JP2020041841A - Change rate estimating apparatus, change rate estimating method, state estimating device, and state estimating method - Google Patents

Abstract

To provide a change rate estimating apparatus with which it is possible to easily estimate a change rate of a shape in the thickness direction of rice without requiring a device of complex configuration.SOLUTION: The change rate estimating apparatus comprises: an imaging device 2 for imaging the state of rice immersed in water; an image data acquisition unit 11 for acquiring the two-dimensional image data of the rice imaged by the imaging device 2 as seen from the thickness direction of the rice; a crack rate calculation unit 16 for referring to the two-dimensional image data acquired by the image data acquisition unit 11 and calculating a vertical crack rate; and an estimation unit 19 for acquiring the vertical crack rate from the crack rate calculation unit 16, and estimating the change rate of shape of the rice in the thickness direction on the basis of the correlation of the vertical crack rate with a predetermined vertical crack rate and change rate of shape in the thickness direction of rice.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a change rate estimating apparatus and a change rate estimating method for estimating information on a target rice, and further calculates a specific gravity, density or water absorption of the target rice to estimate a state of the target rice based on the estimated information. And a state estimating method.

  For example, in sake brewing, a processing step for raw rice, such as a rice polishing step, a water absorption step, and a steaming step, is an important process that determines the subsequent brewing step. In the water absorption process when manufacturing Daiginjo sake etc., so-called limited water absorption is sometimes performed, which is to limit the water absorption of the raw rice by shortening the immersion time so that the raw rice does not absorb too much water. In this case, it is necessary to adjust the water absorption in units of 1%. Therefore, in order to improve the quality of liquor to be produced, it is extremely important to grasp the state of water absorption of raw rice in detail in real time.

  It is known that the change in water absorption of raw rice varies greatly depending on the state of the raw rice (variety, polishing rate, moisture content, etc.) and immersion conditions (time, temperature, concentration, etc.). As a method of evaluating the change in water absorption, a method of measuring the weight of raw rice before and after immersion with a weighing machine and calculating the water absorption from the difference in weight before and after immersion was generally used. However, with such a method, it is not possible to accurately grasp the change over time in water absorption and the distribution of water absorption. Therefore, at an actual brewing site, the water absorption distribution is estimated visually from the change in the color and shape of the raw rice, and the change in water absorption is estimated in combination with the above-described calculation result of the water absorption. The development of a method that can be evaluated in real time was required.

  Incidentally, the applicant of the present application has proposed a water-containing state determination method described in Patent Literature 1 as a method of determining the water absorption and the water absorption distribution of rice to be cooked by image observation. According to this water content determination method, since the rice to be cooked placed in a state of being immersed in water is optically monitored, the water content related information on the water content of the rice to be cooked can be measured over time. In addition, it is possible to evaluate the water absorption and the water absorption distribution of the rice to be cooked based on the water content-related information. Specifically, a change in the size of the rice to be cooked due to water absorption of the rice to be cooked is measured over time, and the size of the rice to be cooked after water absorption is compared with the size of the rice to be cooked before water absorption. Since the ratio of the change (expansion) can be used as the expansion rate, the water absorption and the water absorption distribution of the rice to be cooked can be evaluated based on the expansion rate.

JP-A-2006-105074

  According to the method for determining a water-containing state described in Patent Document 1, it is possible to evaluate the water absorption amount and water absorption distribution (water absorption state) of the rice to be cooked. However, the water absorption state of rice varies greatly with rice varieties, and rice to be cooked expands mainly in the direction perpendicular to the vertical direction due to water absorption, while rice to be brewed is orthogonal to the vertical direction due to water absorption. Expand not only in the vertical direction but also in the vertical direction. Therefore, when the method for determining a water-containing state described in Patent Document 1 described above is used, for example, simply monitoring the rice to be cooked placed in a state of being immersed in water optically from above in the vertical direction, the direction orthogonal to the vertical direction. It is possible to know the degree of expansion in the vertical direction, but not to know the degree of expansion in the vertical direction. Therefore, when the water-containing state of the rice expanding in the vertical direction (eg, rice for sake brewing) is evaluated using the water-containing state determination method described in Patent Document 1, the water content-related information reflects the vertical expansion. As a result, a difference occurs between the measured moisture content-related information and the actual moisture content of the rice to be brewed, and as a result, the water absorption state of the rice to be brewed cannot be accurately evaluated.

  In other words, in order to accurately evaluate the water absorption state of rice irrespective of rice varieties, etc., it was considered based only on the rate of change in the two-dimensional shape of the rice to be evaluated (change rate of the two-dimensional shape). Is not sufficient, and it is important to consider the change rate of the shape in the direction orthogonal to the two-dimensional direction and the change rate of the shape in the three-dimensional direction (change rate of the three-dimensional shape).

  If the rice is optically monitored from a direction perpendicular to the vertical direction using the water content determination method described in Patent Document 1, the rate of change of the shape in the vertical direction and the rate of change of the three-dimensional shape can be reduced. In this case, a device for optically monitoring the rice (such as a lighting device for irradiating the target rice with light or a camera device for capturing the transmitted light from the target rice) may be provided in each direction. Therefore, there is a concern that the device configuration becomes complicated and the device cost increases.

  In addition, the rate of change of the shape and the rate of change of the three-dimensional shape in the direction orthogonal to the two-dimensional direction can be used not only for evaluating the water absorption state of rice, but also for example, in the direction orthogonal to the two-dimensional direction (rice thickness). Calculating the specific gravity of rice based on the rate of change of the shape in the direction) or the rate of change of the three-dimensional shape, it is possible to evaluate the whiteness of the rice. There are many advantages of obtaining the rate of change of the rice and the rate of change of the three-dimensional shape of rice, and there is a situation in which the development of a method that can easily obtain these rates is required.

  SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and can easily determine the rate of change in the shape of rice in the thickness direction or correct the rate of change of the two-dimensional shape of rice without the need for a device having a complicated configuration. Of a change rate estimating apparatus and a change rate estimating method that can easily obtain a change rate of a three-dimensional shape by performing the method, and a change rate of a shape and a change rate of a three-dimensional shape in a direction orthogonal to the two-dimensional direction. It is an object of the present invention to provide a state estimating apparatus and a state estimating method which can easily estimate the specific gravity, density or water absorption of rice based on the above.

The characteristic configuration of the change rate estimation device according to the present invention for achieving the above object is as follows.
Photographing means for photographing the target rice immersed in water,
Image data acquisition means for acquiring two-dimensional image data of the target rice photographed by the photographing means, as viewed from the thickness direction of the target rice;
The two-dimensional image data is analyzed with reference to the two-dimensional image data acquired by the image data acquiring means, and a longitudinal crack having an open longitudinal section along a long axis direction of the target rice among a plurality of target rice is opened. Crack rate calculating means for calculating a vertical crack rate which is a ratio of target rice in which cracks have occurred,
The vertical cracking rate is obtained from the cracking rate calculating means, and the vertical cracking rate, and the predetermined correlation between the vertical cracking rate of the target rice and the change rate of the shape in the thickness direction are determined based on the correlation. Estimating means for estimating the rate of change in the shape of rice in the thickness direction.
Further, the characteristic configuration of the change rate estimation method according to the present invention for achieving the above object,
A photographing process of photographing the target rice immersed in water,
On the basis of a two-dimensional image of the target rice that has been photographed and the target rice viewed from the thickness direction, a vertical crack having an open longitudinal section along the long axis direction of the target rice among a plurality of target rice occurs. Cracking ratio calculating step of calculating a vertical cracking ratio which is a ratio of target rice having
Based on the calculated vertical cracking rate, and a predetermined correlation between the vertical cracking rate of the target rice and the change rate of the shape in the thickness direction, the target rice, the change rate of the shape in the thickness direction is calculated. And an estimating step of estimating.

  Here, the inventor of the present application, in the course of studying a method of determining the water absorption rate and water absorption distribution of rice by image observation, the manner of cracking of the rice immersed in water differs depending on the rice, and the rate of water absorption depends on the way the rice cracks. And the state of water absorption, in other words, the rate of change in the three-dimensional shape of rice was different. As a method of determining how to break, there is a method of visually checking, but it is difficult to objectively record the changing manner of rice that changes every moment as a numerical value. Therefore, the inventors of the present application have conducted intensive studies, and as a result, succeeded in numerically analyzing the cracking of rice by performing automatic analysis using two-dimensional image data on the two-dimensional image data. New knowledge was obtained that the ratio of cracked rice was correlated with the rate of shape change in the thickness direction of the rice.

