A Rotary Tiller Soil Crushing Quality Detection System and Method Based on Image Technology

Abstract

The present invention provides a rotary tiller soil crushing quality detection system and method based on image technology. The method includes obtaining soil images during the working process of the rotary tiller and preprocessing them. The soil images are multi-channel color images; extracting soil feature regions from the preprocessed soil images through image segmentation technology; constructing a regularization model for soil quality evaluation based on the texture information of the soil feature regions, and accordingly constructing a minimization problem for soil quality evaluation; calculating parameters for soil quality evaluation based on the minimization problem, including indicators such as soil particle size distribution and soil humidity; and evaluating the soil crushing quality of the rotary tiller according to the calculated soil quality evaluation parameters to obtain the final soil crushing quality evaluation result. The present invention can achieve rapid and accurate evaluation of key indicators such as soil particle size distribution and humidity, and improve the efficiency of agricultural mechanized operations and the quality of soil tillage.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural mechanization. More specifically, the present invention relates to a rotary tiller soil fragmentation quality detection system and method based on image technology. Background Art

[0002] The existing rotary tiller soil fragmentation quality detection technology mainly relies on manual observation and empirical judgment. This method is not only inefficient but also easily affected by human factors, resulting in limited accuracy and consistency of detection results. In addition, traditional detection means cannot accurately measure key indicators such as soil particle size distribution and humidity, which to a certain extent restricts the further development of agricultural mechanization and the refinement of soil management. With the development of computer vision and image processing technology, people have begun to explore methods for automatically detecting soil quality using these technologies. However, the existing image-based detection technologies still face problems such as complex image acquisition conditions, inaccurate soil feature extraction, and imperfect evaluation models in practical applications. The existence of these problems makes it difficult for the existing technology to meet the requirements of rapid and accurate soil quality detection in modern agricultural production.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the existing technology: The soil image acquisition and preprocessing technology cannot well adapt to the changing field environment; The methods for extracting and analyzing soil features are not precise and comprehensive enough to fully reflect the actual quality of the soil; The existing soil quality evaluation models are often too simplified and cannot comprehensively consider various influencing factors, resulting in insufficient practicality and accuracy of the evaluation results. Summary of the Invention

[0004] The present invention provides a rotary tiller soil fragmentation quality detection system and method based on image technology.

[0005] In the first aspect of the present invention, a rotary tiller soil fragmentation quality detection method based on image technology is provided, including:

[0006] Obtain soil images during the operation of the rotary tiller and perform preprocessing, where the soil images are multi-channel color images;

[0007] Extract the soil feature regions from the preprocessed soil images through image segmentation technology;

[0008] Construct a regularization model for soil quality evaluation based on the texture information of the soil feature regions, and accordingly construct a minimization problem for soil quality evaluation;

[0009] Calculate the parameters for soil quality evaluation based on the minimization problem, including indicators such as soil particle size distribution and soil humidity;

[0010] Based on the calculated soil quality evaluation parameters, the soil crushing quality of the rotary tiller is evaluated to obtain the final soil crushing quality evaluation result.

[0011] Further, the texture information of the soil feature area includes:

[0012] The distribution information of soil particle size;

[0013] The distribution information of soil humidity;

[0014] The uniformity information of soil color;

[0015] The integrity information of soil structure.

[0016] Further, calculating the soil quality evaluation parameters based on the minimization problem includes:

[0017] Applying an optimization algorithm to solve the minimization problem; wherein, the optimization algorithm includes:

[0018] Introducing an auxiliary variable into the objective function of the minimization problem to convert the minimization problem into an unconstrained optimization problem;

[0019] Decomposing the unconstrained optimization problem into separate sub-problems and alternately iteratively solving the separate sub-problems to obtain the soil quality evaluation parameters.

[0020] Further, the sub-problem of the auxiliary variable y is solved by an iterative algorithm, including:

[0021] Introducing a dual variable to convert the solution of the sub-problem of the auxiliary variable y into a saddle point problem;

[0022] Solving the saddle point problem by an iterative algorithm to obtain the optimal auxiliary variable y.

