control unit (signal processing unit) 20 that receives and processes light reception signals from the CCD linear sensors 9, 12, and 15 will be described (see FIG. 1). The control unit 20 has a central processing unit (hereinafter referred to as "CPU") 21 as its central component, and is composed of an input / output circuit (hereinafter referred to as "I / O") 22 electrically connected to the CPU 21, a read-only memory unit (hereinafter referred to as "ROM") 23, and a read / write memory unit (hereinafter referred to as "RAM") 24.

[0026] The I / O 22 is electrically connected to the CCD linear sensors 9, 12, and 15, and is also connected to a display unit 25 that displays the quality, grain image, grain size weight ratio, etc. determined by the CPU 21, and a setting start button 22a (not shown). The ROM 23 stores an operating program and a look-up table for controlling the grain quality discriminating device 1. The control unit 20 is also electrically connected to a circuit (not shown) that controls the start of the motor 5 and the lighting output of the irradiating units 10, 13, 16, and 17 (detailed description of known means is omitted).

[0027] The I / O 22 is also connected to an input unit 26. The input unit 26 receives the value of the test weight measured by the Brawell grain meter 30. The volume weight measuring device for measuring the volume weight is not limited to the Brawell grain meter, and a hectoliter kilogram meter or an electric grain meter may also be used.

[0028] A method for imaging rice grains using the grain quality discriminator 1 will now be described. In this embodiment, the case where the rice grains to be inspected are brown rice will be described. Fig. 6 shows the steps of the method for discriminating grain quality and the method for calculating the weight ratio of whole grains using the granular material appearance quality discriminator 1.

[0029] (1) Volumetric weight measurement step (step S1): First, the bulk density of the grains to be inspected (sample rice) is measured using a Brawell grain meter 30. The measurement is preferably carried out using the method described in the Standard Measurement Method based on the Agricultural Products Inspection Act (Ministry of Agriculture, Forestry and Fisheries Notification No. 332 of 2001). The volume weight measuring step (step S1) may be performed after the measurement by the grain quality discriminating device 1.

[0030] (2) Imaging process (step S2): Next, the sample rice is measured using the grain quality discrimination device 1. When the sample rice grains S are supplied to the rice grain supply unit 7 and a measurement start button (not shown) is turned on, the disk 4 rotates, and the rice grains S from the rice grain supply unit 7 enter the groove 4a one by one and are transported to the first and second optical detection units 2a and 2b. In the first optical detection unit 2a, light is irradiated onto the rice grains S from the irradiation units 10 and 13, and the rice grains S are scanned by the CCD linear sensors 9 and 12. This obtains image information of the top and side surfaces of the rice grains S, and each image information is stored sequentially in the RAM 24.

[0031] The rice grain S that has passed through the first optical detection unit 2a is imaged by the second optical detection unit 2b. That is, in the second optical detection unit 2b, light is irradiated onto the rice grain S from the irradiation units 16 and 17, and the rice grain S is scanned by the CCD linear sensor 15. As a result, image information of the back side of the rice grain S is obtained, and the image information is sequentially stored in the RAM 24.

[0032] (3) Image processing step (step S3): The CPU 21 sequentially reads out the image information of each rice grain S stored in the RAM 24 and processes it in the image processing unit 47 to obtain image information for each grain, and extracts measurement quality information related to the quality of each rice grain, such as its outer shape, area, length, width, color, and cracks.

[0033] (4) Discrimination step (step S4): Next, the discrimination unit 48 of the CPU 21 compares the measured quality information with preset quality information (e.g., threshold values ​​or calibration curves (quality relational expressions)) to determine the quality of each rice grain (whole grain, broken grain, dead grain, discolored grain, green immature grain, rice damaged by pests, etc.). The extraction and discrimination can be performed using a general analytical method. For example, methods such as those described in JP 2011-242284 A and JP 2000-304702 A can be used.

[0034] (5) Calculation step (step S5) At this stage, the volume weight value measured in the volume weight measurement step (step S1) is input to the input unit 26 of the grain quality discrimination device 1. The input volume weight value is then sent via I / O 22 to CPU 21, which calculates the sieved weight ratio. This calculation is performed by substituting the volume weight value measured with the Blawell grain meter and the values ​​of the measured quality information into a calibration curve (calculation formula) that has been created in advance to calculate the sieved weight ratio. The calibration curve is created by the method described below. The values ​​to be substituted may be those described below.

[0035] (6) Image creation process (step S6): The image information is used to create an image of each grain of rice in the image processing section of the CPU 21. The image creating step may be performed in the image processing step (step S3) described above.

[0036] (7) Display step (step S7): The image of each rice grain created in the image creation step is displayed on a display unit 25 such as a display. The method of display is not particularly limited.