  The change rate estimating apparatus and the change rate estimating method according to the present invention are based on such knowledge, and in the above-described feature configuration, first, the target rice in a state of being immersed in water is viewed from the thickness direction. A two-dimensional image is photographed by the photographing means, and two-dimensional image data photographed by the photographing means is acquired by the image data acquiring means. Next, the cracking ratio calculating means analyzes the two-dimensional image data with reference to the two-dimensional image data, and calculates a vertical cracking ratio, which is a ratio of target rice having a vertical crack, among a plurality of target rices. calculate.

  Thereafter, the estimating means obtains the vertical cracking rate, and based on the vertical cracking rate, and a predetermined correlation between the vertical cracking rate of the target rice and the shape change rate in the thickness direction, the target rice The rate of change of the shape in the thickness direction is estimated.

  As described above, according to the characteristic configuration, the vertical crack rate having a correlation with the rate of change in the shape of the target rice in the thickness direction is calculated, and based on the vertical crack rate and the correlation, the thickness direction of the target rice is calculated. Since the rate of change of the shape of the target rice can be estimated, the rate of change of the shape of the target rice in the thickness direction can be easily obtained without optically monitoring the target rice from a direction orthogonal to the thickness direction as in the related art. be able to. Therefore, for example, the water absorption state of the target rice, the whiteness of the target rice, and the like can be evaluated based on the obtained change rate of the shape in the thickness direction.

The feature configuration of the state estimation device according to the present invention for achieving the above object is as follows.
The change rate estimation device,
A point calculating unit that obtains the shape change rate in the thickness direction from the estimating unit and calculates the specific gravity, density, or water absorption of the target rice based on the obtained shape change rate in the thickness direction. It is in.
The feature configuration of the state estimation method according to the present invention for achieving the above object is as follows.
After performing the change rate estimation method, based on the change rate of the shape in the thickness direction estimated by the change rate estimation method, performing a state calculation step of calculating the specific gravity, density, or water absorption of the target rice. It is in.

  According to the above feature configuration, it is possible to calculate the specific gravity, density, or water absorption of the target rice based on the shape change rate in a direction orthogonal to the thickness direction, which is easily obtained without the need for an optical monitoring device. Therefore, the state of the target rice such as specific gravity, density and water absorption can be more easily estimated than before.

Further, a further characteristic configuration of the state estimation device according to the present invention is as follows.
Area information acquisition for analyzing the two-dimensional image data with reference to the two-dimensional image data acquired by the image data acquisition means and acquiring the area of the target rice in a direction along a plane orthogonal to the thickness direction With means,
In the state calculating means, the area is obtained from the area information obtaining means, and the obtained shape change rate in the thickness direction, the obtained area, and the thickness of the target rice before immersion in pre-stored water. The specific gravity or density of the target rice is calculated based on the volume of the target rice calculated based on the following formula.
Further, a further characteristic configuration of the state estimation method according to the present invention includes:
An area information acquisition step of acquiring an area of the target rice in a direction along a plane orthogonal to the thickness direction based on a two-dimensional image of the target rice that has been photographed and viewing the target rice from the thickness direction. ,
In the state calculation step, based on the shape change rate in the thickness direction, the acquired area, and the volume of the target rice calculated based on the thickness of the target rice before immersion in water, the target rice The point is to calculate the specific gravity or density.

  According to the above feature configuration, the area of the target rice in a direction along a plane orthogonal to the thickness direction is obtained based on the two-dimensional image data, and the obtained area and the shape in the thickness direction that are more easily obtained than before are obtained. In addition to the rate of change, the thickness of the target rice before immersion in water is determined in advance, the volume is calculated based on these, and the specific gravity or density of the target rice is calculated based on this volume. . Therefore, the state of specific gravity or density of the target rice can be estimated more easily than before.

The characteristic configuration of the change rate estimation device according to the present invention for achieving the above object is as follows.
Photographing means for photographing the target rice immersed in water,
Image data acquisition means for acquiring two-dimensional image data of the target rice photographed by the photographing means, as viewed from the thickness direction of the target rice;
The two-dimensional image data is analyzed with reference to the two-dimensional image data acquired by the image data acquiring means, and the rate of change of the two-dimensional shape of the target rice in a direction along a plane orthogonal to the thickness direction is calculated. A two-dimensional shape change rate calculating means for calculating;
The rate of change of the two-dimensional shape is obtained from the two-dimensional shape change rate calculating means, and a target rice having a vertical crack having an open longitudinal section along a long axis direction of the target rice among a plurality of target rices The rate of change of the two-dimensional shape is corrected based on the correlation between the vertical cracking rate, which is the ratio of the shape, and the rate of change of the shape in the thickness direction, and the obtained rate of change of the two-dimensional shape. Estimating means for estimating the rate of change.
Further, the characteristic configuration of the change rate estimation method according to the present invention for achieving the above object,
A photographing process of photographing the target rice immersed in water,
Calculating a change rate of a two-dimensional shape of the target rice in a direction along a plane orthogonal to the thickness direction based on a two-dimensional image of the target rice taken in the thickness direction of the target rice; Dimensional shape change rate calculating step,
Among a plurality of target rice, a correlation between a vertical cracking rate which is a ratio of target rice in which a vertical section along a longitudinal direction along the long axis direction of the target rice is open and a rate of shape change in the thickness direction. And an estimation step of correcting the change rate of the two-dimensional shape based on the calculated change rate of the two-dimensional shape and estimating the change rate of the two-dimensional shape as the change rate of the three-dimensional shape.

  As described above, the inventors of the present invention differ in the rate of change in the three-dimensional shape of rice depending on how the rice is cracked, and the ratio of rice that has vertical cracks in the rice varies in the thickness direction of the rice. New knowledge that there is a correlation with the shape change rate in.

  The change rate estimating apparatus and the change rate estimating method according to the present invention are based on such knowledge, and in the above configuration, first, a two-dimensional image of the target rice in a state of being immersed in water, viewed from the thickness direction. Is photographed by the photographing means, and the image data acquiring means acquires two-dimensional image data photographed by the photographing means. Next, the two-dimensional shape change rate calculating means analyzes the two-dimensional image data with reference to the two-dimensional image data, and calculates a two-dimensional shape change rate of the target rice in a direction along a plane orthogonal to the thickness direction. calculate.

  Thereafter, the estimating means obtains the change rate of the two-dimensional shape, and calculates the two-dimensional shape based on the correlation between the vertical crack rate and the change rate of the shape in the thickness direction and the obtained change rate of the two-dimensional shape. The change rate of the two-dimensional shape calculated by the change rate calculating means is estimated as the change rate of the three-dimensional shape.

  As described above, according to the characteristic configuration, the change rate of the two-dimensional shape can be corrected to the change rate of the three-dimensional shape based on the correlation and the obtained change rate of the two-dimensional shape. Regardless of the variety of the target rice, the change rate of the three-dimensional shape of the target rice can be easily estimated. Therefore, based on the estimated change rate of the three-dimensional shape, for example, the water absorption state of the target rice, the whiteness of the target rice, and the like can be evaluated.

A further characteristic configuration of the change rate estimating apparatus according to the present invention is that the estimating means multiplies the degree of influence of the vertical crack determined based on the correlation by the obtained change rate of the two-dimensional shape to obtain the two-dimensional shape. The point is that the rate of change is corrected to the rate of change of the three-dimensional shape.
Further, a further characteristic configuration of the change rate estimating method according to the present invention is that, in the estimating step, the degree of vertical crack influence determined based on the correlation is multiplied by the calculated change rate of the two-dimensional shape to obtain the two-dimensional shape. The point is that the change rate of the three-dimensional shape is corrected to the change rate of the three-dimensional shape.

  Here, the “vertical crack influence degree” is an index indicating the influence of the presence of a vertical crack on the rate of change of the three-dimensional shape, and specifically, a database obtained from measurement results for a plurality of varieties in advance. A graph showing the correlation between the vertical cracking rate and the rate of change in the thickness direction is created based on the above, and the inclination of this graph is determined in advance as the vertical cracking influence degree.

  According to the above feature configuration, it is possible to correct the change rate of the two-dimensional shape by multiplying the change rate of the two-dimensional shape of the target rice by the degree of vertical cracking to obtain the change rate of the three-dimensional shape of the target rice. it can. Therefore, similarly to the above, the change rate of the three-dimensional shape of the target rice can be easily obtained, and the water absorption state of the rice and the whiteness of the rice can be evaluated.

Further, the characteristic configuration of the state estimation device according to the present invention for achieving the above object, any one of the above change rate estimation device,
The three-dimensional shape change rate is obtained from the estimating means, and a state calculating means for calculating the specific gravity, density or water absorption of the target rice based on the obtained three-dimensional shape change rate. .
In addition, a characteristic configuration of the state estimation method according to the present invention for achieving the above object includes, after performing any one of the change rate estimation methods described above, calculating the change rate of the three-dimensional shape estimated by the change rate estimation method. Based on this, a state calculation step of calculating the specific gravity, density or water absorption of the target rice is performed.