[0023] In the second aspect of the present invention, a rotary tiller soil crushing quality detection system based on image technology is provided, including:

[0024] An image acquisition module for real-time acquisition of soil images during the operation of the rotary tiller;

[0025] An image preprocessing module for preprocessing operations such as noise removal and contrast enhancement on the acquired soil images;

[0026] An image segmentation module for extracting the soil feature area from the preprocessed soil images;

[0027] A texture analysis module for analyzing the texture information of the soil feature area and constructing a regularization model for soil quality evaluation;

[0028] A parameter calculation module for calculating soil quality evaluation parameters according to the regularization model, including indicators such as soil particle size distribution and soil humidity;

[0029] A quality assessment module is used to comprehensively evaluate the soil fragmentation quality of the rotary tiller based on the calculated soil quality evaluation parameters and output the evaluation results.

[0030] According to the above embodiments of the present invention, it has at least the following beneficial effects: The soil fragmentation quality detection method of the rotary tiller of the present invention can collect and preprocess the soil images in real time during the operation of the rotary tiller by adopting advanced image processing technology, effectively removing image noise and enhancing contrast, thereby improving the recognition accuracy of the soil feature area. This method uses image segmentation technology to accurately extract the soil feature area and constructs a regularization model based on soil texture information, which can comprehensively evaluate key indicators such as soil particle size distribution and humidity, and achieve accurate evaluation of soil quality.

[0031] In addition, this method can efficiently calculate soil quality evaluation parameters, including soil particle size and humidity, by constructing a minimization problem and applying an optimization algorithm. This method can not only improve the automation and intelligence level of the evaluation process, but also dynamically adjust the evaluation parameters according to different soil conditions and operation requirements to achieve personalized evaluation of the soil fragmentation quality of the rotary tiller. Through the comprehensive evaluation results, it can provide scientific decision-making support for agricultural mechanization operations, thereby optimizing the soil tillage process and improving agricultural production efficiency and crop yields. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By referring to the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown by way of illustration and not limitation, wherein:

[0033] Figure 1 It is a schematic flow chart of a method for detecting the soil fragmentation quality of a rotary tiller based on image technology provided by an embodiment of the present invention;

[0034] Figure 2 It is a schematic structural diagram of a system for detecting the soil fragmentation quality of a rotary tiller based on image technology provided by an embodiment of the present invention;

[0035] Figure 3 It schematically shows a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and then implement the present invention, rather than limiting the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to be able to fully convey the scope of the present invention to those skilled in the art.

[0037] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, equipment, method, or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0038] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0039] The following reference Figure 1 , Figure 1 is a schematic flow chart of a method for detecting the soil crushing quality of a rotary tiller based on image technology provided for an embodiment of the present invention. As Figure 1 shown, a method 100 for detecting the soil crushing quality of a rotary tiller based on image technology includes:

[0040] Step 101, obtaining a soil image during the operation of the rotary tiller and performing preprocessing, where the soil image is a multi-channel color image;

[0041] Step 102, extracting the soil feature region from the preprocessed soil image through image segmentation technology;

[0042] Step 103, constructing a regularization model for soil quality evaluation based on the texture information of the soil feature region, and accordingly constructing a minimization problem for soil quality evaluation;

[0043] Step 104, calculating the parameters of soil quality evaluation based on the minimization problem, including indicators such as soil particle size distribution and soil humidity;

[0044] Step 105, evaluating the soil crushing quality of the rotary tiller according to the calculated soil quality evaluation parameters to obtain the final soil crushing quality evaluation result.

[0045] It should be noted that obtaining a soil image during the operation of the rotary tiller means that a high-resolution camera needs to be installed on the rotary tiller to capture the soil image in real time. The preprocessing steps may include denoising, contrast enhancement, etc. to improve the image quality. Image segmentation technology can use methods such as edge detection and threshold segmentation to extract the soil feature region, which is the basis for subsequent analysis.

[0046] Preferably, advanced machine learning algorithms can be used for the preprocessing of soil images to automatically identify and remove noise in the images while enhancing the contrast of soil features. During the image segmentation process, convolutional neural networks based on deep learning can be used to improve the recognition accuracy of feature regions. In addition, when constructing a regularization model, multiple soil quality evaluation indicators, such as the particle size distribution, humidity, color uniformity, and structural integrity of the soil, can be introduced to form a comprehensive evaluation system.

