Spike number prediction device

JPWO2023176196A5Pending Publication Date: 2024-11-21
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024507566
Authority / Receiving Office
JP · JP
Patent Type
Applications
Filing Date
2024-08-22
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing methods for predicting the number of ears of grain, such as wheat, using image recognition technology struggle with accuracy due to hidden or densely packed ears behind leaves, which are difficult to read from images.

Method used

A panicle number prediction device that employs image recognition to count ears within a two-dimensional unit area, followed by statistical correction using correlation information between detected and true values, and relative size relationships to enhance prediction accuracy.

Benefits of technology

The device improves prediction accuracy by compensating for unreadable information, effectively counting hidden or crowded ears, and provides a more accurate estimation of the number of ears in a wheat cultivation field.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A spike number prediction device (10) includes: a spike number counting unit (14A) and the like that count, from data of an image capturing a two-dimensional unit region (frame F) disposed in a wheat cultivation field (30), the number of spikes within the frame F on the basis of an image recognition technology; a correction unit (14B) that corrects a spike number count value obtained by counting, on the basis of correlation information between a spike number count value obtained in advance by statistical processing and a spike number true value; and a spike number prediction unit (14C) that predicts a wheat spike number in a target range on the basis of a relative size relation between the predetermined target range and the frame F in the wheat cultivation field and a corrected spike number value after correction.
Need to check novelty before this filing date? Find Prior Art

Description

Ear number prediction device

[0001] The present disclosure relates to an ear number prediction device that predicts the number of grain ears. Note that "grain" refers to agricultural crops whose seeds are consumed by humans, such as wheat, rice, millet, and barnyard millet. An "ear of grain" refers to a cluster of flowers or fruits at the tip of a grain stalk, and when counting the number of grain ears, a cluster of flowers or fruits at the tip of a grain stalk is counted as one.

[0002] In wheat cultivation, top dressing is essential to provide the necessary nutrients as the wheat grows, and determining the appropriate amount of top dressing requires knowing the number of wheat ears. However, actually counting the number of wheat ears manually is an enormous amount of work. For this reason, it is conceivable to use image recognition technology using artificial intelligence (AI) to recognize wheat ears from image data of wheat cultivation fields and predict the number of wheat ears. Similar circumstances also exist when predicting the number of ears of other grains.

[0003] Japanese Patent Application Laid-Open No. 2020-087286

[0004] Patent Literature 1 describes an example of counting the number of small objects (e.g., crop stalks) using image recognition technology that utilizes artificial intelligence. However, in order to accurately predict the number of small objects from the count value obtained by the above-described counting process, it is desirable to make the prediction by supplementing information that cannot be read or is difficult to read from the image, such as the number of small objects hidden behind the leaves of the crops and the density of the crops. However, Patent Literature 1 does not take such circumstances into consideration, and there is room for improvement in terms of improving prediction accuracy.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to predict the number of panicles more accurately by supplementing information that cannot be read or is difficult to read from an image in the panicle number prediction process.

[0006] The ear number prediction device according to the present disclosure is an ear number prediction device that predicts the number of grain ears, and includes an ear number counting unit that counts the number of grain ears in a two-dimensional unit area set in a grain cultivation field based on image recognition technology from image data of the grain reflected in the two-dimensional unit area; a correction unit that corrects the ear number count value obtained by counting by the ear number counting unit based on correlation information between the ear number count value and the true ear number value obtained in advance by statistical processing; and an ear number prediction unit that predicts the number of grain ears in a predetermined target range in the grain cultivation field based on the relative size relationship between the two-dimensional unit area and the ear number correction value corrected by the correction unit.

[0007] In the ear number prediction device, the ear number counting unit counts the number of grain ears in a two-dimensional unit area based on image data of the grains captured in the two-dimensional unit area using image recognition technology, and the correction unit corrects the ear number count value obtained by counting based on correlation information between the ear number count value and the true ear number value obtained in advance through statistical processing. As an example, the correction unit may correct the ear number count value by applying the ear number count value obtained by counting to a "relationship equation between ear number count value and true ear number value" previously generated from a graph plotting numerous combinations of true ear number values ​​obtained by counting in a grain cultivation field and the number of detections (ear number count values) obtained by detecting and counting grain ears from image data of the same cultivation field. The ear number prediction unit then predicts the number of grain ears in the target area based on the relative size relationship between the target area and the two-dimensional unit area and the corrected ear number correction value. As described above, the number of grain ears is predicted after correcting the ear count value, so information that cannot be read or is difficult to read from the image, such as grain ears hidden behind leaves or grain ears that are densely packed, can be supplemented, making it possible to predict the number of grain ears more accurately.