  According to the above-mentioned feature configuration, the specific gravity of the target rice is determined based on the change rate of the three-dimensional shape obtained by correcting the change rate of the two-dimensional shape, that is, the change rate of the three-dimensional shape easily obtained than before. Since the density or the water absorption can be calculated, the state of the target rice such as specific gravity, density, and water absorption can be more easily estimated than before.

A further characteristic configuration of the state estimation device according to the present invention is that, in the state calculation means, the acquired rate of change of the three-dimensional shape, and the thickness and thickness direction of the target rice before immersion in water stored in advance. The point is that the specific gravity or density of the target rice is calculated based on the volume of the target rice calculated based on the area in the direction along the orthogonal plane.
Further, a further characteristic configuration of the state estimation method according to the present invention includes, in the state calculation step, a change rate of the three-dimensional shape, and a thickness and a thickness direction of the target rice before immersion in pre-stored water. The point is that the specific gravity or density of the target rice is calculated based on the volume of the target rice calculated based on the area in the direction along the orthogonal plane.

  According to the above-mentioned characteristic configuration, in addition to the change rate of the three-dimensional shape, which is more easily obtained than before, the thickness of the target rice before immersion in water and the area in a direction along a plane orthogonal to the thickness direction are obtained in advance. In advance, the volume is calculated based on these, and the specific gravity or density of the target rice is calculated based on the volume. Therefore, the state of specific gravity and density of the target rice can be more easily estimated than before.

In the present application, “specific gravity” means specific gravity (cm ³ / g) as a reciprocal of density (g / cm ³ ).

It is a figure showing composition of a state estimating device and a change rate estimating device concerning a 1st embodiment. It is a flowchart explaining the process of a state estimation method and a change rate estimation method. FIG. 4 is a diagram illustrating an example of image processing. It is a figure showing an example of a form of sake rice. It is a graph which shows the relationship between the rate of change of the two-dimensional shape and the rate of change of the three-dimensional shape of sake rice, and water absorption. It is a graph which shows the relationship between a vertical crack rate and a thickness change rate. It is a flowchart explaining a machine learning process. It is a graph which shows the relationship between the number of times of learning and the correct answer rate. It is a graph which shows the relationship between the number of times of learning and the correct answer rate. It is a graph which shows the relationship between the number of times of learning and the correct answer rate. It is a figure showing composition of a state estimating device and a change rate estimating device concerning a 2nd embodiment. It is a flowchart explaining a machine learning process. It is a flowchart explaining a validity determination process. It is a figure showing composition of a state estimating device and a change rate estimating device concerning a 3rd embodiment. It is a flowchart explaining the process of a state estimation method and a change rate estimation method.

[First Embodiment]
Hereinafter, a state estimation device 1 and a change rate estimation device 20A according to an embodiment of the present invention will be described with reference to the drawings. In this embodiment, a case where the rate of change in the shape of sake rice in the thickness direction is estimated will be described as an example.

  As shown in FIG. 1, a state estimation device 1A (1) according to the present embodiment includes an imaging device (imaging unit) 2, an image data obtaining unit (image data obtaining unit) 11, an image processing unit 12, an area An information acquisition unit (area information acquisition unit) 13, a crack ratio calculation unit (crack ratio calculation unit) 16, an estimation unit (estimation unit) 19, and a state calculation unit (state calculation unit) 30. Is provided. In FIG. 1, the state estimating device 1 includes a change rate estimating device 20A (20) including a photographing device 2, an image data acquiring unit 11, an image processing unit 12, a crack rate calculating unit 16, an estimating unit 19, and a storage device 35; It is configured to include an area information acquisition unit 13 and a state calculation unit 30. The image data acquisition unit 11, the image processing unit 12, the area information acquisition unit 13, the crack rate calculation unit 16, and the estimation unit 19 , The state calculator 30 and the storage device 35 are shown in a form that constitutes the rice analyzer 10A (10).

  The rice analyzing apparatus 10A constituting a part of the state estimating apparatus 1 is realized using one or a plurality of computer devices having an information processing function, an information input / output function, an information storage function, and the like. . In that case, the function of the image data acquisition unit 11, the function of the image processing unit 12, the function of the area information acquisition unit 13, and the crack rate calculation unit 16 (the function of the form determination unit 17, the function of the form data analysis unit 18) A program that causes a computer device to implement the functions of the above, the function of the estimation unit 19, and the function of the state calculation unit 30 may be installed in the computer device.

  The storage device 35 of the present embodiment includes an image data storage unit 35a that stores image data captured by the imaging device 2, an area storage unit 35b that stores the area information acquired by the area information acquisition unit 13, and machine learning. It includes a training data storage unit 35c that stores training data to be used, and a form storage unit 35d that stores the determination result of the form determination unit 17. Here, the area storage unit 35b and the form storage unit 35d store the area information acquired by the area information acquisition unit 13 and the determination result of the form determination unit 17 in one target rice after the start of immersion in water. May be stored in a state in which the change history of the area and the form can be understood. The thickness of the sake rice before immersion in water is stored in the storage device 35 of the present embodiment in advance.

  The photographing device 2 is arranged vertically above a rice holding unit that arranges sake rice in a state of being immersed in water, and shoots a two-dimensional image of the sake rice by photographing transmitted light from the sake rice. It is a device that can do.

  The output device 3 is a display device that can display image information, character information, and the like, and a printing device that can print the image information, character information, and the like on paper.

  FIG. 2 shows a step of acquiring the area information of the sake rice (area information acquisition step) and a step of calculating the crack rate of the sake rice (crack rate calculation) included in the change rate estimation method and the state estimation method according to the present embodiment. Step), a step of estimating a change in the shape of the sake rice in the thickness direction (estimation step), and a step of calculating the specific gravity of the sake rice based on the estimated rate of change of the shape in the thickness direction (state calculation step). FIG. After the process # 10 of FIG. 2 is performed, that is, after the photographing process of photographing the sake rice arranged in the state of being immersed in water by the photographing device 2 is performed, in the process # 11, the image data acquiring unit 11 is performed. Acquires two-dimensional image data obtained by photographing the sake rice arranged in a state of being immersed in water with the photographing device 2 from the thickness direction of the sake rice. That is, an image data acquisition step of acquiring two-dimensional image data of the sake rice placed in a state of being immersed in water as viewed from the thickness direction of the sake rice is executed. For example, the imaging device 2 obtains one two-dimensional image data by imaging a plurality of sake rice arranged in a state of being immersed in water. Then, the two-dimensional image data is transmitted to the rice analyzing apparatus 10A, and the image data acquiring unit 11 acquires the two-dimensional image data. The two-dimensional image data acquired by the image data acquisition unit 11 can be stored in the image data storage unit 35a of the storage device 35. For example, image data A shown in FIG. 3 is two-dimensional image data acquired by the image data acquisition unit 11.

  Next, in step # 12 of FIG. 2, the image processing unit 12 obtains image data prior to the process of obtaining area information by the area information obtaining unit 13 and the process of calculating the vertical cracking ratio by the cracking ratio calculation unit 16. In the two-dimensional image data acquired by the unit 11, image processing including a process of extracting two-dimensional sake rice image data of an area including only one sake rice is performed, and a processed two-dimensional image of the one sake rice is performed. Generate data. That is, in the two-dimensional image data acquired in the image data acquiring step, image processing including a process of extracting two-dimensional sake rice image data of a region including only one sake rice is performed, and the processing of the one sake rice is completed. An image processing step of generating the two-dimensional image data is performed. FIG. 3 is a diagram illustrating an example of image processing. In the present embodiment, as the image processing performed by the image processing unit 12, in the image data acquired by the image data acquiring unit 11, a process of extracting two-dimensional sake rice image data of an area including only one sake rice, The process includes a process of rotating the two-dimensional image of the sake rice so that the sake rice has a predetermined posture, and a process of adjusting the contrast of the image.

  For example, the two-dimensional image data B illustrated in FIG. 3 is an image obtained by extracting data including a two-dimensional whole image of one sake rice from the image data A acquired by the image data acquiring unit 11. The two-dimensional image data C shown in FIG. 3 is obtained by removing a two-dimensional image of sake rice other than one sake rice from the two-dimensional image data B to obtain a two-dimensional sake of an area including only one sake rice. It is an image after performing the process of extracting rice image data. The two-dimensional image data D illustrated in FIG. 3 is an image obtained by performing a process on the two-dimensional image data C to rotate the two-dimensional image of the sake rice so that one sake rice has a predetermined posture. . In this case, the sake rice image is rotated so that the long axis of the sake rice is along the vertical direction. The two-dimensional image data E illustrated in FIG. 3 is an image obtained by performing a process of adjusting the direction of sake rice (for example, a vertical inversion process or a horizontal inversion process) on the two-dimensional image data. The two-dimensional image data F shown in FIG. 3 is an image obtained by performing a process of adjusting the contrast of the image on the two-dimensional image data E.