[0037] In the image processing step (step S3), values ​​such as length, width, thickness, surface area, lateral area, volume, length / thickness ratio, etc. of the input brown rice (sample rice) are determined for each grain, and the standard deviation of length, width, thickness, surface area, lateral area, volume, and length / thickness ratio are calculated for each sample rice. These calculated values ​​and standard deviations may be displayed in the display step (step S7). In order to obtain the above value from the image information of the kernel, a known method such as that disclosed in Japanese Patent Application Laid-Open No. 2003-042963 can be used.

[0038] In the above calculation, the length is the length in the X direction of Fig. 7, which is the longitudinal direction of the kernel, the width is the Y direction, and the thickness is the Z direction. Furthermore, the surface area is the area of ​​the kernel shown in Fig. 7(a), and the side area is the area of ​​the kernel shown in Fig. 7(b). The weight ratio of the sample rice grains that have been sized is calculated from the value, the standard deviation, and a volume weight value that is separately measured, using a method described below. The weight ratio of the sized rice grains may then be displayed in the display step (step S7). The weight ratio of whole grains is the percentage of whole grains in a given amount of brown rice, expressed as a weight ratio. For example, if a 100g sample of rice contains 70g of whole grains, the weight ratio of whole grains in that sample is 70%.

[0039] In the embodiment of the present invention, the calculation process (step S5) is performed using the CPU 21 of the grain quality discrimination device 1, but the calculation process (step S5) can also be performed using a PC (personal computer), tablet terminal, etc. that can input image information captured by the grain quality discrimination device 1 and the volume weight value measured by a Brawell grain meter.

[0040] In the present invention, in order to increase the accuracy of the sieved weight ratio calculated from image information (image capture data) of grains, a calibration curve (regression equation) is created using the actual measured volume weight value and items measurable by the grain quality discrimination device 1 (length, width, thickness, etc.) as explanatory variables and the sieved weight ratio as the objective variable. The calibration curve is then used to calculate the sieved weight ratio. A known analytical method may be used to create the calibration curve. The calibration curve may be created by multivariate analysis such as multiple regression analysis or PLS regression analysis. The explanatory variables use the values ​​of the measured quality information.

[0041] The method for creating the calibration curve will be described. Multiple sample rice (brown rice) was prepared, and the volume weight of each sample rice was measured. This measurement was performed using a well-known Blawell Grain Meter 30 (manufactured by Fujiwara Seisakusho Co., Ltd.). A total of 34 varieties of domestically produced brown rice (short grain) from the 2018 or 2019 harvest were used as sample rice. Each sample of rice was measured three times, resulting in 102 measurement data points.

[0042] The measurement method using the Brawell Grain Meter 30 was carried out in accordance with the standard measurement method based on the Agricultural Products Inspection Law (Ministry of Agriculture, Forestry and Fisheries Notification No. 332 of 2001).

[0043] Next, using the 102 pieces of measurement data, a calibration curve was created by multiple regression analysis, with the sized grain weight ratio as the objective variable and the bulk density and the values ​​of each item determined from the image information of the grain quality discrimination device 1 as explanatory variables. In this case, the 102 pieces of measurement data were divided into two, with 51 pieces of measurement data being used for calibration and the remaining 51 pieces of measurement data being used for validation (prediction).

[0044] In the embodiment of the present invention, a PC (personal computer) was separately prepared into which image information captured by the grain quality discrimination device 1 and the volume weight values ​​measured by the Brawell grain meter could be input, and the calibration curve was created, calibrated, and verified using the PC.

[0045] The target variable, the sized grain weight ratio, is preferably determined by measuring the weight of each grain, but in this example, the weight ratio was determined by measuring the weight using a grain discriminator (manufactured by Satake Corporation, model: RGQI100B).

[0046] Next, the results of creating calibration curves when the number of explanatory variables was 8 (Example 1), 15 (Example 2), and 4 (Example 3) are shown below.

[0047] Example 1 (8 explanatory variables) Objective variable: granulation weight ratio Explanatory variables: test weight, average length, average width, average thickness, average surface area, average lateral area, average volume, average length / thickness ratio

[0048] The measurement data (volume weight and image information) for the 102 rice samples were divided into two, and 51 of the measurement data were used for calibration. Multiple regression analysis was performed to determine coefficients and constants to create calibration curve A (regression equation). The remaining half of the measurement data (51) was then substituted into calibration curve A for verification (prediction). The results are shown in Figure 8.

[0049] 8 shows the accuracy of the sized grain weight ratio calculated from the image information of rice grains and the volume weight value of the grain quality discriminating device 1 using the calibration curve A. As a result of calculation under the conditions of Example 1 (8 explanatory variables), the correlation coefficient (R) was 0.528.