[0008] According to the present disclosure, the number of spikes can be predicted with higher accuracy by supplementing information that cannot be read or is difficult to read from an image.

[0009] 1 is a functional block diagram of an ear number prediction device and related equipment according to an embodiment of the present invention; 2 is a flow diagram showing processing related to ear number prediction; 3 is a diagram for explaining processing up to ear detection; 4 (a) is a diagram for explaining a case in which a rectangular area set in a wheat cultivation field, the directions of each side of which are predetermined and which is identified from the coordinates of both ends of the diagonal, is used as a two-dimensional unit area; 5 (b) is a diagram for explaining a case in which a region is used as a two-dimensional unit area by arranging multiple rectangles, the length and width of which are known, along two predetermined axis directions, based on the coordinates of a single point set in the grain cultivation field; 6 is a diagram for explaining an example in which a correction formula is generated to represent the degree of discrepancy between the true value (actual number) of ear numbers obtained by counting in a wheat cultivation field and the detected number (ear number count value) obtained by counting from image data of the corresponding wheat cultivation field, and the correction formula is applied to the detection results of artificial intelligence; 7 is a diagram showing the correlation between the true value (actual number) of ear numbers obtained by counting in a wheat cultivation field and the row width of the corresponding wheat cultivation field; 8 is a diagram showing an example of the hardware configuration of a ear number prediction device;

[0010] Hereinafter, an embodiment of the panicle number predicting device according to the present disclosure will be described with reference to the drawings.

[0011] The ear number prediction device 10 according to one embodiment shown in Figure 1 acquires image data of an area image captured by a camera 20 of an area including a rectangular frame F set in a wheat cultivation field 30 as an example of a two-dimensional unit area set in a grain cultivation field, and predicts the number of grain ears (here, the number of wheat ears) in a target range (for example, the entire wheat cultivation field 30) from the image data.

[0012] The spike number prediction device 10 includes, as functional blocks for realizing the functions according to the present disclosure, a corner detection unit 11, an exclusion unit 12, an object detection unit 13, and a prediction unit 14. The functions of each unit will be outlined below.

[0013] The corner detection unit 11 is a functional unit that acquires image data of an area image captured by a camera 20 of an area including a rectangular frame F set in the wheat cultivation field 30, divides the area image into a grid by image processing of the acquired image data, and detects the corners of the frame F in the obtained first divided image using existing image recognition technology. By detecting the corners of the frame F here, position coordinate information for each of the four corners of the frame F in a two-dimensional coordinate system set in the image data of the area image is obtained. In this embodiment, an example is shown in which an actual rectangular frame F, rather than a virtual frame, is used as the two-dimensional unit area set in the cultivation field, but other variations will be described later using Figures 4(a) and (b).

[0014] The exclusion unit 12 is a functional unit that obtains the position of the frame F in the wheat cultivation field 30 from the positions of the corners detected by the corner detection unit 11 (i.e., the position coordinates of each of the four corners of the frame F in the two-dimensional coordinate system set in the image data of the area image), and excludes image data of the area outside the frame F from the image data of the first divided image.

[0015] The object detection unit 13 is a functional unit that divides an image based on the image data after elimination by the exclusion unit 12 into grids with a division size corresponding to the size of the object (ear of wheat) and detects the object (ear of wheat) from the resulting second divided image. Here, "grid division with a division size corresponding to the size of the object (ear of wheat)" refers to dividing the image into grids to obtain a "second divided image" in which the proportion of pixels occupied by the object is appropriately increased while avoiding situations in which the features of the object become unclear or disappear. For example, the grid division is performed with a division size corresponding to the size of the object (ear of grain) that is obtained empirically or experimentally so that the object can be detected without its features becoming unclear or disappearing. In other words, the image based on the image data after elimination by the exclusion unit 12 becomes the entire image, and this entire image is again divided into grids to obtain a "second divided image" in which the proportion of pixels occupied by the object is appropriately increased so that the object can be detected without its features becoming unclear or disappearing. The object detection unit 13 detects the object (ear of wheat) from this second divided image.