  Next, the area information acquisition unit 13 analyzes the two-dimensional image data with reference to the two-dimensional image data acquired by the image data acquisition unit 11, and outputs the sake in a direction along a plane orthogonal to the thickness direction of the sake rice. Calculate the area of rice.

  In the present embodiment, the area information acquisition unit 13 refers to the two-dimensional image data acquired by the image data acquisition unit 11 and processed by the image processing unit 12, and refers to a direction along a plane orthogonal to the thickness direction of the sake rice. Can be automatically determined using various image recognition techniques.

  As described above, the two-dimensional image data as the input data referred to by the area information acquisition unit 13 to determine the area of the sake rice is the image data acquired by the image data acquisition unit 11 by the image processing unit 12. A process of extracting two-dimensional sake rice image data of an area including only one sake rice, a process of rotating a two-dimensional image of sake rice so that one sake rice has a predetermined posture, and adjusting the image contrast. This is processed two-dimensional image data after performing image processing including processing. That is, the area information acquisition unit 13 can determine the area of the sake rice from the two-dimensional image data (processed two-dimensional image data) that is in a state suitable for determining the area of the sake rice, and the determination result is More likely to be appropriate.

  In step # 13 of FIG. 2, the area information acquisition unit 13 determines the area of the sake rice. That is, an area determining step (area information obtaining step) for determining the area of one sake rice in the processed two-dimensional image data generated in the image processing step is executed. In the present embodiment, the area of one sake rice in the processed two-dimensional image data (two-dimensional image data F in FIG. 3) generated by the area information acquisition unit 13 is determined. When there are a plurality of sake rice immersed in water, how many of the sake rice the area information acquisition unit 13 determines the area of the sake rice can be appropriately set. Then, the determination result by the area information acquisition unit 13 is stored in the area storage unit 35b for each sake rice in association with the identifier assigned to each of the plurality of sake rice. The area storage unit 35b stores the area information acquired by the area information acquisition unit 13 in a state where the change history of the area of one sake rice after the start of immersion in water is known.

  As described above, in the present embodiment, the area information obtaining unit 13 determines the area of one sake rice in the processed two-dimensional image data generated by the image processing unit 12. As a result, a determination result can be obtained in a short time for the area of sake rice.

  The area information acquired by the area information acquiring unit 13 can be output from the output device 3.

  Next, the cracking rate calculating unit 16 analyzes the two-dimensional image data with reference to the two-dimensional image data acquired by the image data acquiring unit 11, and a vertical crack has occurred in a plurality of sake rice. Calculate the vertical cracking rate, which is the ratio of sake rice. The “vertical crack” is a state in which the gap of the vertical crack generated in the sake rice is large.

  In the present embodiment, the crack rate calculation unit 16 includes a shape determination unit 17 and a shape data analysis unit 18. For example, the morphological determination unit 17 can automatically determine whether cracked rice wine has been generated using various image recognition techniques. In the example described later, a case will be described where the morphological determination unit 17 predicts and determines the morphology of sake rice using an algorithm such as CNN (Convolutional Neural Network). Also, a case will be described where the morphological determination unit 17 performs machine learning such that when processed two-dimensional image data generated by the image processing unit 12 is input, information on the morphology of sake rice is output. For example, in machine learning, a plurality of weight parameters of the CNN constituting the algorithm of the form determination unit 17 are adjusted.

  As described above, the two-dimensional image data as the input data that is referred to by the morphological determination unit 17 to determine the morphology of the sake rice is such that the image processing unit 12 uses one of the image data acquired by the image data acquisition unit 11. A process of extracting two-dimensional sake rice image data of an area including only one sake rice, a process of rotating a two-dimensional image of sake rice so that one sake rice has a predetermined posture, and a process of adjusting image contrast Is processed two-dimensional image data after image processing including. That is, the morphological determination unit 17 can determine the morphology of the sake rice from the two-dimensional image data (processed two-dimensional image data) in a state suitable for determining the morphology of the sake rice. Is more likely to be.

  In step # 14 of FIG. 2, the form determining unit 17 determines whether there is a crack. That is, a form determination step (crack rate calculation step) for determining whether or not one sake rice in the processed two-dimensional image data generated in the image processing step has a crack is executed. In the present embodiment, the form determination unit 17 uses the execution result of the machine learning of the accumulated training data to process the two-dimensional image data generated by the image processing unit 12 (two-dimensional image data F in FIG. 3). It is determined whether or not one sake rice has cracks.

  In the present embodiment, the morphological determination unit 17 determines whether the sake rice is in an intact form, the sake rice has a crack, or a vertical crack having an open longitudinal section along the long axis direction of the sake rice occurs. One of the four types is determined, that is, the type of the brewed rice, or the type in which the horizontal cross section along the short axis direction of the sake rice has an open lateral crack. Among them, as shown in Table 1 below, the form in which the sake rice is not damaged and the form in which the sake rice has cracks are determined to be "no crack", and the form in which the sake rice has vertical cracks. , And the mode in which the sake rice has horizontal cracks is determined to be "cracked". When there are a plurality of sake rice immersed in water, it is possible to appropriately set how many of the sake rice the form determination unit 17 determines the form of the sake rice.

  The algorithm of the morphological determination unit 17 is constructed such that when the processed two-dimensional image data generated by the image processing unit 12 is input, information on the morphology of sake rice is output. For example, the CNN executed by the form determination unit 17 performs processing on the input layer to which the processed two-dimensional image data generated by the image processing unit 12 is input and the processed two-dimensional image data input to the input layer. A hierarchical network including a convolution layer that performs a convolution process to obtain a feature map, a pool layer that reduces a feature map output from the convolution layer, a fully connected layer that connects all units, and an output layer. Make up. The combination of the convolutional layer and the pool layer is repeatedly provided a plurality of times, and the total connection layer is also provided in a plurality of layers.

  In the present embodiment, four types of information (no damage, cracks, vertical cracks, and horizontal cracks) are assumed as the information on the morphology of the sake rice, which is the determination result of the morphological determination unit 17, and therefore, the unit of the output layer of the CNN. The number becomes four. In the output layer, a softmax function can be used as a likelihood function. The softmax function divides the value of each unit by the value obtained by accumulating the values of all units in the output layer and normalizes the value, so that the likelihood of each unit in the output layer takes a value between 0 and 1. Then, the form determining unit 17 determines the class having the maximum likelihood as the classification class, that is, the form (determination result) of the sake rice. The determination result by the form determining unit 17 is stored in the form storage unit 35d for each sake rice in association with the identifier given to each of the plurality of sake rice. The determination result of the configuration determination unit 17 is stored in the configuration storage unit 35d in a state where the change history of the configuration of one sake rice after the start of immersion in water is known.

  FIG. 4 is a diagram illustrating an example of a state of sake rice. Specifically, FIG. 4A is a view showing a form in which the sake rice is not damaged, and FIG. 4B is a form in which cracks are generated in the sake rice along the short axis direction. 4 (c) and 4 (d) show a form in which cracks are generated in a state in which the gaps of cracks generated in sake rice are large, and the form of the cracks is along the short axis direction of the sake rice. There is a form in which a horizontal crack with an open cross section is generated (FIG. 4C), and a form in which a vertical crack with an open vertical cross section along the long axis direction of sake rice is generated (FIG. 4D). As described above, the shape determining unit 17 may determine whether the crack is in a form in which a horizontal cross section along the short axis direction of the sake rice is open or a vertical cross section along the long axis direction of the sake rice. It is also determined whether or not is a form in which an open vertical crack has occurred.

  As described above, in the present embodiment, instead of the human judging the form of the sake rice, the form judging unit 17 generates the image processing unit 12 using the machine learning execution result of the accumulated training data. The form of one sake rice in the processed two-dimensional image data is determined. As a result, a consistent determination result can be obtained in a short time with respect to the form of sake rice.

  In step # 15 of FIG. 2, the morphological data analysis unit 18 performs vertical cracking based on the determination result of the morphological determination unit 17 for each of the plurality of sake rice included in the two-dimensional image data acquired by the image data acquisition unit 11. Calculate the rate. That is, a form data analysis step (a crack rate calculation step) of calculating a vertical crack rate based on the determination result of the form determination step is executed. In the form storage unit 35d, the determination result of the form determination unit 17 for each of the plurality of sake rice is stored in a state where the change history of the form of one sake rice after the start of immersion in water is known. I have. Further, in the form storage unit 35d, at least two or more forms from the start of immersion in water to the present (latest) of each sake rice are stored as a change history. Then, the morphological data analysis unit 18 reads out the current (latest) morphology of the plurality of sake rice stored in the morphology storage unit 35d, and among the plurality of sake rice, the one of the sake rice in which the vertical crack occurs. Calculate the vertical crack ratio which is a ratio.