[0047] In some embodiments, the texture information of the soil feature region includes:

[0048] Distribution information of soil particle size;

[0049] Distribution information of soil humidity;

[0050] Uniformity information of soil color;

[0051] Integrity information of soil structure.

[0052] It should be noted that the texture information of the soil feature region is one of the key factors for soil quality evaluation. This information includes not only the distribution information of soil particle size and humidity, but also the uniformity of soil color and the integrity of soil structure. The accurate extraction of these features is crucial for constructing an accurate soil quality evaluation model.

[0053] Specifically, the distribution information of soil particle size can be obtained through particle size analysis algorithms in image processing technology, such as using the gray-level co-occurrence matrix of the image to evaluate the distribution of different particle sizes. The distribution information of soil humidity can be inferred by analyzing the brightness and contrast of soil images. The uniformity of soil color can be evaluated through the distribution of color histograms, while the integrity of soil structure can be determined by analyzing the connection and arrangement between soil particles.

[0054] Preferably, the extraction of the soil feature region can adopt image segmentation algorithms based on deep learning, such as convolutional neural networks (CNNs), to improve the accuracy of feature extraction. In terms of parameter settings, the hyperparameters of the algorithm, such as the learning rate and batch size, can be adjusted according to different soil types and operating conditions to obtain the best model performance. In addition, the weight coefficients of the regularization terms can be dynamically adjusted according to the actual conditions of the soil to ensure the adaptability and robustness of the evaluation model. As an alternative, other image processing techniques, such as edge detection or texture analysis, can also be used to extract the texture information of the soil feature region.

[0055] In some embodiments, the objective function of the minimization problem is:

[0056]

[0057] In the formula, f is the preprocessed soil image, x is the soil quality evaluation parameter; ||·||2 is the Euclidean norm, R(x) is the regularization term of soil quality, and λ and μ are penalty term weight coefficients greater than 0.

[0058] It should be noted that the objective function of the minimization problem is constructed based on the mathematical model of soil image processing and quality evaluation. This objective function minimizes the difference between the soil image and the soil quality evaluation parameter, and at the same time adds a regularization term to improve the robustness of the model. The regularization term here helps to prevent the model from overfitting and ensures the generalization ability of the evaluation results.

[0059] Specifically, R(x) in the objective function represents the regularization term of soil quality, which is usually composed of multiple subterms, and each subterm corresponds to a soil quality evaluation index. For example, soil particle size distribution, humidity and other indicators can all be part of the regularization term. The parameters λ and μ are regularization parameters, which control the trade-off between the model's fitting to the data and the regularization term. The settings of these parameters need to be determined according to the actual soil characteristics and evaluation requirements to achieve the best model performance.

[0060] Preferably, the soil image f in the objective function can be specifically preprocessed, such as removing noise through filtering and improving image quality through contrast enhancement. The soil quality evaluation parameter x can include but is not limited to soil particle size distribution parameters, humidity parameters, etc., and these parameters can be obtained through image analysis. For example, soil particle size distribution can be determined through texture analysis of the image, and humidity can be estimated through the reflectance characteristics of the image. The regularization parameters λ and μ can be optimized through methods such as cross-validation to ensure that the model can maintain high evaluation accuracy and robustness under different conditions. In addition, the specific form of the regularization term R(x) can be adjusted according to different soil characteristics and evaluation requirements to adapt to different application scenarios.

[0061] In some embodiments, the regularization term based on the texture information of the soil feature region is shown in the following formula:

[0062]

[0063] where r i (x) is the regularization function of the i-th soil quality evaluation index, w i is the corresponding weight coefficient, and n is the number of evaluation indexes.

[0064] It should be noted that the regularization term is a key component. It combines the regularization functions of multiple evaluation indicators to form a comprehensive regularization model. This model not only considers the importance of individual indicators but also balances their influence on the final evaluation result through weight coefficients.