[0016] The prediction unit 14 is a functional unit that counts the number of ears of the object (wheat ears) detected by the object detection unit 13 and predicts the number of wheat ears in a predetermined target range (here, for example, the entire wheat cultivation field 30) based on the relative size relationship between the frame F and a predetermined target range in the grain cultivation field, and the ear number count value obtained by counting. The prediction unit 14 also includes a counting unit 14A, a correction unit 14B, and an ear number prediction unit 14C. Of these, the counting unit 14A is a functional unit that counts the number of wheat ears detected by the object detection unit 13, and the correction unit 14B is a functional unit that corrects the ear number count value obtained by counting by the counting unit 14A using a method described below in order to improve the accuracy of the ear number prediction. The ear number prediction unit 14C is a functional unit that predicts the number of wheat ears in a predetermined target range based on the relative size relationship between a predetermined target range (for example, the entire wheat cultivation field 30) and the frame F, and the ear number correction value after correction by the correction unit 14B. Note that the "ear number counting unit" in the claims corresponds to the corner detection unit 11, the exclusion unit 12, the object detection unit 13, and the counting unit 14A in Figure 1.

[0017] Next, the process for predicting the number of ears will be described with reference to the flow chart in Fig. 2. As shown in Fig. 2, the process includes the process by the ear number prediction device 10 (steps S3 to S10) and the process by the operator (steps S1 and S2) that corresponds to the preparation for the process.

[0018] First, the operator sets up a frame F in the cultivation field to be counted (step S1), and takes a photograph using the camera 20 so that the entire set frame is visible (step S2). This provides image data of an area image that includes the frame F, as shown in the example image P1 in Fig. 3 .

[0019] Next, as processing by the spike number prediction device 10, the corner detection unit 11 acquires image data of an area image that depicts an area including the above-mentioned frame F (step S3), divides the above-mentioned area image into, for example, a 3x3 grid as shown in image example P2 in Figure 3, and detects the corners of frame F in the obtained first divided image based on existing image recognition technology (step S4).

[0020] The exclusion unit 12 then acquires the position of the frame F based on the detected corner positions (position coordinate data), and excludes the area outside the frame F from the image data of the first divided image based on the acquired position of the frame F (step S5). Image example P3 in Fig. 3 shows an example of the image after the exclusion of the area outside the frame F. In reality, if the frame F is slightly distorted from its rectangular shape, the exclusion process of step S5 is performed so that at least the entire area of ​​the frame F remains, although it is not possible to completely exclude the entire area outside the frame F.

[0021] Next, the object detection unit 13 divides the image based on the image data after elimination by the exclusion unit 12 into, for example, a 3 x 3 grid as shown in image example P4 in Fig. 3, and detects the object (ear of wheat) from the obtained second divided image based on existing image recognition technology (step S6). In this "grid division", for example, the grid division is performed with a division size according to the size of the object (ear of grain) that is obtained in advance empirically or experimentally, etc., so that the object can be detected without its features becoming unclear or disappearing, and a "second divided image" is obtained in which the proportion of pixels occupied by the object is appropriately increased.

[0022] It should be noted that the division into 3x3 grids is not required in both steps S4 and S6, but may be into NxM grids (N and M are both arbitrary natural numbers). Furthermore, the division into the same NxM grids (N and M are both arbitrary natural numbers) in steps S4 and S6 is not required, and the division into different grids may be required.

[0023] Next, the counting unit 14A counts the number of wheat ears detected by the object detection unit 13 (step S7), and the correction unit 14B corrects the ear count value obtained by counting by the counting unit 14A using the following method (step S8).

[0024] For example, the correction unit 14B uses statistical processing to previously determine a correction formula (true value = A × detection count + M) for the simple regression model to correct the detection count (the ear count value obtained by counting with the counting unit 14A) to approximate the true value (real number). In step S8, the ear count value is applied to the previously determined correction formula for the simple regression model to obtain a corrected ear count value that is expected to approximate the true value (real number). The correction formula (true value = A × detection count + M) for the simple regression model is generated from a graph plotting numerous combinations of true ear count values ​​(real numbers) obtained by manually counting ears in a wheat field and detection counts (ear count values) obtained by detecting and counting ears from image data of the same wheat field, as shown in FIG. 5 . FIG. 5 shows an example of a generated correction formula for the simple regression model, represented by a dashed line segment: "true value y = 0.90 × detection count (ear count value) + 110.65."

[0025] As another example, because the ridge width of sown wheat seeds generally correlates with the density of the wheat, the ridge width of sown wheat seeds may be used as the basic information for correction instead of the detection number (count value of ear count). Note that a "ridge" refers to each of multiple rows of areas formed by mounding soil in elongated, linear rows at intervals to produce grain in a cultivation field, and "ridge width" refers to the width (dimension perpendicular to the longitudinal direction) of the elongated, linear area. When ridge width is used, a correction formula for a simple regression model (true value = A1 × ridge width + B) represented by a dashed line segment may be generated from a graph, such as that shown in Figure 6, which plots numerous combinations of true values ​​(actual numbers) of ear counts obtained by manual counting in a wheat cultivation field and ridge widths (units: cm) in the same wheat cultivation field. In this case, the correction unit 14B can correct the spike count value obtained by counting by the counting unit 14A using any method, for example, a formula (such as a weighting function) determined in advance empirically or through experiments, etc., so as to bring the spike count value obtained by counting by the counting unit 14A closer to the true value obtained by applying the ridge width to the above correction formula.