  Note that the determination result of the form determination unit 17 can be output from the output device 3. For example, the ratio (crack rate) of cracked rice among a plurality of sake rice may be calculated as needed and output from the output device 3 in real time. If the determination result of the shape determination unit 17 is stored in the form storage unit 35d in a state in which the change history of the shape of one sake rice after the start of immersion in water is known, the cracking rate Can be calculated and output from the output device 3.

  The estimating unit 19 obtains the vertical cracking rate of the sake rice calculated by the cracking rate calculation unit 16 and obtains the obtained vertical cracking rate, and the predetermined vertical cracking rate of the sake rice and the change rate of the shape in the thickness direction. The change rate of the shape of the sake rice in the thickness direction is estimated based on the correlation.

  The inventor of the present application has studied a method of determining a water absorption state such as a water absorption rate and a water absorption distribution of sake rice by image observation. , A new finding that there is a correlation with the rate of change of the shape of sake rice in the thickness direction. Specifically, the inventor of the present application has determined that the rate of change of the two-dimensional shape in a plane orthogonal to the thickness direction of the sake rice measured based on the two-dimensional image data for a plurality of sake rice arranged in a state of being immersed in water. By calculating the average value and calculating the average value of the water absorption rates of a plurality of sake rice measured using a moisture meter at the time when the change rate of the two-dimensional shape is measured, the change in the two-dimensional shape of the sake rice is calculated. The relationship between the percentage and the water absorption was plotted. In addition, for a plurality of sake rice arranged in a state of being immersed in water, the average value of the change rate of the three-dimensional shape measured using a known method is calculated, and at the time when the change rate of the three-dimensional shape is measured. The relationship between the change rate of the three-dimensional shape of the sake rice and the water absorption was plotted by calculating the average value of the water absorption of a plurality of sake rice measured using a moisture meter (see FIG. 5). As shown in FIG. 5, the coefficient of determination of the approximate line for the change rate of the two-dimensional shape and the water absorption is 0.30, whereas the coefficient of determination of the approximate line for the change rate of the three-dimensional shape and the water absorption is 0.91, which indicates that the change rate of the two-dimensional shape has a lower correlation with the water absorption rate than the change rate of the three-dimensional shape.

  The inventor of the present application calculates the shape change rate in the thickness direction (thickness change rate) based on the two-dimensional shape change rate and the three-dimensional shape change rate, and calculates the shape change rate in the thickness direction. Then, the lateral cracking ratio and the vertical cracking ratio were plotted (see FIG. 6), and the correlation between the lateral cracking ratio and the vertical cracking ratio with respect to the shape change rate in the thickness direction was examined. As a result, as shown in FIG. 6, it was found that there was a high correlation between the vertical cracking rate and the shape change rate in the thickness direction. Therefore, for a plurality of varieties, a database relating to the correlation between the vertical cracking rate and the shape change rate in the thickness direction is created in advance by the same method as described above, and stored in the storage device 35 as appropriate. Based on the correlation, the change rate of the shape in the thickness direction can be estimated from the vertical crack rate.

  Therefore, in step # 16 of FIG. 2, the estimating unit 19 determines the vertical cracking rate calculated by the cracking rate calculating unit 16 and the above-described correlation between the vertical cracking rate and the shape change rate in the thickness direction. Then, the change rate of the shape of the sake rice in the thickness direction is estimated. Thus, the change rate of the shape in the thickness direction estimated by the estimation unit 19 corresponds to the change rate of the shape in the thickness direction estimated by the change rate estimation device 20A. The rate of change of the shape in the thickness direction estimated in this way can be stored in the storage device 35 and output from the output device 3. Further, it is also possible to monitor the temporal change of the shape change rate in the thickness direction and to output from the output device 3 when the shape change rate in the thickness direction becomes a predetermined value.

  The state calculation unit 30 determines the specific gravity, density, or water absorption of the sake rice based on the shape change rate in the thickness direction estimated by the estimation unit 19 (the shape change rate in the thickness direction estimated by the change rate estimation device 20A). Calculate the rate. In the present embodiment, the specific gravity of sake rice is calculated.

  Specifically, in step # 17 of FIG. 2, the state calculation unit 30 acquires the area of the sake rice from the area information acquisition unit 13 and acquires the rate of change in the shape of the sake rice in the thickness direction from the estimation unit 19. Further, the thickness of the sake rice before immersion in water stored in the storage device 35 is obtained, and the volume of the sake rice is multiplied by the obtained rate of change in shape, area, and thickness in the thickness direction. The specific gravity of the sake rice can be calculated by dividing the calculated volume by the weight of the sake rice that is separately measured and input as appropriate. The calculated specific gravity of the sake rice can be stored in the storage device 35 and output from the output device 3.

  FIG. 7 is a flowchart illustrating a machine learning process that is part of the change rate estimation method and the state estimation method. In step # 20, the image data acquisition unit 11 acquires two-dimensional image data of sake rice. That is, an image data acquisition step of acquiring two-dimensional image data of the sake rice placed in the state of being immersed in water is executed. Then, in step # 21, the image processing unit 12 performs image processing to generate processed two-dimensional image data. Thereafter, in step # 22, the morphological determination unit 17 is configured by a combination of the processed two-dimensional image data generated by the image processing unit 12 and information on the morphology of the sake rice contained in the processed two-dimensional image data. The training data is stored in the training data storage unit 35c, and machine learning of the training data is executed in step # 23. For example, for one processed two-dimensional image data, the correct data determined by a human is labeled with respect to one form of sake rice included in the processed two-dimensional image data, One training data is created. The training data storage unit 35c stores a plurality of training data created in this way.

  In the present embodiment, a softmax function is used as a likelihood function in the output layer of the CNN executed by the form determining unit 17. Since the softmax function divides the value of each unit by the value obtained by accumulating the values of all units in the output layer and normalizes it, the likelihood of each unit in the output layer takes a value between 0 and 1. Then, the form determination unit 17 reduces the error between the estimated data (“likelihood”) generated in the output layer and the correct data (“1”) by using an error back propagation method. The algorithm that corrects a plurality of weight parameters of the CNN to be executed and outputs information on the morphology of the sake rice when the processed two-dimensional image data generated by the image processing unit 12 is input is executed. Get as.

  As described above, the machine learning is performed by using the two-dimensional image data (processed two-dimensional image data) in a state suitable for determining the morphology of the sake rice as the training data, and determining the morphology of the sake rice described above. This is also performed on the two-dimensional image data (processed two-dimensional image data) that is in a state suitable for determining the form of sake rice. As a result, the possibility that the determination result of the form determination unit 17 is appropriate is increased.

  Next, effects obtained by performing machine learning using the processed two-dimensional image data as the training data as in the present embodiment will be described. 8 to 10 are graphs showing the relationship between the number of times of learning and the correct answer rate. Specifically, FIG. 8 is a graph showing the relationship between the number of times of learning and the correct answer rate when machine learning is performed using training data in which correct answer data is labeled on the two-dimensional image data B in FIG. FIG. 9 is a graph showing the relationship between the number of times of learning and the accuracy rate when machine learning is performed using the two-dimensional image data D of FIG. 3 as training data. FIG. 10 is a graph showing the relationship between the number of times of learning and the accuracy rate when machine learning is performed using the two-dimensional image data F of FIG. 3 as training data.

  As shown in FIG. 8, when the two-dimensional image data B is used as training data, the correct answer rate is less than 80% even if the number of times of learning reaches 20,000. The correct answer rate of 80% corresponds to, for example, the correct answer rate in the case where a somewhat skilled person determines the form of sake rice from two-dimensional image data. On the other hand, when machine learning is performed using training data in which the correct answer data is labeled on the two-dimensional image data D, the number of times of learning is about 12,000, and the correct answer rate exceeds 80%. When machine learning is performed using training data in which the correct data is labeled on the two-dimensional image data F, the number of times of learning is about 5,000 and the correct answer rate exceeds 80%. As described above, the processed two-dimensional image data after the image processing is performed, that is, the two-dimensional image data in a state suitable for determining the form of sake rice is used as training data, so that the number of times of learning is reduced. At least, the effect of increasing the correct answer rate was obtained.

  As described above, according to the change rate estimating apparatus 20A and the change rate estimating method, the two-dimensional image data of the plurality of sake rice arranged in the state of being immersed in water is referred to, and among the plurality of sake rice, After calculating the vertical cracking rate, which is the percentage of cracked rice, the calculated vertical cracking rate, and the correlation between the predetermined vertical cracking rate of sake rice and the rate of change in shape in the thickness direction. Based on the relationship, the change rate of the shape of the sake rice in the thickness direction can be estimated. Therefore, the change rate of the shape of the sake rice in the thickness direction can be easily estimated without the need for a device having a complicated configuration including a plurality of photographing devices 2 and the like. In addition, it is naturally possible to estimate the change rate of the shape in the thickness direction of the rice to be cooked, which expands mainly in the direction perpendicular to the vertical direction due to water absorption, using the change rate estimation device 20A and the change rate estimation method. It is.

