Wheat yield estimation method and device, computer equipment and storage medium

Abstract

The invention discloses a wheat yield estimation method and device, computer equipment and a storage medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring a wheat ear image, preprocessing the wheat ear image by adopting an illumination normalization algorithm, and extracting illumination invariant features; inputting the illumination invariant features into an image semantic segmentation model to obtain wheat ear area sub-images; using a support vector regression algorithm to calculate the wheat ear density of each wheat ear area sub-image, and constructing a wheat ear density distribution map; extracting wheat ear morphological features from the wheat ear region sub-images; and estimating the wheat yield by using a pre-trained random forest regression model according to the wheat ear density distribution diagram and the wheat ear morphological characteristics. The invention also relates to the technical field of block chains, and the wheat ear image is stored on a block chain node. According to the method, the influence of human factors and natural environment changes on the wheat yield estimation result is remarkably reduced, and the estimation reliability and practicability are improved.

Description

Technical Field

[0001] The present application belongs to the field of artificial intelligence technology, and specifically relates to a wheat yield estimation method, device, computer equipment and storage medium. Background Art

[0002] Wheat yield estimation is an important part of agricultural production, which helps farmers and agricultural departments to rationally plan production and market strategies. At present, wheat yield estimation uses a variety of technical means, including traditional visual estimation, field sampling and bag cutting. These methods rely on experience and field observation, and have the advantages of being simple and easy to implement, but there may be certain errors.

[0003] Taking the field sampling method as an example, agricultural experts will randomly select multiple representative areas in the wheat field as sample points, and then record the wheat growth conditions in these areas in detail, such as plant height, number of ears, number of grains, etc. Through field measurement and observation, combined with past experience, experts can preliminarily estimate the average yield of the entire wheat field. However, although this method is simple and easy to implement, it is affected by human factors and changes in the natural environment, such as changes in lighting conditions, which may lead to certain errors in the estimated results.

[0004] Therefore, how to reduce the impact of human factors and natural environmental changes in the process of wheat yield estimation and improve the accuracy of yield estimation has become an urgent problem to be solved. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose a wheat yield estimation method, device, computer equipment and storage medium to reduce the impact of human factors and natural environmental changes in the wheat yield estimation process and improve the accuracy of yield estimation.

[0006] In order to solve the above technical problems, the present application embodiment provides a method for estimating wheat yield, which adopts the following technical solution:

[0007] A wheat yield estimation method, comprising:

[0008] Acquire wheat ear images, and preprocess the wheat ear images using an illumination normalization algorithm to extract illumination invariant features;

[0009] The illumination invariant features are input into the image semantic segmentation model to obtain the wheat ear region sub-image;

[0010] The wheat ear density of each wheat ear region sub-image is calculated using support vector regression algorithm to construct a wheat ear density distribution map;

[0011] Extracting wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area;

[0012] Wheat yield was estimated using a pre-trained random forest regression model based on the wheat ear density distribution map and wheat ear morphological characteristics.

[0013] Furthermore, the step of preprocessing the wheat ear image using an illumination normalization algorithm to extract illumination invariant features specifically includes:

[0014] Obtain the pixel value of each pixel point on the wheat ear image to obtain the original pixel value;

[0015] Calculate the Gaussian surround function value of each pixel on the RGB color channel;

[0016] Calculate the difference between the original pixel value and the Gaussian surround function value of each pixel to obtain the wheat ear image after illumination adjustment;

[0017] Feature extraction is performed on the wheat ear image after illumination adjustment to obtain illumination invariant features.

[0018] Furthermore, the step of calculating the Gaussian surround function value of each pixel point on the RGB color channel specifically includes:

[0019] Performing Gaussian blur processing on the wheat ear image to obtain a Gaussian blurred image of the wheat ear;

[0020] Calculate the Gaussian weight of each pixel on the Gaussian blurred image;

[0021] Determine the Gaussian kernel of the Gaussian blurred image, and determine the coverage of the Gaussian kernel for each pixel on the Gaussian blurred image;

[0022] The Gaussian surround function value is calculated based on the Gaussian weight of each pixel and the Gaussian kernel coverage of each pixel.

[0023] Furthermore, the step of using a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image and constructing a wheat ear density distribution map specifically includes:

[0024] Density features are extracted for each wheat ear region sub-image to obtain wheat ear density features;

[0025] Perform vector transformation on the wheat ear density feature to obtain the wheat ear density feature vector;

[0026] Input the wheat ear density feature vector into the pre-trained support vector regression model, and output the wheat ear density label of each wheat ear region sub-image;

[0027] The wheat ear density label of each wheat ear region sub-image is mapped to the wheat ear part image to obtain the wheat ear density distribution map.