[0065] Specifically, each r i (x) in the regularization term represents the regularization function of the u-th soil quality evaluation indicator. These functions may include the distribution of soil particle size, the uniformity of soil moisture, the consistency of soil color, etc. The weight coefficient w i is used to adjust the relative importance of each indicator in the evaluation process. These parameters can be set according to the actual soil characteristics and the requirements of agricultural operations to ensure the accuracy and applicability of the evaluation results.

[0066] Preferably, the setting of the regularization term can be further refined. For example, the regularization function of soil particle size distribution can use a Gaussian mixture model to describe the distribution of different particle size levels, while the regularization function of soil moisture can be designed based on the statistical characteristics of soil moisture. In addition, the uniformity of soil color can be measured by the equalization degree of the color histogram, and the integrity of soil structure can be evaluated by texture analysis methods. In practical applications, appropriate regularization functions and weight coefficients can be selected according to the specific conditions of the soil and the working requirements of the rotary tiller to achieve the best evaluation effect.

[0067] In some embodiments, calculating the soil quality evaluation parameters based on the minimization problem includes:

[0068] Applying an optimization algorithm to solve the minimization problem; wherein, the optimization algorithm includes:

[0069] Introducing an auxiliary variable into the objective function of the minimization problem to transform the minimization problem into an unconstrained optimization problem;

[0070] Decomposing the unconstrained optimization problem into separate sub-problems and alternately iteratively solving the separate sub-problems to obtain the soil quality evaluation parameters.

[0071] It should be noted that applying an optimization algorithm to solve the minimization problem involves using specific mathematical methods to find the optimal solution that meets specific conditions. In the present invention, the optimization algorithm is used to minimize the objective function while considering the regularization term and the penalty term weight coefficient of soil quality evaluation.

[0072] Specifically, the optimization algorithm can be the gradient descent method, Newton's method, or other advanced optimization techniques. When setting the parameters of these algorithms, the characteristics of the soil image and the complexity of the evaluation index need to be considered. For example, the choice of the step size parameter will directly affect the convergence speed and stability of the algorithm. In addition, the setting of the regularization term and the penalty term weight coefficient needs to be determined through experiments to ensure that the model can capture the key features of the soil without losing the generalization ability due to overfitting.

[0073] Preferably, the optimization algorithm can adopt an adaptive step size adjustment strategy to dynamically adjust the step size according to the improvement of each iteration, so as to accelerate the convergence speed and avoid falling into local optima. More specifically, a momentum term can be introduced to help the algorithm jump out of saddle points, or quasi-Newton methods such as L-BFGS can be used to improve the calculation efficiency. In addition, parallel computing technology can also be considered to process large-scale data sets to adapt to the operation requirements of rotary tillers of different scales. In practical applications, the composition of the regularization term can also be adjusted according to the specific conditions of the soil to better adapt to different soil types and working environments.

[0074] In some embodiments, the auxiliary variable is:

[0075] y = D(x)

[0076] where D is the transformation operator of the soil quality evaluation parameter.

[0077] It should be noted that the auxiliary variable is a key concept introduced in the optimization algorithm. It is used to transform the minimization problem into an unconstrained problem, thereby simplifying the problem-solving process. The auxiliary variable is usually associated with some parameters or constraint conditions in the original problem. By introducing the auxiliary variable, the solution space can be explored more effectively and the solution efficiency can be improved.

[0078] Specifically, the auxiliary variable y is defined as the output of the transformation operator D of the soil quality evaluation parameter x, that is, y = D(x). Here, the transformation operator D can be linear or non-linear, and the specific form depends on the characteristics of the soil quality evaluation parameter and the requirements of the problem. For example, if the soil quality evaluation parameter x includes soil particle size and humidity, the transformation operator D may include normalization or standardization processing of these parameters.

[0079] Preferably, the choice of the transformation operator D can be customized according to the prior knowledge of the soil characteristics. For example, if the soil particle size distribution has specific statistical characteristics, a transformation operator can be designed to enhance the expression of these characteristics. In addition, the introduction of the auxiliary variable y can also include regularization processing of the soil quality evaluation parameter to avoid overfitting and improve the generalization ability of the model. In practical applications, L1 or L2 regularization methods can be adopted, and the model complexity can be controlled by adjusting the regularization coefficient.