  Further, according to the state estimation device 1A and the state estimation method, it is possible to calculate the specific gravity of sake rice based on the estimated change rate of the shape in the thickness direction, and to easily estimate the specific gravity of sake rice. Can be. The specific gravity of general rice, including sake rice, is largely related to the density of the starch in the rice, and the more sparse the starch, such as heart white, the greater the specific gravity. Therefore, by estimating the specific gravity, it is possible to evaluate the state of white heart (for example, the size). And, for example, sake rice suitable for sake brewing is said to have a large grain and a large white heart, but if the white heart is too large, cracking during rice polishing will increase, so the size of the white heart is within a certain range Since it is considered preferable to be within the range, it is also possible to evaluate the whiteness of the heart based on the estimated specific gravity, and to use the evaluation result to select rice suitable for sake brewing.

[Second embodiment]
The state estimating device 1B (1) of the second embodiment differs from the above embodiment in the content of the machine learning process. Hereinafter, a state estimation device 1B according to the second embodiment will be described, but the description of the same configuration as the above embodiment will be omitted.

  FIG. 11 is a diagram illustrating a configuration of a state estimation device 1B according to the second embodiment. As shown in the figure, the state estimation device 1B of the second embodiment further includes a validity determining unit 36 that determines the validity of the rice rice form determined by the form determining unit 17. This validity determination unit 36 is provided in the crack rate calculation unit 16 of the rice analyzer 10B (10). Then, the morphological determination unit 17 compares the combination of the information on the morphology of the sake rice determined to have high validity by the validity determination unit 36 and the processed two-dimensional image data whose morphology has been determined to the existing training data. And perform machine learning.

  Also in the present embodiment, the rice analyzer 10B is realized using one or a plurality of computer devices having an information processing function, an information input / output function, and an information storage function. In that case, the function of the image data acquisition unit 11, the function of the image processing unit 12, the function of the area information acquisition unit 13, and the crack rate calculation unit 16 (the function of the form determination unit 17, the function of the form data analysis unit 18) A program that causes a computer device to implement the functions of the above, the function of the estimation unit 19, and the function of the state calculation unit 30 may be installed in the computer device.

  The form storage unit 35d stores the determination result of the form determination unit 17 in a state where the change history of the form of one sake rice after the start of immersion in water is known. For example, in the form storage unit 35d, the determination result of the form determination unit 17 from the time when the immersion in water is started to the present (latest) for each of a plurality of sake rice to be determined for the form is The change history of the form is stored in a state where it can be understood. As a result, it is possible to know what change history the form of one sake rice shows.

  FIG. 12 is a flowchart illustrating a machine learning process according to the second embodiment. This machine learning process is performed in addition to the machine learning process (FIG. 7) described in the first embodiment. More specifically, in step # 30, the validity determining unit 36 determines the validity of the current form of the sake rice determined by the form determining unit 17. That is, a validity determination step (crack rate calculation step) of determining the validity of the form of the sake rice determined in the form determination step performed by the form determination unit 17 is performed. FIG. 13 is a flowchart illustrating a validity determination process performed by the validity determination unit 36. In step # 40, the validity determination unit 36 determines whether the certainty factor of the current form of the sake rice determined by the form determination unit 17 is equal to or greater than a predetermined value. For example, when the maximum likelihood calculated by the softmax function used in the output layer of the form determination unit 17 is set as the certainty, the maximum likelihood calculated by the softmax function is equal to or more than a predetermined value (for example, 0). If it is .8 or more, the validity determination unit 36 determines that the certainty factor of the current form of the sake rice is equal to or greater than a predetermined value, and proceeds to step # 41.

  Next, in step # 41, the validity determination unit 36 determines whether or not the change history of the form of the sake rice conforms to a predetermined standard. The predetermined criterion is the following six patterns. If the change history of the shape of the sake rice conforms to any of the six patterns, the validity determination unit 36 proceeds to step # 42 and determines that the validity is high.

  More specifically, for example, when one sake rice is continuously viewed, the sake rice does not return from the cracked form to the non-damaged form, and cracks are formed in the sake rice. There is no return to the form that occurred. That is, if the morphology of the sake rice after the start of immersion in water is appropriately determined by the morphology determination unit 17, the change history should conform to the predetermined change criterion. Therefore, in the present characteristic configuration, the validity determination unit 36 determines that the degree of certainty of the morphology of the sake rice determined by the morphology determination unit 17 is equal to or more than a predetermined value, and that the morphology of the sake rice after starting immersion in water If the change history conforms to a predetermined change criterion, it is determined that the form of the sake rice determined by the form determining unit 17 is highly appropriate. As a result, accumulation of appropriate training data and machine learning can be performed automatically.

  On the other hand, the validity determination unit 36 determines that the certainty factor is not equal to or more than the predetermined value in step # 40, and that the change history of the sake rice form does not conform to the predetermined standard in step # 41. When the determination is made, the process proceeds to step # 43, where it is determined that “the validity is low”.

  As described above, the validity determination unit 36 determines that the degree of certainty of the morphology of the sake rice determined by the morphology determination unit 17 is equal to or greater than the predetermined value, and that the morphology of the sake rice after the start of immersion in water. If the change history meets the predetermined change criterion, it is determined that the form of the sake rice determined by the form determination unit 17 is high in validity, and the certainty factor of the form of the sake rice determined by the form determination unit 17 Is less than a predetermined value, or when the change history of the morphology of the sake rice after the start of immersion in water does not meet the predetermined change criteria, the sake rice determined by the form determination unit 17 Is determined to be low in validity.

  Next, in step # 31, the combination of the information on the morphology of the sake rice determined to be highly relevant and the processed two-dimensional image data is used as training data. Then, in step # 32, the training data is added to the existing training data to execute machine learning.

  As described above, the validity determination unit 36 determines the level of validity of the sake rice form determined by the form determination unit 17, and the form determination unit 17 determines that the validity is high by the validity determination unit 36. Machine learning is performed by adding a combination of the information on the shape of the sake rice and the processed two-dimensional image data in which the shape is determined to the existing training data. That is, accumulation of appropriate training data and machine learning can be automatically performed.

[Third embodiment]
The state estimating device 1C (1) of the third embodiment differs from the above embodiment in that the estimating unit 19 estimates the rate of change of the three-dimensional shape. Hereinafter, a state estimating device 1C according to the third embodiment will be described, but the description of the same configuration as the above embodiment will be omitted.

  FIG. 14 is a diagram illustrating a configuration of a state estimation device 1C according to the third embodiment. As illustrated, the state estimation device 1C of the third embodiment includes a two-dimensional shape change rate calculation unit (two-dimensional shape change rate calculation means) 40 provided in the rice analysis device 10C. The two-dimensional shape change rate calculating unit 40 analyzes the two-dimensional image data with reference to the two-dimensional image data acquired by the image data acquiring unit 11, and calculates a direction along a plane orthogonal to the thickness direction of the sake rice. The change rate of the two-dimensional shape of the sake rice in is calculated.

  Therefore, in the present embodiment, the rice analyzer 10C is realized using one or more computer devices having an information processing function, an information input / output function, an information storage function, and the like. The function of the image data acquisition unit 11, the function of the image processing unit 12, the function of the two-dimensional shape change rate calculation unit 40 (the function of the area determination unit 41, the function of the area data analysis unit 42), and the function of the estimation unit 19 A program for causing the computer device to realize the function of the state calculation unit 30 may be installed in the computer device.

  In the present embodiment, the two-dimensional shape change rate calculation unit 40 includes an area determination unit 41 and an area data analysis unit 42. The area determination unit 41 acquires the image data With reference to the processed two-dimensional image data, the area of the sake rice in a direction along a plane orthogonal to the thickness direction of the sake rice can be automatically determined using various image recognition techniques.

  As described above, the two-dimensional image data as the input data referred to by the area determination unit 41 to determine the area of the sake rice is one in the image data acquired by the image data acquisition unit 11 by the image processing unit 12. A process of extracting two-dimensional sake rice image data of an area including only one sake rice, a process of rotating a two-dimensional image of sake rice so that one sake rice has a predetermined posture, and a process of adjusting image contrast Is processed two-dimensional image data after image processing including. That is, since the area determination unit 41 can determine the area of the sake rice from the two-dimensional image data (processed two-dimensional image data) that is in a state suitable for determining the area of the sake rice, the determination result is appropriate. Is more likely to be.

  In the present embodiment, the thickness of the sake rice before immersion in water and the area along a plane orthogonal to the thickness direction are stored in the storage device 35 in advance.

