CN103810522B - Counting method and device for corn ear grains - Google Patents

Abstract

The present invention provides a method and device for counting corn ear kernels, the method comprising: step S1, obtaining an estimation model of corn ear kernels; wherein the corn ear kernel estimation model is used to represent the number of corn ear kernels on one side and all the ear kernels The functional relation between numbers; Step S2, obtain the single-sided image of corn, and carry out image preprocessing to described single-sided image, obtain corresponding gray scale image; Threshold segmentation to obtain the corresponding binary image; step S4, count the connected regions with the same gray value, and obtain the number of connected regions corresponding to the gray value used to represent the corn grain; step S5, according to step S4 The number obtained in and the corn grain estimation model in step S1 determine the number of grains in all ears. By adopting the method for counting corn ear grains provided by the invention, the fast and simple counting of corn ear grains can be realized.

Description

Method and device for counting corn ear kernels

technical field

The invention relates to image processing technology, in particular to a method and device for counting corn ear kernels.

Background technique

Maize breeding is to use the heterozygosity of maize cross-pollination and genetic basis to carry out hybridization experiments on a large number of inbred lines and hybrid combinations, and finally select individual excellent hybrids for production. Among them, the seed yield rate of the ear is an important indicator for judging the quality of corn varieties.

At present, the work of counting corn ears and grains in my country is mainly manual. This method not only has a huge workload, but also has large errors in reading data, and it is impossible to re-read the read data repeatedly, and it is difficult to verify its correctness when a problem occurs. Developed countries such as Europe and the United States study and use test equipment. However, the price of this kind of instrument is too high, the volume and weight are large, and it is not easy to flow and use, so it is not suitable for promotion and use in actual production.

Therefore, it is very necessary to develop an automatic computer aid to improve the counting efficiency of corn ear kernels, improve the counting method, and speed up the examination and promotion of varieties.

Contents of the invention

The invention provides a method and device for counting corn ear grains, which can realize fast and simple counting of corn ear grains,

The invention provides a method for counting corn ear grains, the method comprising:

A corn ear grain counting method is characterized in that, comprising:

Step S1, obtaining a corn ear grain estimation model; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of corn ear grains on one side of the corn and the number of all ear grains;

Step S2, obtaining a single-sided image of corn, and performing image preprocessing on the single-sided image to obtain a corresponding grayscale image;

Step S3, performing multi-threshold segmentation on the grayscale image obtained in step S2 to obtain a corresponding binary image;

Step S4, count the connected regions with the same gray value, and obtain the number of connected regions corresponding to the gray value used to represent the corn grain;

Step S5, determine the number of kernels in all ears according to the number obtained in step S4 and the corn kernel estimation model in step S1.

Preferably, the step S1 specifically includes:

Obtain the following formula as the corn ear kernel estimation model:

y=ax+b;

Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears.

Preferably, said step S2 includes:

Step S21, performing median filtering processing on the single-sided image;

Step S22, carrying out Wallis filter processing to the image obtained in step S21;

Step S23, performing histogram enhancement processing on the image obtained in step S22 to obtain a grayscale image corresponding to the single-sided image.

Preferably, said step S3 specifically includes:

A multi-threshold segmentation method combining RBGM and OTSU is used to segment the gray image obtained in step S2 to obtain a binary image of corn kernels.

Preferably, the multi-threshold segmentation method combining RBGM and OTSU is used to perform image segmentation on the corn ear grain grayscale image and then obtain the binary image of the corn grain, which specifically includes:

Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula, and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span>$

in,

j=0,1,...,N-1; m=0,1,...

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

Among them, ω _k and μ _k represent the probability and mean difference corresponding to the k-th type of problem when the n-type problem is divided;

Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2.

The invention provides a corn ear grain counting device, comprising:

The modeling module is used to obtain an estimation model of corn ear grains; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of ear grains on one side of corn and the number of all ear grains;

An image preprocessing module, used to obtain a single-sided image of corn, and perform image preprocessing on the single-sided image to obtain a corresponding grayscale image;

An image segmentation module, configured to perform multi-threshold segmentation on the grayscale image obtained by the image preprocessing module to obtain a corresponding binary image;

The counting module is used to count the connected regions with the same gray value, and obtain the number of connected regions corresponding to the gray value used to represent the corn grain;

The calculation module determines the number of grains in the entire ear according to the number obtained in the counting module and the corn kernel estimation model obtained by the modeling module.

Preferably, the modeling module is specifically used to obtain the following formula as a corn ear grain estimation model:

y=ax+b;

Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears.