[0080] In some embodiments, the separate sub - problems include an auxiliary variable y sub - problem and a soil quality evaluation parameter x sub - problem, as shown in the following formula:

[0081]

[0082] In the formula, <·,·> represents the dot product of vectors, and D(x) is the transformation operator of the soil quality evaluation parameter.

[0083] It should be noted that the separate sub - problems refer to the individual sub - problems that need to be solved independently after decomposing the minimization problem. These sub - problems are independent of each other but also interrelated. By solving them through alternating iteration, the optimal solution of the soil quality evaluation parameter can be gradually approximated.

[0084] Specifically, the auxiliary variable sub - problem and the soil quality evaluation parameter sub - problem respectively correspond to different parts of the minimization problem. The auxiliary variable sub - problem mainly focuses on the difference between the soil image and the image after transformation of the soil quality evaluation parameter, while the soil quality evaluation parameter sub - problem focuses on the optimization of the soil quality evaluation parameter itself. During the solution process, specific iteration times, step - size parameters, and regularization term weight coefficients can be set. The selection of these parameters will directly affect the efficiency of the solution and the accuracy of the result.

[0085] Preferably, in order to improve the solution efficiency and accuracy, some specific iterative algorithms can be adopted, such as the gradient descent method, the Newton method, or the quasi - Newton method, etc. In the solution of the auxiliary variable y sub - problem, dual variables can be introduced to transform the problem into a saddle - point problem and a specific iterative algorithm can be used to solve it. For example, the augmented Lagrangian multiplier method (ALM) can be used to solve it. By iteratively updating the auxiliary variable and the dual variable, the optimal solution can be gradually approximated. In the solution of the soil quality evaluation parameter sub - problem, an appropriate step - size adjustment strategy, such as an adaptive step - size or a momentum method, can be adopted to accelerate convergence and improve the stability of the solution. In addition, some regularization techniques, such as L1 regularization or L2 regularization, can be considered to enhance the generalization ability of the model and prevent the occurrence of overfitting.

[0086] In some embodiments, the auxiliary variable y sub - problem is solved by an iterative algorithm, including:

[0087] Introduce dual variables to transform the solution of the auxiliary variable y sub - problem into a saddle - point problem;

[0088] Solve the saddle - point problem through an iterative algorithm to obtain the optimal auxiliary variable y.

[0089] It should be noted that the solution process of the auxiliary variable sub - problem involves introducing dual variables and transforming the problem into a saddle - point problem. This process is to solve the optimization problem more effectively. By introducing additional variables, different parts in the objective function are balanced, thus simplifying the solution of the problem.

[0090] Specifically, introducing dual variables mentioned in the auxiliary variable sub - problem means adding a new variable in the optimization problem. This variable is associated with a certain constraint condition in the original problem to form the Lagrangian function. The saddle - point problem refers to finding a point in the optimization problem such that this point is both the local minimum and local maximum of the function among all possible variable values. In the present invention, the saddle - point problem is solved by an iterative algorithm, that is, by continuously iteratively updating the values of the variables to gradually approach the optimal solution.

[0091] Preferably, the iterative solution process of the auxiliary variable sub - problem can be further refined. For example, in the iterative process, a specific step - size parameter adjustment strategy can be adopted, such as using an adaptive step - size or a learning rate decay strategy, to accelerate the convergence speed or improve the solution stability. In addition, for the proximal operator of the regularization term, different algorithms can be selected for implementation, such as the soft - threshold algorithm or the gradient projection algorithm. These algorithms can be selected according to the characteristics of the actual problem to achieve a better solution effect. At the same time, the iterative algorithm can be improved, for example, by introducing a momentum term or adopting a non - linear convergence acceleration technique to further improve the efficiency and robustness of the algorithm.

[0092] In some embodiments, the calculation formula for the sub - problem of the auxiliary variable y is:

[0093]

[0094] where prox μR represents the proximal operator of the regularization term R, α is the step - size parameter, is the gradient of the objective function at x l ;

[0095] The calculation formula for the sub - problem of the soil quality evaluation parameter x is:

[0096]

[0097] where β is the step - size parameter, represents the gradient operator of x.