  FIG. 15 shows a process of calculating the change rate of the two-dimensional shape of sake rice (two-dimensional shape change rate calculating step), which is included in the change rate estimation method and the state estimation method according to the present embodiment. Estimating the rate of change of the three-dimensional shape by correcting the three-dimensional shape to the rate of change of the three-dimensional shape (estimating step), and calculating the specific gravity of sake rice based on the estimated rate of change of the three-dimensional shape (state calculation) 3 is a flowchart illustrating a step). Steps # 50 to # 52 in FIG. 15 are the same as steps # 10 to # 12 in FIG.

  In step # 53 of FIG. 15, the area determining unit 41 determines the area of the sake rice. That is, an area determining step (a two-dimensional shape change rate calculating step) of determining an area of one sake rice in the processed two-dimensional image data generated in the image processing step is executed. In the present embodiment, the area of one sake rice in the processed two-dimensional image data (two-dimensional image data F in FIG. 3) generated by the area determination unit 41 is determined. When there are a plurality of sake rice immersed in water, how many of the sake rice the area determination unit 41 determines the area of the sake rice can be appropriately set. The determination result by the area determination unit 41 is stored in the area storage unit 35b for each sake rice in association with the identifier given to each of the plurality of sake rice. The determination result of the area determination unit 41 is stored in the area storage unit 35b in a state where the change history of the area of one sake rice after the start of immersion in water is known.

  As described above, in the present embodiment, the area determination unit 41 determines the area of one sake rice in the processed two-dimensional image data generated by the image processing unit 12. As a result, a determination result can be obtained in a short time for the area of sake rice.

  In step # 54 of FIG. 15, the area data analysis unit 42 determines whether or not the sake rice is based on the determination result of the area determination unit 41 for each of the plurality of sake rice included in the two-dimensional image data acquired by the image data acquisition unit 11. Of the two-dimensional shape are calculated. That is, an area data analysis step (two-dimensional shape change rate calculation step) of calculating a change rate of the two-dimensional shape based on the determination result of the area determination step is executed. In the area storage unit 35b, the determination result of the area determination unit 41 for each of the plurality of sake rice (that is, the area for each sake rice) is stored as the area of one sake rice after the start of immersion in water. The change history is stored in a state where the change history can be understood. Further, in the area storage unit 35b, for each sake rice, at least two or more areas from the start of immersion in water to the present (latest) are stored as a change history. Then, the area data analysis unit 42 reads the area of the plurality of sake rice at the time of starting the immersion in water and the area of the current (latest) plurality of sake rice stored in the area storage unit 35b, Based on these two areas for each sake rice, the change rate of the two-dimensional shape of each sake rice is calculated.

  The determination result of the area determination unit 41 can be output from the output device 3, or the change rate of the two-dimensional shape of the sake rice can be calculated at any time and output from the output device 3 in real time. Furthermore, when the determination result of the area determination unit 41 is stored in the area storage unit 35b in a state where the change history of the area of one sake rice after the start of immersion in water is known, the area and the area are determined. The change rate of the two-dimensional shape over time can be calculated and output from the output device 3.

  The estimating unit 19 according to the present embodiment acquires the two-dimensional shape change rate of the sake rice calculated by the two-dimensional shape change rate calculating unit 40, and obtains the two-dimensional shape change rate and a predetermined vertical crack rate. The present embodiment is different from the above embodiments in that the change rate of the two-dimensional shape is corrected to the change rate of the three-dimensional shape based on the correlation with the change rate of the shape in the thickness direction.

  As described above, the inventor of the present application has confirmed that the rate of change of the two-dimensional shape has a lower correlation with the water absorption rate than the rate of change of the three-dimensional shape. The correlation between the vertical cracking rate and the vertical cracking rate was examined, and it was confirmed that there was a high correlation between the vertical cracking rate and the shape change rate in the thickness direction. And the inventor of the present application has found that the degree of vertical cracking, which is an index indicating the effect of the vertical cracking rate on the change rate of the three-dimensional shape of sake rice, can be determined based on the correlation. In the present embodiment, the degree of vertical cracking is a graph showing a correlation between the vertical cracking rate and the rate of change in shape in the thickness direction based on a database obtained in advance from measurement results for a plurality of types (see FIG. 6). And the inclination of this graph is determined in advance as the degree of vertical crack influence.

  Therefore, in # 55 of FIG. 15, the estimation unit 19 corrects the change rate of the two-dimensional shape by multiplying the change rate of the two-dimensional shape calculated by the two-dimensional shape change rate calculation unit 40 by the degree of vertical crack influence. Then, the change rate of the three-dimensional shape is estimated. Thus, the change rate of the three-dimensional shape estimated by the estimating unit 19 corresponds to the change rate of the three-dimensional shape estimated by the change rate estimating device 20C. The change rate of the three-dimensional shape estimated in this manner can be stored in the storage device 35 and output from the output device 3. Further, it is also possible to monitor the temporal change of the change rate of the three-dimensional shape and output the change from the output device 3 when the change rate of the three-dimensional shape becomes a predetermined value.

  The state calculator 30 of the present embodiment calculates the specific gravity and density of sake rice based on the change rate of the three-dimensional shape obtained by the estimator 19 (the change rate of the three-dimensional shape estimated by the change rate estimating device 20C). Alternatively, the water absorption is calculated. In the present embodiment, the specific gravity of sake rice is calculated.

  Specifically, in step # 56 of FIG. 15, the state calculation unit 30 acquires the change rate of the three-dimensional shape of the sake rice from the estimation unit 19, and further immerses the rice in water stored in the storage device 35 in advance. The area of the sake rice in the direction along the plane perpendicular to the thickness direction and the thickness direction of the sake rice is obtained, and the rate of change of the obtained three-dimensional shape is multiplied by the thickness and area before immersion in water. By calculating the volume of rice and dividing the calculated volume by the weight of sake rice that is separately measured and appropriately input, the specific gravity of sake rice can be calculated. Note that the calculated specific gravity of the sake rice can be stored in the storage device 35 and output from the output device 3.

  As described above, according to the change rate estimating apparatus 20C and the change rate estimating method, the change in the two-dimensional shape of the sake rice is referred to by referring to the two-dimensional image data of the plurality of sake rice arranged in a state of being immersed in water. After calculating the rate, by multiplying the change rate of the two-dimensional shape by the vertical crack influence degree determined by the correlation between the predetermined vertical crack rate of sake rice and the change rate of the shape in the thickness direction, The change rate of the two-dimensional shape can be corrected and estimated as the change rate of the three-dimensional shape. Therefore, sake rice that expands not only in the direction perpendicular to the vertical direction but also in the vertical direction due to water absorption can be easily formed into a two-dimensional shape without the need for a device having a complicated configuration including a plurality of photographing devices 2 and the like. Is corrected, the change rate of the three-dimensional shape of the sake rice can be easily estimated. In addition, using the change rate estimating apparatus 20A and the change rate estimating method, the change rate of the two-dimensional shape is corrected by correcting the change rate of the two-dimensional shape for the rice to be cooked, which expands mainly in the direction perpendicular to the vertical direction due to water absorption. Obviously it is also possible to determine the rate.

  Further, according to the state estimation device 1C and the state estimation method, the specific gravity of the sake rice can be calculated based on the estimated change rate of the three-dimensional shape, and the specific gravity of the sake rice can be easily estimated. it can. Therefore, as described above, by estimating the specific gravity, it is possible to evaluate the state of the white heart (for example, the size). For example, the state of the white heart is evaluated based on the estimated specific gravity. Utilizing the evaluation results, rice suitable for sake brewing can be selected.

[Another embodiment]
[1] In the first and second embodiments, the change rate of the shape of the sake rice in the thickness direction is estimated, and in the third embodiment, the change rate of the three-dimensional shape of the sake rice is estimated. However, the present invention is not limited to this, and it is also possible to estimate the rate of change in the shape in the thickness direction and the rate of change in the three-dimensional shape of rice other than sake rice, for example, rice to be cooked.

[2] In each of the above embodiments, the specific gravity of the sake rice is calculated by calculating the volume of the sake rice in the state calculation unit 30 and separately measuring the calculated volume and dividing the calculated volume by the weight of the sake rice that is appropriately input. Although the calculation is performed, the density of the sake rice may be calculated by dividing the weight of the sake rice by the volume.