[0028] Furthermore, the step of mapping the wheat ear density label of each wheat ear region sub-image to the wheat ear part image to obtain the wheat ear density distribution map specifically includes:

[0029] Determine the position area of ​​each wheat ear region sub-image in the wheat ear part image to obtain the sub-image position area;

[0030] Map the wheat ear density label of each wheat ear region sub-image to the corresponding sub-image position area;

[0031] In each sub-image position area, density values ​​are filled based on the corresponding wheat ear density labels to obtain a wheat ear density distribution map.

[0032] Furthermore, the steps of estimating wheat yield using the pre-trained random forest regression model according to the wheat ear density distribution map and morphological characteristics include:

[0033] Extract the characteristics of the wheat ear density distribution map to obtain the wheat ear density distribution characteristics;

[0034] Combining the wheat ear density distribution characteristics and the wheat ear morphological characteristics to generate the wheat ear density-morphological combination characteristics;

[0035] Based on the wheat ear density-morphology combination characteristics, the wheat yield was estimated using a pre-trained random forest regression model.

[0036] Furthermore, the step of combining the wheat ear density distribution feature and the wheat ear morphological feature to generate the wheat ear density-morphological combination feature specifically includes:

[0037] Performing feature interaction processing on the wheat ear density distribution feature and the wheat ear morphological feature to obtain the wheat ear interaction feature, wherein the feature interaction processing includes calculating feature products, calculating feature ratios, and calculating feature differences;

[0038] The wheat ear interaction characteristics are taken as the wheat ear density-morphology combined characteristics.

[0039] In order to solve the above technical problems, the present application also provides a wheat yield estimation device, which adopts the following technical solution:

[0040] A wheat yield estimation device, comprising:

[0041] An illumination normalization module is used to obtain wheat ear images, and pre-process the wheat ear images using an illumination normalization algorithm to extract illumination invariant features;

[0042] An image semantic segmentation module is used to input the illumination invariant features into the image semantic segmentation model to obtain a wheat ear region sub-image;

[0043] A wheat ear density calculation module is used to calculate the wheat ear density of each wheat ear region sub-image using a support vector regression algorithm to construct a wheat ear density distribution map;

[0044] A morphological feature extraction module is used to extract wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area;

[0045] The wheat yield estimation module is used to estimate wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and wheat ear morphological characteristics.

[0046] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0047] A computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the wheat yield estimation method as described in any one of the above items are implemented.

[0048] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0049] A computer-readable storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement the steps of the wheat yield estimation method as described in any one of the above.

[0050] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0051] The present application discloses a wheat yield estimation method, device, computer equipment and storage medium, which belongs to the field of artificial intelligence technology. The present application obtains wheat ear images, and pre-processes the wheat ear images using an illumination normalization algorithm to extract illumination-invariant features; inputs the illumination-invariant features into an image semantic segmentation model to obtain wheat ear region sub-images; uses a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and constructs a wheat ear density distribution map; extracts wheat ear morphological features from the wheat ear region sub-images, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area; and estimates wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological features. The present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features and ensuring the stability and accuracy of feature extraction. Then, the image semantic segmentation technology is combined to accurately locate the wheat ear area, and further calculate the wheat ear density and extract the wheat ear morphological features. Support vector regression and random forest regression models are used to integrate the wheat ear density distribution and morphological features to achieve high-precision prediction of wheat yield, significantly reduce the impact of human factors and natural environment changes on the wheat yield prediction results, and improve the reliability and practicality of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 An exemplary system architecture diagram is shown in which the present application can be applied;

[0054] Figure 2 A flow chart showing an embodiment of a wheat yield estimation method according to the present application is shown;

[0055] Figure 3 A schematic structural diagram of an embodiment of a wheat yield estimation device according to the present application is shown;

[0056] Figure 4 A schematic structural diagram of an embodiment of a computer device according to the present application is shown. DETAILED DESCRIPTION

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0058] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0059] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0060] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0061] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0062] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0063] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0064] It should be noted that the wheat yield estimation method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the wheat yield estimation device is generally set in the server / terminal device.

[0065] It should be understood that Figure 1 The numbers of terminal devices, networks and servers in the embodiment are merely illustrative. The above system may have any number of terminal devices, networks and servers according to implementation requirements.

[0066] Continue to refer Figure 2 , shows a flow chart of an embodiment of a method for estimating wheat yield according to the present application. The method for estimating wheat yield comprises the following steps:

[0067] S201, obtaining a wheat ear image, and preprocessing the wheat ear image using an illumination normalization algorithm to extract illumination invariant features.