Preferably, the image preprocessing module includes:

The first processing submodule is used to perform median filtering processing on the single-sided image;

The second processing submodule is used to perform Wallis filter processing on the image obtained by the first processing submodule;

The third processing submodule is configured to perform histogram enhancement processing on the image obtained by the second processing submodule to obtain a grayscale image corresponding to the single-sided image.

Preferably, the image segmentation module is specifically used to perform image segmentation on the grayscale image obtained by the image preprocessing module by adopting a multi-threshold segmentation method combining RBGM and OTSU to obtain a binary image of corn kernels.

Preferably, when the image segmentation module is used to perform image segmentation on the grayscale image obtained by the image preprocessing module using a multi-threshold segmentation method combined with RBGM and OTSU to obtain a binary image of corn kernels, it is specifically used to perform :

Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula, and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span>$

in,

j=0,1,...,N-1; m=0,1,...

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

Among them, ω _k and μ _k represent the probability and mean difference corresponding to the k-th type of problem when the n-type problem is divided;

Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2.

By adopting the method for counting corn ear grains provided by the invention, the fast and simple counting of corn ear grains can be realized.

Description of drawings

Fig. 1 is the schematic flow sheet of a kind of corn ear grain counting method that the embodiment of the present invention provides;

Fig. 2 is the schematic flow sheet of the grayscale processing in the method for counting corn ear grains provided by the embodiment of the present invention;

Fig. 3 is the schematic flow chart of image segmentation in the method for counting corn ear grains provided by the embodiment of the present invention;

Fig. 4a is the single-sided corn image after the corn ear is marked using the corn ear grain counting method provided by the embodiment of the present invention;

Fig. 4b is a local single-sided corn image after the corn ear is marked using the corn ear grain counting method provided by the embodiment of the present invention;

Fig. 5 is a schematic structural diagram of a corn ear kernel counting device provided by an embodiment of the present invention.

detailed description

The specific implementation manners of the present invention will be further described below in conjunction with the drawings and examples. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings. The following examples serve to illustrate the present invention, but do not limit the scope of the present invention.

Embodiment one

A kind of corn ear grain counting is provided, as shown in Figure 1, the method comprises:

Step S1, obtaining a corn ear grain estimation model; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of corn ear grains on one side of the corn and the number of all ear grains;

Step S2, obtaining a single-sided image of corn, and performing image preprocessing on the single-sided image to obtain a corresponding grayscale image;

Step S3, performing multi-threshold segmentation on the grayscale image obtained in step S2 to obtain a corresponding binary image;

Step S4, count the connected regions with the same gray value, and obtain the number of connected regions corresponding to the gray value used to represent the corn grain;

Step S5, determine the number of kernels in all ears according to the number obtained in step S4 and the corn kernel estimation model in step S1.

By adopting the method for counting corn ear grains provided by the invention, the fast and simple counting of corn ear grains can be realized.

In the process of realizing the present invention, the inventors of the present application found that: (1) the number of rows of corn ears and kernels is generally fixed, usually 14 or 16 rows; (2) when sampling corn images, the corn kernels are The visible part of the image occupies a fixed area of corn kernels. On this basis, in a preferred embodiment of the present invention, the step S1 can be specifically: obtain the following formula (1) as a corn ear kernel estimation model:

y=ax+b; (1)

Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears.

Here, a and b can be obtained through training, that is, by manually counting some corn ears and assigning values to a and b. It should be pointed out that the training model established here can be used for multiple times in steps S2-S5, so as to facilitate the rapid counting of other corn ears.

Since step S4 is the processing result of the single-sided image of corn, the obtained grain is also one-sided, which is equivalent to x in the above relational expression. At this time, x is brought into the above-mentioned relational expression to obtain corn The total number of kernels in the ear.

Of course, in practical applications, those skilled in the art can also use other mathematical models, and the specific model used does not affect the protection scope of the present invention.

Preferably, the method for performing image preprocessing on the single-sided image in the above step S2 may include as shown in Figure 2:

Step S21, performing median filtering processing on the single-sided image;

Step S22, carrying out Wallis filter processing to the image obtained in step S21;

Step S23, performing histogram enhancement processing on the image obtained in step S22 to obtain a grayscale image corresponding to the single-sided image.

In practical applications, before step S21, a grayscale image f(x, y) of corn ear grains is also obtained, where f(x, y) is the grayscale value corresponding to the (x, y) pixel.