[3] In the third embodiment, the state calculator 30 calculates the correlation between the estimated change rate of the three-dimensional shape and the previously stored three-dimensional shape of the sake rice and the water absorption of the sake rice. , The moisture content of sake rice can be calculated. That is, for a plurality of sake rice arranged in a state of being immersed in water, the average value of the change rate of the three-dimensional shape obtained in advance by a known method was calculated and used to obtain the change rate of the three-dimensional shape. Calculate the average value of the water absorption rate of a plurality of sake rice measured using a moisture meter at the time when the change rate of the three-dimensional shape is measured, and plot the relationship between the change rate of the three-dimensional shape of the sake rice and the water absorption rate. Keep it. Then, the approximate lines for the plotted points are stored in the storage device 35 as the correlation between the change rate of the three-dimensional shape of the sake rice and the water absorption referred to by the state calculation unit 30. Then, the water absorption of the sake rice can also be calculated based on the estimated change rate of the three-dimensional shape and the correlation.
In the first and second embodiments, if the area along a plane orthogonal to the thickness direction of the sake rice before immersion in water is stored in advance, the state calculation unit 30 can store the stored area in advance. The rate of change of the three-dimensional shape is calculated by multiplying the rate of change of the two-dimensional shape calculated based on the area obtained by the area information obtaining unit 13 and the rate of change of the shape in the thickness direction estimated by the estimating unit 19. With such a configuration, the water absorption of the sake rice can be calculated in the same manner as described above.

[4] In the above embodiment, the configurations of the state estimation devices 1A, 1B, and 1C and the change rate estimation devices 20A, 20B, and 20C have been described with specific examples. However, the configurations can be appropriately changed.

[5] In the first and second embodiments, an example has been described in which the morphological determining unit 17 predicts and determines the morphology of sake rice using a CNN (Convolutional Neural Network). Is not limited to the above-described CNN, and can be configured using various other algorithms such as a support vector machine (SVM), a k-nearest neighbor algorithm, and a discriminant function.
Further, the form determining unit 17 may be configured to determine whether or not cracked rice has been generated using various image recognition techniques different from the above-described embodiment. Alternatively, the form determining unit 17 may be configured to receive a result of determination made by a human as to whether or not cracked rice is generated, and use the result as its own determination result.

[6] The configuration disclosed in the above embodiment (including another embodiment) can be applied in combination with the configuration disclosed in another embodiment unless there is a contradiction. The embodiment disclosed in the present specification is an exemplification, and the embodiment of the present invention is not limited thereto, and can be appropriately modified without departing from the object of the present invention.

  INDUSTRIAL APPLICABILITY The present invention can be used for a change rate estimating device capable of estimating a change rate of a shape in a thickness direction of rice or a change rate of a three-dimensional shape of rice. It can be used for a state estimating device that can estimate the specific gravity, density and water absorption of rice based on the rate of change of the rice.

1 (1A, 1B, 1C) State estimation device 2 Imaging device 3 Output device 10 (10A, 10B, 10C) Rice analyzer 11 Image data acquisition unit 13 Area information acquisition unit 16 Crack rate calculation unit 19 Estimation unit 20 (20A, 20B, 20C) Change rate estimating device 30 State calculating unit 40 Two-dimensional shape change rate calculating unit

Claims (14)

Photographing means for photographing the target rice immersed in water,
Image data acquisition means for acquiring two-dimensional image data of the target rice photographed by the photographing means, as viewed from the thickness direction of the target rice;
The two-dimensional image data is analyzed with reference to the two-dimensional image data acquired by the image data acquiring means, and a longitudinal crack having an open longitudinal section along a long axis direction of the target rice among a plurality of target rice is opened. Crack rate calculating means for calculating a vertical crack rate which is a ratio of target rice in which cracks have occurred,
The vertical cracking rate is obtained from the cracking rate calculating means, and the vertical cracking rate, and the predetermined correlation between the vertical cracking rate of the target rice and the change rate of the shape in the thickness direction are determined based on the correlation. A change rate estimating device comprising: estimating means for estimating a change rate of the shape of rice in the thickness direction. A change rate estimating device according to claim 1,
A state calculating unit that obtains the shape change rate in the thickness direction from the estimating unit, and calculates a specific gravity, density, or water absorption of the target rice based on the obtained shape change rate in the thickness direction. Estimation device. Area information acquisition for analyzing the two-dimensional image data with reference to the two-dimensional image data acquired by the image data acquisition means and acquiring the area of the target rice in a direction along a plane orthogonal to the thickness direction With means,
In the state calculating means, the area is obtained from the area information obtaining means, and the obtained shape change rate in the thickness direction, the obtained area, and the thickness of the target rice before immersion in pre-stored water. The state estimating device according to claim 2, wherein the specific gravity or density of the target rice is calculated based on the volume of the target rice calculated based on the following. Photographing means for photographing the target rice immersed in water,
Image data acquisition means for acquiring two-dimensional image data of the target rice photographed by the photographing means, as viewed from the thickness direction of the target rice;
The two-dimensional image data is analyzed with reference to the two-dimensional image data acquired by the image data acquiring means, and the rate of change of the two-dimensional shape of the target rice in a direction along a plane orthogonal to the thickness direction is calculated. A two-dimensional shape change rate calculating means for calculating;
The rate of change of the two-dimensional shape is obtained from the two-dimensional shape change rate calculating means, and a target rice having a vertical crack having an open longitudinal section along a long axis direction of the target rice among a plurality of target rices The rate of change of the two-dimensional shape is corrected based on the correlation between the vertical cracking rate, which is the ratio of the shape, and the rate of change of the shape in the thickness direction, and the obtained rate of change of the two-dimensional shape. A change rate estimating device comprising: estimating means for estimating a change rate.   The estimation means corrects the change rate of the two-dimensional shape to the change rate of the three-dimensional shape by multiplying the degree of influence of a vertical crack determined based on the correlation by the obtained change rate of the two-dimensional shape. Item 5. The change rate estimation device according to item 4. A change rate estimating device according to claim 4 or 5,
A state estimating device comprising: a state calculator that obtains the change rate of the three-dimensional shape from the estimator; and calculates a specific gravity, density, or water absorption of the target rice based on the obtained change rate of the three-dimensional shape. .   The state calculation means calculates the change rate of the acquired three-dimensional shape, and the thickness of the target rice before immersion in water stored in advance and the area in a direction along a plane orthogonal to the thickness direction of the target rice. The state estimating device according to claim 6, wherein a specific gravity or a density of the target rice is calculated based on a volume of the target rice. A photographing process of photographing the target rice immersed in water,
On the basis of a two-dimensional image of the target rice that has been photographed and the target rice viewed from the thickness direction, a vertical crack having an open longitudinal section along the long axis direction of the target rice among a plurality of target rice occurs. Cracking ratio calculating step of calculating a vertical cracking ratio which is a ratio of target rice having
Based on the calculated vertical cracking rate, and a predetermined correlation between the vertical cracking rate of the target rice and the change rate of the shape in the thickness direction, the target rice, the change rate of the shape in the thickness direction is calculated. A rate-of-change estimating method for performing an estimating step of estimating.   A state calculation for calculating a specific gravity, density, or water absorption of the target rice based on the change rate of the shape in the thickness direction estimated by the change rate estimation method after performing the change rate estimation method according to claim 8. State estimation method for executing the process. An area information acquisition step of acquiring an area of the target rice in a direction along a plane orthogonal to the thickness direction based on a two-dimensional image of the target rice that has been photographed and viewing the target rice from the thickness direction. ,
In the state calculation step, based on the shape change rate in the thickness direction, the acquired area, and the volume of the target rice calculated based on the thickness of the target rice before immersion in water, the target rice The state estimating method according to claim 9, wherein the specific gravity or the density is calculated. A photographing process of photographing the target rice immersed in water,
Calculating a change rate of a two-dimensional shape of the target rice in a direction along a plane orthogonal to the thickness direction based on a two-dimensional image of the target rice taken in the thickness direction of the target rice; Dimensional shape change rate calculating step,
Among a plurality of target rice, a correlation between a vertical cracking rate which is a ratio of target rice in which a vertical section along a longitudinal direction along the long axis direction of the target rice is open and a rate of shape change in the thickness direction. And an estimation step of correcting the change rate of the two-dimensional shape based on the calculated change rate of the two-dimensional shape to estimate the change rate of the two-dimensional shape as the change rate of the three-dimensional shape.   In the estimating step, a vertical crack influence degree determined based on the correlation is multiplied by the calculated two-dimensional shape change rate, and the two-dimensional shape change rate is corrected to the three-dimensional shape change rate. Item 12. The change rate estimation method according to Item 11.   A state in which after performing the change rate estimation method according to claim 11 or 12, the specific gravity, density, or water absorption of the target rice is calculated based on the change rate of the three-dimensional shape estimated by the change rate estimation method. A state estimating method for executing the calculating step. In the state calculation step, the change rate of the three-dimensional shape, and the thickness of the target rice before immersion in water stored in advance and the area in a direction along a plane orthogonal to the thickness direction of the target rice is calculated based on the area 14. The state estimation method according to claim 13, wherein the specific gravity or density of the target rice is calculated based on the volume of the target rice.