[0068] Specifically, use a digital camera, smart phone or dedicated agricultural image acquisition equipment to take photos or video frames of wheat ears, and ensure that the image is clear and can clearly distinguish the details of the wheat ears, such as the number, color, shape, etc. of grains. Then, the wheat ear image is preprocessed using an illumination normalization algorithm, such as an algorithm based on color constancy theory, namely the Retinex algorithm, which is used to eliminate the problem of uneven illumination in image processing.

[0069] Through the above steps, the illumination normalization process can be effectively performed on the wheat ear image, and features that are insensitive to illumination changes, namely, illumination invariant features, can be extracted.

[0070] S202, inputting the illumination invariant features into the image semantic segmentation model to obtain a wheat ear region sub-image.

[0071] Specifically, a semantic segmentation model suitable for wheat ear region segmentation is selected, which can be FCN (Fully Convolutional Network), U-Net, DeepLab series, etc. These models automatically extract image features through convolutional neural networks (CNN) and use these features for pixel-level classification, thereby achieving semantic segmentation of the image.

[0072] The selected semantic segmentation model is trained using a dataset containing wheat ear region annotations. During the training process, the model will learn how to classify pixels in the input image into wheat ear regions and other regions. The extracted illumination invariant feature image is input into the trained semantic segmentation model. The model will classify each pixel in the image and output a segmentation map of the same size as the input image. Each pixel in the segmentation map is assigned a category label, indicating whether the pixel belongs to the wheat ear region or the non-wheat ear region.

[0073] S203, using a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and construct a wheat ear density distribution map.

[0074] Specifically, for each wheat ear region sub-image, a corresponding wheat ear density label is required in order to subsequently construct a wheat ear density distribution map. The system uses the SVR model (i.e., support vector regression model) to calculate the wheat ear density of each wheat ear region sub-image. SVR is a regression method based on statistical learning theory that minimizes the prediction error by finding an optimal hyperplane in a high-dimensional space. Use a labeled feature vector data set to train the SVR model, and adjust the parameters of the SVR model (such as kernel function type, regularization parameter C, kernel function parameter γ, etc.) to optimize model performance. During the training process, the model will learn how to map input features to corresponding wheat ear density labels. For a new wheat ear region sub-image, its feature vector is input into the trained SVR model, and the SVR model will output a wheat ear density label, that is, the wheat ear density of the wheat ear region sub-image.

[0075] S204, extracting wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area.

[0076] Specifically, edge detection algorithms (such as Canny edge detector, Sobel operator, etc.) are used to identify the edges in the wheat ear region sub-image. Edge detection helps to define the outline of the wheat ear. Based on edge detection, contour extraction algorithms (such as Hough transform, contour search, etc.) are used to find the complete outline of the wheat ear. If there are multiple outlines in the image, the most likely wheat ear outline needs to be screened out based on the contour area, shape, etc.

[0077] It can be approximated by calculating the longest straight line segment or longest path of the contour, or by traversing the contour points and finding the line segment with the largest distance between two points, or by using image morphological operations (such as skeletonization) to extract the central axis, and then calculating the length of the axis to obtain the length of the ear of wheat.

[0078] Since the ear of wheat may not be a completely regular shape, the width needs to be defined as the average width of the contour at multiple points or directions. This can be achieved by uniformly sampling points on the contour, then calculating the width of the ear of wheat in the normal direction of the contour at each point, and finally taking the average value.

[0079] The area of ​​the ear of wheat can be obtained by directly calculating the number of pixels in the inner area of ​​the contour. This can be achieved through the contour filling algorithm, and then the total number of pixels in the filled area is calculated to obtain the area of ​​the ear of wheat.

[0080] S205, estimating wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and wheat ear morphological characteristics.

[0081] Specifically, the information of the wheat ear density distribution map is integrated with the wheat ear morphological characteristics data to form a complete data set for subsequent yield estimation. Select a pre-trained random forest regression model, which should have been trained using similar wheat yield data and related features and has shown good prediction performance. Load the pre-trained random forest regression model into the computing environment and prepare for prediction. Input the integrated data set (including the wheat ear density distribution map and the wheat ear morphological characteristics) into the pre-trained random forest regression model to output the wheat yield estimation result.

[0082] In the above embodiment, the present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features, ensuring the stability and accuracy of feature extraction, and then combines image semantic segmentation technology to accurately locate the wheat ear area, and further performs wheat ear density calculation and wheat ear morphological feature extraction. Support vector regression and random forest regression models are used to integrate wheat ear density distribution and morphological features, thereby achieving high-precision prediction of wheat yield, significantly reducing the impact of human factors and natural environment changes on wheat yield prediction results, and improving the reliability and practicality of the prediction.