Wherein, the formula used in the median filter processing in step S21 is the following formula (2):

${\mathrm{<span\; class="notranslate">\; f</span>}}^{<span\; class="notranslate">\; \text{'}</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; [</span>\underset{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Element;</span>\mathrm{<span\; class="notranslate">\; S</span>}}{\mathrm{<span\; class="notranslate">\; Sort</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>}_{\frac{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Where S is the neighborhood set of pixel (x ₀ , y ₀ ), which contains (x ₀ , y ₀ ), (x, y) represents any element in S, and f(x, y) represents (x, y) point The gray value of , |S| indicates the number of elements in the set S, Sort indicates sorting, and ω indicates the weight added when calculating the intermediate pixel value.

The formula used in the Wallis filter processing in step S22 is the following formula (3):

$\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>\frac{\mathrm{<span\; class="notranslate">\; A</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}{\mathrm{<span\; class="notranslate">\; A\&sigma;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Where [f(i,j)] _M×N represents the original image, [g(i,j)] _M×N represents the enhanced image, Represents the local mean, m _d represents the mean value, σ _d represents the standard deviation, A represents the gain coefficient, and α is a proportionality constant that controls the composition of edges and backgrounds in the enhanced image.

The formula used in the histogram enhancement processing in step S23 is the following formula (4)

z=G ^-1 [T(r)] (4)

r represents the pixel gray level before image enhancement, the transformation function s=T(r) represents the pixel gray value after enhancement, and the inverse transformation function z=G ⁻¹ (s) represents the desired new gray level.

Preferably, said step S3 specifically includes:

A multi-threshold segmentation method combining RBGM and OTSU is used to segment the gray image obtained in step S2 to obtain a binary image of corn kernels.

Specifically, step S3 may include various steps as shown in Figure 3:

Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula (5), and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

in,

j=0,1,...,N-1; m=0,1,...

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

Among them, ω _k and μ _k respectively represent the probability and mean difference corresponding to the k-th type of problem when n-type problems are divided.

Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2.

The principle of the above step S31 is:

Assume that for the n-class segmentation problem, the total variance is marked as σ ² , and the probability, mean and variance of each class are marked as ω _k , μ _k , (k=0,1,...,n-1), the combination of each segmentation threshold is marked as t={t _k |k=1,2,...,n-1}, then:

where k=0,1,....,n-1 (6)

t ₀ ＝0, t _n ＝L is the histogram boundary (7)

Where 1<t _k <L(k=1,2,...,n-1) is each segmentation threshold, (8)

Then the intra-class variance of multiple thresholds is:

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; w</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Similar to the single-threshold problem, the constraint equation for the multi-threshold problem is established:

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&mu;</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; i</span>}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

According to the above formula, it can be concluded that ${\mathrm{\&sigma;}}^2-{\mathrm{\&sigma;}}_w^2=\underset{k=0}{\overset{\mathrm{no}-1}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_k{\mathrm{\&mu;}}_k^2+\underset{k=0}{\overset{\mathrm{no}-1}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_k{\mathrm{\&mu;}}_k\underset{j\&NotEqual;k}{\mathrm{\&Sigma;}}{\mathrm{\&omega;}}_j{\mathrm{\&mu;}}_j;$

After derivation, it can be concluded that ${\mathrm{\&sigma;}}_B^2={\mathrm{\&sigma;}}^2-{\mathrm{\&sigma;}}_w^2=\underset{k=0}{\overset{\mathrm{no}-1}{\mathrm{\&Sigma;}}}\underset{j=k+1}{\overset{\mathrm{no}-1}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_k{\mathrm{\&omega;}}_j{({\mathrm{\&mu;}}_k-{\mathrm{\&mu;}}_j)}^2---\left(10\right);$ obviously is a function relative to the independent variable t _i .

Then replace _{xi in gradient descent method (RBGM) formula (11) with t i ,} and replace f(x) with The descendant solution formula (12) can be obtained:

in,

j=0,1,...,N-1; m=0,1,....

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 12</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

The multi-threshold segmentation method combined with RBGM and OTSU aims to deal with the problem that in the traditional OSTU multi-threshold segmentation method, as the number of segmentation thresholds increases from single threshold to multi-threshold, the amount of calculation for the optimal threshold will increase geometrically. The method can greatly reduce the calculation amount of segmentation under the same precision, and then shorten the calculation time of segmentation.