[0050] Example 2 (15 explanatory variables) Objective variable: granulation weight ratio Explanatory variables: test weight, mean length, mean width, mean thickness, mean surface area, mean lateral area, mean volume, mean length / thickness ratio, standard deviation of length, standard deviation of width, standard deviation of thickness, standard deviation of surface area, standard deviation of lateral area, standard deviation of volume, standard deviation of length / thickness ratio

[0051] In Example 2, multiple standard deviation values ​​were added to the explanatory variables of Example 1. The measurement data (volume weight and image information) for the 102 sample rice samples were divided into two, and 51 of the measurement data were used for calibration. Multiple regression analysis was performed to determine coefficients and constants to create calibration curve B (regression equation). The remaining half of the measurement data, 51, was then substituted into calibration curve B for verification (prediction). The results are shown in Figure 9.

[0052] Fig. 9 shows the accuracy of the refined grain weight ratio calculated from the image information of rice grains and the volume weight value of the grain quality discrimination device 1 using the calibration curve B. As a result of calculation under the conditions of Example 2 (15 explanatory variables), the correlation coefficient (R) was 0.738. It is possible to improve the accuracy by adding a standard deviation value to the explanatory variables.

[0053] Example 3 (four explanatory variables) Objective variable: granulation weight ratio Explanatory variables: test weight, average length, average width, average thickness

[0054] The measurement data (volume weight and image information) for the 102 rice samples were divided into two, and 51 of the measurement data were used for calibration. Multiple regression analysis was performed to determine coefficients and constants to create calibration curve C (regression equation). The remaining half of the measurement data (51 data) was then substituted into calibration curve C for verification (prediction). The results are shown in Figure 10.

[0055] Fig. 10 shows the accuracy of the whole grain weight ratio calculated from the image information of rice grains and the volume weight value of the grain quality discrimination device 1 using the calibration curve C. As a result of calculation under the conditions of Example 3 (four explanatory variables), the correlation coefficient (R) was 0.460. It seems difficult to accurately calculate the whole grain weight ratio under these conditions.

[0056] Although the embodiments of the present invention have been described above, it goes without saying that the present invention is not limited to the above-described embodiments, and the configuration can be appropriately modified without departing from the scope of the invention. [Industrial Applicability]

[0057] The method for calculating the whole grain weight ratio using the grain quality discrimination device of the present invention makes it possible to accurately determine the whole grain weight ratio of a sample (sample rice). This is extremely useful because it allows the whole grain weight ratio of the sample rice to be calculated with high accuracy without actually measuring the weight of each grain of the sample rice, and the sample rice can be quantitatively evaluated using the calculated whole grain weight ratio. [Explanation of symbols]

[0058] 1 Grain quality determination device 2. Imaging unit 2a First optical detection unit 2b Second optical detection unit 3. Conveyor 4 disks 4a groove 4b bottom 4c Open area 5 motors 5a Output shaft 6 Base plate 6a Weir 6b Mounting stand 7 Rice grain supply section 8 Condenser Lens 9 CCD linear sensor (light receiving sensor) 10 Irradiation unit 11 Condenser lens 12 CCD linear sensor (light receiving sensor) 13 Irradiation unit 14 Condenser lens 15 CCD linear sensor (light receiving sensor) 16 Irradiation unit 17 Irradiation unit 20 Control Unit 21 Central Processing Unit (CPU) 21 22 Input / Output Circuit (I / O) 23 Read-only storage 24 Read / write memory unit 25 Display section 26 Input section 30 Brawell Grain Meter

Claims

1. A method for calculating a sieved weight ratio using a grain quality discriminator, A volume weight measurement step of measuring the volume weight of the grain to be inspected; an imaging step of imaging the grains using the grain quality discrimination device and acquiring image information of each grain; A calculation step of calculating the sized grain weight ratio of the grains based on the volume weight value measured in the volume weight measurement step and the image information acquired in the imaging step; Equipped with In the calculation step, when determining the size-reduced weight ratio of the grains, the value of the volume weight measured in the volume weight measurement step and the value of measurement quality information related to the quality determined from the image information acquired in the imaging step are substituted into a calibration curve, A method for calculating a sieved weight ratio using a grain quality discrimination device, characterized in that the calibration curve is created using the volume weight value and the measured quality information value as explanatory variables and the sieved weight ratio as a target variable.

2. The values ​​of the measured quality information relating to the quality obtained from the image information are the average length, average width, average thickness, average surface area, average side area, average volume and average length / width ratio of the kernels, 2. A method for calculating a sized grain weight ratio using the grain quality discriminating device according to claim 1, wherein one or more of these average values ​​are selected and used.

3. The values ​​of the measurement quality information relating to the quality obtained from the image information are a standard deviation value of the length of the kernel, a standard deviation value of the width, a standard deviation value of the thickness, a standard deviation value of the surface area, a standard deviation value of the side area, a standard deviation value of the volume, and a standard deviation value of the length / width ratio, 3. A method for calculating a sized grain weight ratio using the grain quality discriminator according to claim 1 or 2, wherein one or more of these standard deviation values ​​are selected and used.

Images