[0083] Furthermore, the step of preprocessing the wheat ear image using an illumination normalization algorithm to extract illumination invariant features specifically includes:

[0084] Obtain the pixel value of each pixel point on the wheat ear image to obtain the original pixel value;

[0085] Calculate the Gaussian surround function value of each pixel on the RGB color channel;

[0086] Calculate the difference between the original pixel value and the Gaussian surround function value of each pixel to obtain the wheat ear image after illumination adjustment;

[0087] Feature extraction is performed on the wheat ear image after illumination adjustment to obtain illumination invariant features.

[0088] In this embodiment, the algorithm based on the color constancy theory (Retinex algorithm) can simulate the adaptability of the human eye to changes in illumination, thereby eliminating the problem of uneven illumination in the image.

[0089] Specifically, first, the pixel value of each pixel on the wheat ear image is obtained to obtain the original pixel value. Then, the Gaussian surround function value of each pixel on the three color channels is calculated. The Gaussian surround function value can be regarded as the weighted average of the area around each pixel in the image. Finally, the corresponding Gaussian surround value is subtracted from the original pixel value to obtain the pixel value after illumination adjustment. This step simulates the adaptability of the human eye to illumination and removes the uneven illumination in the image.

[0090] In digital image processing, images are usually represented as a collection of pixels, each of which contains certain color information. For color images, this color information is usually represented by the RGB color model. The RGB color model is an additive color model that generates various colors by combining different intensities of the three basic colors red, green, and blue.

[0091] Since the Retinex algorithm may change the color balance of the image while removing illumination, color restoration may be required, which is usually achieved by introducing a color restoration factor (CRF), which is related to the average brightness and contrast of the image.

[0092] In the above embodiment, the Retinex algorithm can effectively perform illumination normalization processing on the wheat ear image and extract features that are insensitive to illumination changes, thereby providing a reliable image analysis basis for the subsequent wheat yield estimation task.

[0093] Furthermore, the step of calculating the Gaussian surround function value of each pixel point on the RGB color channel specifically includes:

[0094] Performing Gaussian blur processing on the wheat ear image to obtain a Gaussian blurred image of the wheat ear;

[0095] Calculate the Gaussian weight of each pixel on the Gaussian blurred image;

[0096] Determine the Gaussian kernel of the Gaussian blurred image, and determine the coverage of the Gaussian kernel for each pixel on the Gaussian blurred image;

[0097] The Gaussian surround function value is calculated based on the Gaussian weight of each pixel and the Gaussian kernel coverage of each pixel.

[0098] In this embodiment, when calculating the Gaussian surround function value of each pixel on the three color channels (red R, green G, blue B), the image is actually subjected to Gaussian blur processing, but the purpose is to estimate the lighting conditions around each pixel, not just to blur the image. The Gaussian surround function is the core of the Gaussian blur algorithm, which calculates the new pixel value by considering the pixel values ​​around the pixel and assigning different weights according to their distance from the central pixel.

[0099] Specifically, first, the wheat ear image is Gaussian blurred to obtain a Gaussian blurred image of the wheat ear. Then, based on the calculation of the Gaussian weight of each pixel on the Gaussian blurred image, the weight of each pixel in the Gaussian kernel can be calculated using the Gaussian function calculation formula.

[0100] Then, determine the Gaussian kernel of the Gaussian blurred image. The Gaussian kernel is a two-dimensional array (matrix) whose element values ​​are calculated by the Gaussian function. The shape of the Gaussian function is a bell-shaped curve, with the largest value in the center and gradually decreasing toward the surroundings. The size of the kernel (such as 3x3, 5x5, etc.) and the standard deviation (σ) determine the degree and range of the blur. The larger the standard deviation, the more obvious the blur effect, and the greater the influence of the surrounding pixels on the central pixel.

[0101] Next, for each pixel on the Gaussian blurred image, determine the coverage of the Gaussian kernel. Then, multiply each weight in the Gaussian kernel by the pixel value at the corresponding position, and add all the products together. The result is the new value of the pixel after being processed by the Gaussian surround function. This process requires performing a pixel value weighting operation on each pixel of the wheat ear image.

[0102] Since the image is in color and has three color channels (R, G, B), the above Gaussian surround function calculation needs to be performed on each channel separately. That is, the same Gaussian kernel and calculation steps are applied to the R, G, and B values ​​of each pixel to obtain three new color values.

[0103] In the above embodiment, illumination normalization is performed based on the Retinex algorithm, and the color value of each pixel is updated to the weighted average of the surrounding pixels, thereby achieving illumination smoothing. This can be used for illumination normalization in image preprocessing to reduce the impact of uneven illumination on subsequent image processing steps.