The binary image of corn ear grains that can be obtained after step S3, wherein each corn grain is divided into a connected area, at this time, the connected areas with the same gray value are counted to obtain the The number of connected regions corresponding to the gray value of the grain, and then calculate the total number of corn ear grains according to the pre-established corn ear grain estimation model. As shown in Figure 4(a)(b), each corn kernel in the binary image of corn ear kernels is divided into a connected area, that is, adjacent pixels in this area have the same gray value 0 or 1. Before marking, initialize a counter C with an initial value of 1, which is used to count the number of corn kernels on one side, and p(x, y) represents the pixel value of the pixel at point (x, y). Starting from any initial point of the image (x ₀ , y ₀ ), scan each pixel line by line. When p(x, y) is not 1, it means that it is the image background or the identified pixel skips this pixel; when When encountering a pixel where p(x,y) is 1, identify the pixel as the counter C value, and scan all pixels adjacent to the pixel with a value of 1 and identify its value as C, and add 1 to the C value after the identification is completed , continue to the next pixel identification. Identify the pixels one by one until all the pixels have been scanned, and finally get the number of corn kernels on one side.

The beneficial effects of the present invention include: (1) Using computer graphic processing technology to replace the manual counting method of corn ear grains, the total number of corn ear grains can be quickly, easily and accurately obtained, which overcomes the time required for counting corn ear grains in the prior art scheme long, subjectivity, large error in reading data, etc.; (2) The present invention uses the collected images of some single-sided corn cobs as samples, and compares the number of single-sided corn cobs in the image with the actual total number of corn cobs. Regression analysis obtains the corn ear grain estimation model y=ax+b, and according to the corn seed estimation model, the total number of corn ear grains can be estimated from the number of corn grains on one side in the image, and the accuracy rate reaches 93%; (3) the present invention adopts The multi-threshold segmentation method combining RBGM and OTSU aims to deal with the problem that in the traditional OSTU multi-threshold segmentation method, as the number of segmentation thresholds increases from single threshold to multi-threshold, the amount of calculation for the optimal threshold will increase geometrically. This method is adopted. With the same accuracy, the amount of segmentation calculation can be greatly reduced, thereby shortening the segmentation calculation time.

Embodiment two

Based on the same design, the present invention also provides a corn ear grain counting device, as shown in Figure 5, the device includes:

The modeling module 501 is used to obtain an estimation model of corn ear grains; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of ear grains on one side of corn and the number of all ear grains;

Image preprocessing module 502, is used for obtaining the single-sided image of corn, and carries out image preprocessing to described single-sided image, obtains corresponding grayscale image;

An image segmentation module 503, configured to perform multi-threshold segmentation on the grayscale image obtained by the image preprocessing module 502, to obtain a corresponding binary image;

The counting module 504 is used to count the connected regions with the same gray value, and obtain the number of connected regions corresponding to the gray value used to represent the corn grain;

The calculation module 505 determines the number of grains in all ears according to the number obtained in the counting module 504 and the corn kernel estimation model obtained by the modeling module 501 .

Preferably, the modeling module 501 is specifically used to obtain the following formula as a corn ear grain estimation model:

y=ax+b;

Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears.

Preferably, the image preprocessing module 502 specifically includes:

The first processing submodule is used to perform median filtering processing on the single-sided image;

The second processing submodule is used to perform Wallis filter processing on the image obtained by the first processing submodule;

The third processing submodule is configured to perform histogram enhancement processing on the image obtained by the second processing submodule to obtain a grayscale image corresponding to the single-sided image.

Preferably, the image segmentation module 503 is specifically used to perform image segmentation on the grayscale image obtained by the image preprocessing module 502 by adopting a multi-threshold segmentation method combining RBGM and OTSU to obtain a binary image of corn kernels.

Preferably, when the segmentation module image segmentation module is used to adopt the multi-threshold segmentation method combined with RBGM and OTSU to perform image segmentation on the grayscale image obtained by the image preprocessing module 502 and then obtain the binary image of corn kernels, it is specifically used to perform:

Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula, and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold:

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; n</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span>$

in,

j=0,1,...,N-1; m=0,1,...

${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$

Among them, ω _k and μ _k represent the probability and mean difference corresponding to the k-th type of problem when the n-type problem is divided;

Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the technical principle of the present invention, some improvements and modifications can also be made. These improvements and modifications It should also be regarded as the protection scope of the present invention.