[0098] It should be noted that the solution process of the auxiliary variable sub - problem involves introducing dual variables, transforming the problem into a saddle - point problem, and solving it by an iterative algorithm. This method can effectively find the optimal auxiliary variable, thereby improving the accuracy.

[0099] Specifically, the calculation formula of the auxiliary variable sub-problem involves the proximal operator of the regularization term, the gradient of the objective function, and the setting of the step size parameter. In practical applications, the proximal operator of the regularization term can be a form of projection operation, the gradient of the objective function can be obtained by differentiating the objective function, and the step size parameter needs to be adjusted according to the specific situation of the problem to ensure the convergence and stability of the algorithm.

[0100] Preferably, the iterative update formula of the auxiliary variable can be further refined. For example, an adaptive step size strategy can be adopted to dynamically adjust the step size parameter according to the gradient change in each iteration. In addition, some regularization techniques, such as L1 or L2 regularization, can be considered to enhance the generalization ability of the model. In some cases, non-linear optimization algorithms, such as Newton's method or quasi-Newton's method, can also be considered to improve the solution efficiency. These alternative solutions can be selected and adjusted according to different application scenarios and performance requirements.

[0101] The above embodiments of the present invention have the following beneficial effects: The method for detecting the soil crushing quality of the rotary tiller of the present invention can realize the accurate extraction of soil characteristics and the automation of quality evaluation through the application of image technology. The method first preprocesses the soil image, which can ensure the quality and accuracy of the image data and lay a solid foundation for subsequent soil feature analysis. Through image segmentation technology, the soil feature area can be effectively separated from the complex background, which helps to analyze the texture information of the soil more accurately. Further, the regularization model constructed based on the texture information of the soil feature area can systematically evaluate the soil quality, including key indicators such as particle size distribution and humidity, thus providing a scientific evaluation means for the soil crushing quality of the rotary tiller.

[0102] In addition, the method of the present invention also includes constructing a minimization problem and applying an optimization algorithm to solve the soil quality evaluation parameters. This method can improve the efficiency and accuracy of parameter calculation. By converting the minimization problem into an unconstrained optimization problem and adopting an alternating iterative solution strategy, the soil quality evaluation parameters, including soil particle size and humidity, can be effectively solved. This method can not only improve the accuracy of soil quality evaluation, but also enhance the robustness and adaptability of the model by introducing auxiliary variables and transformation operators, making the evaluation results more reliable. Through this method, the soil crushing quality of the rotary tiller can be evaluated more precisely and comprehensively, thus providing more accurate guidance for agricultural mechanization operations.

[0103] As Figure 2 shown, a rotary tiller soil crushing quality detection system 200 based on image technology in some embodiments, the system 200 includes:

[0104] An image acquisition module 201, configured to acquire soil images in real time during the operation of the rotary tiller;

[0105] The image preprocessing module 202 is used to perform preprocessing operations such as noise removal and contrast enhancement on the collected soil images;

[0106] The image segmentation module 203 is used to extract the soil feature regions from the preprocessed soil images;

[0107] The texture analysis module 204 is used to analyze the texture information of the soil feature regions and construct a regularization model for soil quality evaluation;

[0108] The parameter calculation module 205 is used to calculate soil quality evaluation parameters according to the regularization model, including indicators such as soil particle size distribution and soil moisture;

[0109] The quality assessment module 206 is used to comprehensively evaluate the soil crushing quality of the rotary tiller according to the calculated soil quality evaluation parameters and output the evaluation results.

[0110] It can be understood that the various modules described in the rotary tiller soil crushing quality detection system 200 based on image technology correspond to the respective steps in the rotary tiller soil crushing quality detection method described in the reference. Thus, the operations, features, and beneficial effects described above for the rotary tiller soil crushing quality detection method based on image technology also apply to the rotary tiller soil crushing quality detection system 200 based on image technology and the modules included therein, and will not be elaborated here. Figure 1 The operations, features, and beneficial effects described above for the rotary tiller soil crushing quality detection method based on image technology also apply to the rotary tiller soil crushing quality detection system 200 based on image technology and the modules included therein, and will not be elaborated here.