[0104] Furthermore, the step of using a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image and constructing a wheat ear density distribution map specifically includes:

[0105] Density features are extracted for each wheat ear region sub-image to obtain wheat ear density features;

[0106] Perform vector transformation on the wheat ear density feature to obtain the wheat ear density feature vector;

[0107] Input the wheat ear density feature vector into the pre-trained support vector regression model, and output the wheat ear density label of each wheat ear region sub-image;

[0108] The wheat ear density label of each wheat ear region sub-image is mapped to the wheat ear part image to obtain the wheat ear density distribution map.

[0109] In this embodiment, first, for each wheat ear region sub-image, features that can reflect the density of wheat ears are extracted. These features can be pixel-based (such as color histogram, texture features, etc.) or shape-based (such as the compactness of wheat ears, area proportion, etc.). Then, the extracted features are converted into numerical vectors to serve as inputs to the SVR model. For a new wheat ear region sub-image, its feature vector is input into the trained SVR model, and the SVR model will output a wheat ear density label, that is, the wheat ear density of the wheat ear region sub-image.

[0110] The wheat ear density distribution map is a chart generated by image processing and machine learning technology, which shows the spatial distribution and density of wheat ears in a wheat field. Specifically, this distribution map can reflect the number of wheat ears per unit area and their distribution characteristics, such as dense areas and sparse areas.

[0111] The predicted density value of each wheat ear region sub-image is mapped back to its position in the original image, and the coordinates of the sub-image are converted to the coordinates of the original image to determine the sub-image position. Then, the wheat ear density label of each wheat ear region sub-image is mapped to the wheat ear image, and the density value is filled to obtain the wheat ear density distribution map.

[0112] Through the above steps, the support vector regression algorithm is used to calculate the wheat ear density of each wheat ear area sub-image, and a wheat ear density distribution map is constructed to facilitate the evaluation of wheat ear distribution and assist yield prediction and management.

[0113] Furthermore, the step of mapping the wheat ear density label of each wheat ear region sub-image to the wheat ear part image to obtain the wheat ear density distribution map specifically includes:

[0114] Determine the position area of ​​each wheat ear region sub-image in the wheat ear part image to obtain the sub-image position area;

[0115] Map the wheat ear density label of each wheat ear region sub-image to the corresponding sub-image position area;

[0116] In each sub-image position area, density values ​​are filled based on the corresponding wheat ear density labels to obtain a wheat ear density distribution map.

[0117] In this embodiment, the position area of ​​each wheat ear region sub-image in the wheat ear image is determined to obtain the sub-image position area, ensuring that each wheat ear region sub-image and its position information in the wheat ear image are accurately corresponding. Then, the density label of the sub-image is mapped back to the corresponding position in the wheat ear image, and the corresponding density value is filled in the corresponding position of the wheat ear density distribution map according to the coordinates and density values ​​obtained in the mapping process, which can be achieved by setting pixel values ​​or drawing color blocks. In order to improve the visual effect and readability of the distribution map, the distribution map can be smoothed, which can be achieved by an image filtering algorithm (such as Gaussian blur) to reduce noise and sawtooth effects.

[0118] Through the above steps, the wheat ear density label of each wheat ear region sub-image can be accurately mapped to the wheat ear image, and an intuitive wheat ear density distribution map can be constructed.

[0119] Furthermore, the steps of estimating wheat yield using the pre-trained random forest regression model according to the wheat ear density distribution map and morphological characteristics include:

[0120] Extract the characteristics of the wheat ear density distribution map to obtain the wheat ear density distribution characteristics;

[0121] Combining the wheat ear density distribution characteristics and the wheat ear morphological characteristics to generate the wheat ear density-morphological combination characteristics;

[0122] Based on the wheat ear density-morphology combination characteristics, the wheat yield was estimated using a pre-trained random forest regression model.

[0123] In this embodiment, key features are extracted from the wheat ear density distribution map. These wheat ear density distribution features include the wheat ear density in different areas, the uniformity of density distribution, the area proportion of high-density areas, etc., which can quantitatively describe the distribution of wheat ears in the field. After obtaining the wheat ear density distribution features, the wheat ear density distribution features are combined with the wheat ear morphological features (such as size, shape, etc.) to generate wheat ear density-morphology combination features. This combination can provide more comprehensive information and help the model better understand the growth status and yield potential of wheat ears. Using the pre-trained random forest regression model, wheat yield is estimated based on the wheat ear density-morphology combination features. The random forest model is an integrated learning method that predicts the results by constructing multiple decision trees and outputting the average value or mode, with high prediction accuracy and robustness.

[0124] Through the above steps, by combining image processing, feature engineering and machine learning techniques, multiple decision trees are constructed and the average or mode is output to predict the results, which has high prediction accuracy and robustness.