Claims (8)

1. A corn ear grain counting method is characterized in that, comprising: Step S1, obtaining a corn ear grain estimation model; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of corn ear grains on one side of the corn and the number of all ear grains; Step S2, obtaining a single-sided image of corn, and performing image preprocessing on the single-sided image to obtain a corresponding grayscale image; Step S3, performing multi-threshold segmentation on the grayscale image obtained in step S2 to obtain a corresponding binary image; Step S4, count the connected regions with the same gray value, and obtain the number of connected regions with the gray value used to represent the corn grain; Step S5, according to the number obtained in step S4 and the corn kernel estimation model in step S1, determine the number of grains of all ears; Wherein, the step S1 specifically includes: Obtain the following formula as the corn ear kernel estimation model: y=ax+b; Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears. 2. The method according to claim 1, wherein said step S2 comprises: Step S21, performing median filtering processing on the single-sided image; Step S22, carrying out Wallis filter processing to the image obtained in step S21; Step S23, performing histogram enhancement processing on the image obtained in step S22 to obtain a grayscale image corresponding to the single-sided image. 3. The method according to claim 1, wherein said step S3 specifically comprises: A multi-threshold segmentation method combining the gradient descent method RBGM and the maximum between-class variance method OTSU is used to segment the gray image obtained in step S2 to obtain a binary image of corn kernels. 4. The method according to claim 3, wherein the multi-threshold segmentation method combining RBGM and OTSU is used to perform image segmentation on the corn ear grain grayscale image and then obtain the binary image of the corn grain, specifically comprising: Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula, and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold: ${{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span>$ Among them, x _i ^mN+j+1 = x _i ^mN+j -η _m Δ _i ( _xi ^mN+j ), i=1,2,...,N ${\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{<span\; class="notranslate">\; \&part;</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right.<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ;</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span>$ ${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$ Among them, ω _k and μ _k represent the probability and mean difference corresponding to the k-th type of problem when the n-type problem is divided; Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2. 5. A corn ear grain counting device is characterized in that, comprising: The modeling module is used to obtain an estimation model of corn ear grains; wherein the corn ear grain estimation model is used to represent the functional relationship between the number of ear grains on one side of corn and the number of all ear grains; An image preprocessing module, used to obtain a single-sided image of corn, and perform image preprocessing on the single-sided image to obtain a corresponding grayscale image; An image segmentation module, configured to perform multi-threshold segmentation on the grayscale image obtained by the image preprocessing module to obtain a corresponding binary image; The counting module is used to count the connected regions with the same gray value, and obtains the number of connected regions with the gray value used to represent corn grains; Calculation module, according to the number obtained in the counting module and the corn kernel estimation model obtained by the modeling module to determine the number of grains of all ears; Wherein, the modeling module is specifically used to obtain the following formula as a corn ear grain estimation model: y=ax+b; Among them, x represents the number of corn kernels in the single-sided image of corn ears, a and b are training values, and y is the number of kernels in all ears. 6. The device according to claim 5, wherein the image preprocessing module comprises: The first processing submodule is used to perform median filtering processing on the single-sided image; The second processing submodule is used to perform Wallis filter processing on the image obtained by the first processing submodule; The third processing submodule is configured to perform histogram enhancement processing on the image obtained by the second processing submodule to obtain a grayscale image corresponding to the single-sided image. 7. The device according to claim 5, wherein the image segmentation module is specifically configured to use a multi-threshold segmentation method combining the gradient descent method RBGM and the maximum inter-class variance method OTSU to process the image preprocessing module The obtained grayscale image is segmented to obtain a binary image of corn kernels. 8. The device according to claim 7, wherein the image segmentation module is used to perform image segmentation on the grayscale image obtained by the image preprocessing module using a multi-threshold segmentation method combining RBGM and OTSU to obtain When the binary image of corn kernels is used, it is specifically used to execute: Step S31, select the initial iteration point t _i and the iteration step size η _m to iteratively solve the following multi-threshold segmentation iterative solution formula, and obtain the optimal segmentation threshold The optimal segmentation threshold To make Take the maximum segmentation threshold: ${{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\frac{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ;</span>$ Among them, x _i ^mN+j+1 = x _i ^mN+j -η _m Δ _i ( _xi ^mN+j ), i=1,2,...,N ${\mathrm{<span\; class="notranslate">\; \&Delta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{<span\; class="notranslate">\; \&part;</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; \&part;</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right.<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ;</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span>$ ${\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\underset{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ;</span>$ Among them, ω _k and μ _k represent the probability and mean difference corresponding to the k-th type of problem when the n-type problem is divided; Step S32, using the optimal segmentation threshold obtained in step S31 Perform multi-threshold segmentation on the grayscale image obtained in step S2.