[0026] As another example, the correction unit 14B may use both the number of detections (ear count value) and the ridge width to generate a correction formula for the multiple regression model (true value = A1 × ridge width + A2 × number of detections + B). That is, the correction unit 14B obtains an expected true value by applying both the ridge width and the number of detections (ear count value) to such a correction formula, and then corrects the ear count value using any method, for example, a formula (such as a weighting function) previously determined empirically or experimentally, so as to bring the ear count value closer to this true value. As yet another example, the correction unit 14B may correct the ear count value using the growth rate of wheat (e.g., average stalk thickness, average stalk length, etc.) recognized from image data of the wheat cultivation field as additional basic information.

[0027] Returning to Figure 2, next, the ear number prediction unit 14C predicts the number of ears in a predetermined target range (for example, the entire wheat cultivation field 30) based on the relative size relationship between the frame F and the ear number correction value after correction by the correction unit 14B (step S9), and outputs the predicted ear number value for the target range (step S10). For example, if the target range (for example, the entire wheat cultivation field 30) has an area N times that of the frame F, as a relative size relationship, the corrected ear number correction value is multiplied by N to obtain and output the predicted ear number value for the target range. Note that "output" here includes various forms of output, such as display output on a display, printout on a printer, and data transmission to an external device.

[0028] According to the embodiment described above, image data of the area outside the frame F is excluded from the image data of the first divided image, and the image after exclusion is further divided into grids with a division size according to the size of the object (ear of wheat). This results in a "second divided image" in which the proportion of pixels occupied by the object is appropriately increased while avoiding situations in which the features of the object become unclear or disappear. The object (ear of wheat) is detected from such a second divided image, and at this time, the detection accuracy of the object (ear of wheat) in the second divided image can be improved. The number of ears of wheat detected with accuracy is then counted, and the number of ears of wheat in the target range can be predicted from the relative size relationship between the target range and the frame F and the ear number count value, so the number of ears of wheat in the target range can be predicted with high accuracy.

[0029] Furthermore, the correction process by the correction unit 14B compensates for information that is difficult or impossible to read from the image, such as ears hidden behind leaves or crowded ears, thereby enabling more accurate prediction of the number of ears. Furthermore, when ridge width is used as the basis for correction, there is an advantage that the ear count value can be corrected accurately and easily based on ridge width information that is easily obtainable without being affected by changes in wheat color. Furthermore, in an example where correction is performed based additionally on the wheat growth level (e.g., average stalk thickness, average stalk length, etc.), particularly when image data is obtained by photographing a wheat field from an oblique angle, there is a correlation between the higher the wheat growth level, the higher the likelihood that ears will be hidden behind leaves, etc., so by making corrections that take the wheat growth level into further consideration, there is an advantage that the effect of approaching the true value is increased.

[0030] (Variations of the "Two-Dimensional Unit Area" Set in the Wheat Cultivation Field) In the above embodiment, an example of a "two-dimensional unit area" set in the wheat cultivation field was described using an actual rectangular frame F set in the wheat cultivation field. However, other examples may be used in image processing, using a two-dimensional coordinate system set in an image of the wheat cultivation field. For example, (1) a rectangular area set in the wheat cultivation field, the directions of each side of which are predetermined and which is identified from the coordinates of both ends of a diagonal, may be used, or (2) an area identified by arranging multiple rectangles, the length and width of which are known, along two predetermined axis directions, based on the coordinates of a single point set in the wheat cultivation field. An example of (1) above is a pseudo-rectangular area (white area) as shown in FIG. 4( a), which is identified from the coordinates of both ends of a diagonal of a rectangle, the directions of each side of which are predetermined (the coordinates of the two points marked with a star at the top right and bottom left). As an example of (2) above, as shown in FIG. 4(b), when an object X measuring 8 cm in height and 10 cm in width is photographed, a pseudo-rectangular area can be identified by arranging 10 objects X vertically and 10 objects X horizontally, using the coordinates of a single point marked with a star in the upper left corner as the reference point.