[0125] Furthermore, the step of combining the wheat ear density distribution feature and the wheat ear morphological feature to generate the wheat ear density-morphological combination feature specifically includes:

[0126] Performing feature interaction processing on the wheat ear density distribution feature and the wheat ear morphological feature to obtain the wheat ear interaction feature, wherein the feature interaction processing includes calculating feature products, calculating feature ratios, and calculating feature differences;

[0127] The wheat ear interaction characteristics are taken as the wheat ear density-morphology combined characteristics.

[0128] In this embodiment, feature interaction refers to the relationship or interaction between different features. In agricultural production, there is a complex interaction between the density distribution and morphological characteristics of wheat ears. Through feature interaction processing, these potential, nonlinear relationships can be captured, thereby providing richer information for the model and improving the model's prediction ability.

[0129] Feature interaction processing includes calculating feature products, calculating feature ratios, and calculating feature differences. Feature product is a simple feature interaction method that creates a new feature by calculating the product of two features. This feature interaction can capture the multiplication effect between the two features. Feature ratio creates a new feature by calculating the ratio between two features. This feature interaction can reflect the relative importance or proportional relationship between the two features. Feature difference creates a new feature by calculating the difference between two features. This feature interaction can capture the difference or change between the two features.

[0130] The wheat ear interaction features obtained through the feature interaction processing are used as the wheat ear density-morphology combination features, and these interaction features are used in the subsequent model prediction. These combination features not only contain the information of the original density distribution characteristics and morphological characteristics, but also contain the information of the interaction between them, so they can more comprehensively describe the growth status and yield potential of wheat ears.

[0131] In the above embodiment, the present application discloses a wheat yield estimation method, which belongs to the field of artificial intelligence technology. The present application obtains a wheat ear image, and pre-processes the wheat ear image using an illumination normalization algorithm to extract illumination invariant features; inputs the illumination invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image; uses a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and constructs a wheat ear density distribution map; extracts wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width, and wheat ear area; and estimates wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological features. The present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features and ensuring the stability and accuracy of feature extraction. Then, the image semantic segmentation technology is combined to accurately locate the wheat ear area, and further calculate the wheat ear density and extract the wheat ear morphological features. Support vector regression and random forest regression models are used to integrate the wheat ear density distribution and morphological features to achieve high-precision prediction of wheat yield, significantly reduce the impact of human factors and natural environment changes on the wheat yield prediction results, and improve the reliability and practicality of the prediction.

[0132] In this embodiment, the wheat yield estimation method is executed on an electronic device (eg Figure 1 The server shown in the figure) can receive instructions or obtain data through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0133] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned wheat ear image information, the above-mentioned wheat ear image information can also be stored in a blockchain node.

[0134] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0135] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0136] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0137] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0138] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0139] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a wheat yield estimation device, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0140] like Figure 3 As shown, the wheat yield estimation device 300 described in this embodiment includes:

[0141] The illumination normalization module 301 is used to obtain the wheat ear image, and pre-process the wheat ear image using an illumination normalization algorithm to extract illumination invariant features;

[0142] Image semantic segmentation module 302, used for inputting illumination invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image;

[0143] The wheat ear density calculation module 303 is used to calculate the wheat ear density of each wheat ear region sub-image using a support vector regression algorithm to construct a wheat ear density distribution map;

[0144] A morphological feature extraction module 304 is used to extract wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area;

[0145] The wheat yield estimation module 305 is used to estimate the wheat yield using a pre-trained random forest regression model according to the wheat ear density distribution map and the wheat ear morphological characteristics.

[0146] Furthermore, the illumination normalization module 301 is specifically used for:

[0147] The pixel value of each pixel on the wheat ear image is obtained to obtain the original pixel value; based on the Gaussian surround function value of each pixel on the RGB color channel; the difference between the original pixel value and the Gaussian surround function value of each pixel is calculated to obtain the wheat ear image after illumination adjustment; the feature extraction of the wheat ear image after illumination adjustment is performed to obtain the illumination invariant feature.

[0148] Furthermore, the illumination normalization module 301 is further used for:

[0149] Performing Gaussian blur processing on the wheat ear image to obtain a Gaussian blurred image of the wheat ear;

[0150] Based on calculating the Gaussian weight of each pixel on the Gaussian blurred image;

[0151] Determine the Gaussian kernel of the Gaussian blurred image, and determine the coverage of the Gaussian kernel for each pixel on the Gaussian blurred image;

[0152] The Gaussian surround function value is calculated based on the Gaussian weight of each pixel and the Gaussian kernel coverage of each pixel.