[0111] Next, refer to Figure 3 , which shows a schematic structural diagram of the structure 300 of an electronic device suitable for use in implementing some embodiments of the present invention. The electronic device in some embodiments of the present invention may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal devices shown are merely examples and should not impose any limitations on the functions and usage scopes of the embodiments of the present invention.

[0112] As Figure 3As shown, the electronic device 300 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 301, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage device 308 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0113] Generally, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or wireline to exchange data. Although Figure 3 the electronic device 300 with various devices is shown, it should be understood that it is not required to implement or include all the shown devices. Instead, more or fewer devices may be implemented or included. Figure 3 Each block shown in the figure may represent a device or, as needed, multiple devices.

[0114] Furthermore, the storage medium of the embodiments of the present application stores program instructions capable of implementing all the above methods. Among them, the program instructions may be stored in the above storage medium in the form of a software product, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media that can store program codes, such as a USB flash drive, a removable hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc, or a terminal device such as a computer, a server, a mobile phone, or a tablet.

[0115] The above description is only some preferred embodiments of the present invention and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for detecting the soil crushing quality of a rotary tiller based on image technology, characterized in that, The method includes: Obtaining a soil image during the operation of a rotary tiller and performing preprocessing, where the soil image is a multi-channel color image; Extracting a soil feature region from the preprocessed soil image through image segmentation technology; Constructing a regularization model for soil quality evaluation based on the texture information of the soil feature region, and accordingly constructing a minimization problem for soil quality evaluation; Calculating parameters for soil quality evaluation based on the minimization problem, including soil particle size distribution and soil moisture index; Evaluating the soil breaking quality of the rotary tiller according to the calculated soil quality evaluation parameters to obtain a final soil breaking quality evaluation result; The calculating of the soil quality evaluation parameters based on the minimization problem includes: Applying an optimization algorithm to solve the minimization problem; wherein, the optimization algorithm includes: Introducing an auxiliary variable into the objective function of the minimization problem to convert the minimization problem into an unconstrained optimization problem; The objective function of the minimization problem is: In the formula, f is the preprocessed soil image, x is the soil quality evaluation parameter; ||·||2 is the Euclidean norm, R(x) is the regularization term of soil quality, and λ and μ are penalty term weight coefficients greater than 0; The calculating of the soil quality evaluation parameters based on the minimization problem includes: Decomposing the unconstrained optimization problem into individual sub-problems, and alternately iteratively solving the individual sub-problems to obtain the soil quality evaluation parameters; The individual sub-problems include an auxiliary variable y sub-problem and a soil quality evaluation parameter x sub-problem; the calculation formula for the auxiliary variable y sub-problem is: where prox μR denotes the proximal operator of the regularization term R, α is the step size parameter, is the gradient of the objective function at x l ; The calculation formula for the soil quality evaluation parameter x sub-problem is: where β is the step size parameter, represents the gradient operator with respect to x.

2. The method for detecting the soil crushing quality of a rotary tiller based on image technology according to claim 1, wherein The texture information of the soil feature region includes: Distribution information of soil particle size; Distribution information of soil moisture; Uniformity information of soil color; Integrity information of soil structure.

3. The method for detecting the soil breaking quality of a rotary tiller based on image technology according to claim 2, wherein, The auxiliary variable y sub-problem is solved through an iterative algorithm, including: introducing a dual variable to convert the solution of the auxiliary variable y sub-problem into a saddle point problem; Solving the saddle point problem through an iterative algorithm to obtain the optimal auxiliary variable y.

4. A rotary tiller soil crushing quality detection system based on image technology, which implements the method described in claim 1, characterized in that The system includes: An image acquisition module for real-time obtaining a soil image during the operation of a rotary tiller; An image preprocessing module for performing preprocessing operations such as noise removal and contrast enhancement on the acquired soil image; An image segmentation module for extracting a soil feature region from the preprocessed soil image; a texture analysis module for analyzing the texture information of the soil feature region and constructing a regularization model for soil quality evaluation; A parameter calculation module for calculating soil quality evaluation parameters according to the regularization model, including soil particle size distribution and soil moisture index; A quality evaluation module for comprehensively evaluating the soil breaking quality of the rotary tiller according to the calculated soil quality evaluation parameters and outputting an evaluation result.

Images