[0031] (Variation of grid division in step S6 of Fig. 2) In step S6 of Fig. 2, when the object detection unit 13 divides the image after excluding the external region into a grid to obtain multiple second divided images, the grid division may be performed based not on the grid dividing lines but on lines that add an excess (margin) from the grid dividing lines to the outside of the divided areas, and each second divided image may be obtained with an area extended outward by the excess. The excess is determined according to the size of the object (ear of wheat) to be detected, and an appropriate excess area that is neither too wide nor too narrow is added. In this case, the object detection unit 13 detects the object (ear of wheat) in each second divided image that has been extended outward by the excess, making it possible to detect all objects (ears of wheat) located on the grid dividing lines.

[0032] However, in the above case, the excess areas of adjacent second-division images overlap, potentially resulting in double counting in the count of the number of wheat ears. Therefore, when grid division is performed based on the lines that include the excess areas as described above, the prediction unit 14 corrects the ear count values ​​in adjacent second-division images to reduce the amount of double counting as follows. For example, among the ears detected and counted in each of the adjacent second-division images, image processing can be used to detect those detected (ears) whose two-dimensional coordinates in the two-dimensional coordinate system set in the wheat field image are the same (i.e., identical ears that have been double-counted), and the ear count value can be corrected to reduce the detected number (the excess amount due to double counting) by the amount of the excess. This avoids the adverse effects of double counting and enables the detection of all objects (ears) located on the grid division lines, contributing to improved accuracy in ear number prediction.

[0033] In the above-described embodiment, an example was shown in which the "number of wheat ears" was predicted as the "number of grain ears," but the technology disclosed herein can also be applied to predicting the number of ears of grains other than wheat, such as rice, millet, and barnyard millet, and similar effects can be achieved.

[0034] (Explanation of terms, explanation of hardware configuration (Figure 7), etc.) Note that the block diagrams used in explaining the above embodiments and variant examples show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (for example, using wires, wirelessly, etc.) and these multiple devices. A functional block may be realized by combining software with the single device or multiple devices.

[0035] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0036] For example, the ear number prediction device according to an embodiment of the present disclosure may function as a computer that performs the processing of this embodiment. Fig. 7 is a diagram showing an example of the hardware configuration of the ear number prediction device 10 according to an embodiment of the present disclosure. The ear number prediction device 10 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, etc.

[0037] In the following description, the term "apparatus" can be read as a circuit, a device, a unit, etc. The hardware configuration of the ear number prediction apparatus 10 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0038] Each function of the ear number prediction device 10 is realized by loading specified software (programs) onto hardware such as a processor 1001 and a memory 1002, causing the processor 1001 to perform calculations, control communication via a communication device 1004, and control at least one of reading and writing data in the memory 1002 and the storage 1003.

[0039] The processor 1001 controls the entire computer by running, for example, an operating system, and may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc.

[0040] The processor 1001 also reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. While the above-described various processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may be transmitted from a network via a telecommunications line.

[0041] The memory 1002 is a computer-readable recording medium and may be configured by, for example, at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a wireless communication method according to an embodiment of the present disclosure.

[0042] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database including at least one of memory 1002 and storage 1003, or other appropriate medium.

[0043] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0044] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that performs output to the outside. The input device 1005 and the output device 1006 may be integrated (e.g., a touch panel). Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses for each device.

[0045] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0046] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0047] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0048] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be transmitted to another device.

[0049] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0050] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0051] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0052] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0053] 10...ear number prediction device, 11...corner detection unit, 12...exclusion unit, 13...object detection unit, 14...prediction unit, 14A...counting unit, 14B...correction unit, 14C...ear number prediction unit, 20...camera, 30...wheat cultivation field, 1001...processor, 1002...memory, 1003...storage, 1004...communication device, 1005...input device, 1006...output device, 1007...bus.

Claims

1. A grain ear number prediction device for predicting the number of grain ears, an ear number counting unit that counts the number of ears of the grain in a two-dimensional unit area set in a grain cultivation field based on image recognition technology from image data of the grain depicted in the two-dimensional unit area; a correction unit that corrects the spike number count value obtained by counting by the spike number counting unit based on correlation information between the spike number count value and the spike number true value obtained in advance by statistical processing; a spike number prediction unit that predicts the number of spikes of the grain in the target range based on a relative size relationship between a predetermined target range in the grain cultivation field and the two-dimensional unit area and a spike number correction value corrected by the correction unit; A spike number prediction device comprising:

2. The correction unit performs correction based on a ridge width of a ridge in which the grain seeds are sown, the ridge width having a predetermined correlation with the grain density. The ear number predicting device according to claim 1 .

3. The correction unit performs correction further based on the growth degree of the grain. The ear number predicting device according to claim 1 .