[0153] Furthermore, the wheat ear density calculation module 303 is specifically used for:

[0154] Density features are extracted for each wheat ear region sub-image to obtain wheat ear density features;

[0155] Perform vector transformation on the wheat ear density feature to obtain the wheat ear density feature vector;

[0156] Input the wheat ear density feature vector into the pre-trained support vector regression model, and output the wheat ear density label of each wheat ear region sub-image;

[0157] The wheat ear density label of each wheat ear region sub-image is mapped to the wheat ear part image to obtain the wheat ear density distribution map.

[0158] Furthermore, the wheat ear density calculation module 303 is further used for:

[0159] Determine the position area of ​​each wheat ear region sub-image in the wheat ear part image to obtain the sub-image position area;

[0160] Map the wheat ear density label of each wheat ear region sub-image to the corresponding sub-image position area;

[0161] In each sub-image position area, density values ​​are filled based on the corresponding wheat ear density labels to obtain a wheat ear density distribution map.

[0162] Furthermore, the wheat yield estimation module 305 is specifically used for:

[0163] Extract the characteristics of the wheat ear density distribution map to obtain the wheat ear density distribution characteristics;

[0164] Combining the wheat ear density distribution characteristics and the wheat ear morphological characteristics to generate the wheat ear density-morphological combination characteristics;

[0165] Based on the wheat ear density-morphology combination characteristics, the wheat yield was estimated using a pre-trained random forest regression model.

[0166] Furthermore, the wheat yield estimation module 305 is further used for:

[0167] Performing feature interaction processing on the wheat ear density distribution feature and the wheat ear morphological feature to obtain the wheat ear interaction feature, wherein the feature interaction processing includes calculating feature products, calculating feature ratios, and calculating feature differences;

[0168] The wheat ear interaction characteristics are taken as the wheat ear density-morphology combined characteristics.

[0169] In the above embodiment, the present application discloses a wheat yield estimation device, which belongs to the field of artificial intelligence technology. The present application obtains a wheat ear image, and pre-processes the wheat ear image using an illumination normalization algorithm to extract illumination invariant features; inputs the illumination invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image; uses a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and constructs a wheat ear density distribution map; extracts wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width, and wheat ear area; and estimates wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological features. The present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features and ensuring the stability and accuracy of feature extraction. Then, the image semantic segmentation technology is combined to accurately locate the wheat ear area, and further calculate the wheat ear density and extract the wheat ear morphological features. Support vector regression and random forest regression models are used to integrate the wheat ear density distribution and morphological features to achieve high-precision prediction of wheat yield, significantly reduce the impact of human factors and natural environment changes on the wheat yield prediction results, and improve the reliability and practicality of the prediction.

[0170] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0171] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 4 with a memory 41, a processor 42, and a network interface 43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0172] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0173] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the wheat yield estimation method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0174] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the wheat yield estimation method.

[0175] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0176] In the above embodiment, the present application discloses a computer device, which belongs to the field of artificial intelligence technology. The present application obtains a wheat ear image, and pre-processes the wheat ear image using an illumination normalization algorithm to extract illumination invariant features; inputs the illumination invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image; uses a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and constructs a wheat ear density distribution map; extracts wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width, and wheat ear area; and estimates wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological features. The present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features and ensuring the stability and accuracy of feature extraction. Then, the image semantic segmentation technology is combined to accurately locate the wheat ear area, and further calculate the wheat ear density and extract the wheat ear morphological features. Support vector regression and random forest regression models are used to integrate the wheat ear density distribution and morphological features to achieve high-precision prediction of wheat yield, significantly reduce the impact of human factors and natural environment changes on the wheat yield prediction results, and improve the reliability and practicality of the prediction.

[0177] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the wheat yield estimation method as described above.

[0178] In the above embodiment, the present application discloses a computer-readable storage medium, which belongs to the field of artificial intelligence technology. The present application obtains a wheat ear image, and pre-processes the wheat ear image using an illumination normalization algorithm to extract illumination-invariant features; inputs the illumination-invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image; uses a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image, and constructs a wheat ear density distribution map; extracts wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width, and wheat ear area; and estimates wheat yield using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological features. The present application preprocesses the wheat ear image through illumination normalization, effectively eliminating the interference of changes in illumination conditions on image features and ensuring the stability and accuracy of feature extraction. Then, the image semantic segmentation technology is combined to accurately locate the wheat ear area, and further calculate the wheat ear density and extract the wheat ear morphological features. Support vector regression and random forest regression models are used to integrate the wheat ear density distribution and morphological features to achieve high-precision prediction of wheat yield, significantly reduce the impact of human factors and natural environment changes on the wheat yield prediction results, and improve the reliability and practicality of the prediction.

[0179] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0180] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0181] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A wheat yield estimation method, characterized in that: include: Acquire a wheat ear image, and preprocess the wheat ear image using an illumination normalization algorithm to extract illumination invariant features; Inputting the illumination invariant feature into an image semantic segmentation model to obtain a wheat ear region sub-image; Calculate the density of wheat ears in each of the wheat ear region sub-images using a support vector regression algorithm to construct a wheat ear density distribution map; Extracting wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area; The wheat yield is estimated using a pre-trained random forest regression model based on the wheat ear density distribution map and the wheat ear morphological characteristics.

2. The wheat yield estimation method according to claim 1, characterized in that: The step of preprocessing the wheat ear image using an illumination normalization algorithm to extract illumination invariant features specifically includes: Obtaining the pixel value of each pixel point on the wheat ear image to obtain the original pixel value; Calculate the Gaussian surround function value of each pixel on the RGB color channel; Calculate the difference between the original pixel value of each pixel and the Gaussian surround function value to obtain the wheat ear image after illumination adjustment; Feature extraction is performed on the wheat ear image after the illumination adjustment to obtain the illumination invariant feature.

3. The wheat yield estimation method according to claim 2, characterized in that: The step of calculating the Gaussian surround function value of each pixel point on the RGB color channel specifically includes: Performing Gaussian blur processing on the wheat ear image to obtain a Gaussian blurred image of the wheat ear; Calculating the Gaussian weight of each pixel on the Gaussian blurred image; Determine a Gaussian kernel of the Gaussian blurred image, and determine a Gaussian kernel coverage range for each pixel point on the Gaussian blurred image; The Gaussian surround function value is calculated according to the Gaussian weight of each pixel and the Gaussian kernel coverage of each pixel.

4. The wheat yield estimation method according to claim 1, characterized in that: The step of using a support vector regression algorithm to calculate the wheat ear density of each wheat ear region sub-image and constructing a wheat ear density distribution map specifically includes: Performing density feature extraction on each wheat ear region sub-image to obtain wheat ear density features; Performing vector transformation on the wheat ear density feature to obtain a wheat ear density feature vector; Inputting the wheat ear density feature vector into a pre-trained support vector regression model, and outputting a wheat ear density label of each wheat ear region sub-image; The wheat ear density label of each wheat ear region sub-image is mapped to the wheat ear part image to obtain the wheat ear density distribution map.

5. The wheat yield estimation method according to claim 4, characterized in that: The step of mapping the wheat ear density label of each wheat ear region sub-image to the wheat ear part image to obtain the wheat ear density distribution map specifically includes: Determine the position area of ​​each wheat ear region sub-image in the wheat ear image to obtain a sub-image position area; Mapping the wheat ear density label of each wheat ear region sub-image into the corresponding sub-image position region; In each of the sub-image position areas, density value filling is performed based on the corresponding wheat ear density label to obtain the wheat ear density distribution map.

6. The wheat yield estimation method according to claim 1, characterized in that: The step of estimating wheat yield using a pre-trained random forest regression model according to the wheat ear density distribution map and the morphological characteristics specifically includes: Extracting characteristics of the wheat ear density distribution map to obtain wheat ear density distribution characteristics; Combining the wheat ear density distribution feature and the wheat ear morphological feature to generate a wheat ear density-morphological combination feature; Based on the wheat ear density-morphology combination characteristics, the wheat yield is estimated using a pre-trained random forest regression model.

7. The wheat yield estimation method according to claim 6, characterized in that: The step of combining the wheat ear density distribution feature and the wheat ear morphological feature to generate a wheat ear density-morphological combination feature specifically includes: Performing feature interaction processing on the wheat ear density distribution feature and the wheat ear morphological feature to obtain wheat ear interaction features, wherein the feature interaction processing includes calculating feature products, calculating feature ratios, and calculating feature differences; The wheat ear interaction feature is used as the wheat ear density-morphology combination feature.

8. A wheat yield estimation device, characterized in that: include: An illumination normalization module is used to obtain an image of a wheat ear, and preprocess the image of the wheat ear using an illumination normalization algorithm to extract illumination invariant features; An image semantic segmentation module, used for inputting the illumination invariant features into an image semantic segmentation model to obtain a wheat ear region sub-image; A wheat ear density calculation module, used for calculating the wheat ear density of each wheat ear region sub-image using a support vector regression algorithm, and constructing a wheat ear density distribution map; A morphological feature extraction module, used for extracting wheat ear morphological features from the wheat ear region sub-image, wherein the wheat ear morphological features include wheat ear length, wheat ear width and wheat ear area; The wheat yield estimation module is used to estimate the wheat yield using a pre-trained random forest regression model according to the wheat ear density distribution map and the wheat ear morphological characteristics.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the wheat yield estimation method as described in any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the wheat yield estimation method according to any one of claims 1 to 7 are implemented.

Images