A facility cultivation method for overcoming obstacles to continuous cropping of sweet potato seedlings

Through deep learning technology, the image of sweet potato seedlings and soil state is analyzed, and the amount of intelligent fertilizer is generated, which solves the problem of soil nutrient imbalance in sweet potato seedlings and improves the yield and quality of sweet potato seedlings.

CN118749384BActive Publication Date: 2024-11-15SANYA INST OF HENAN UNIV +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411243452.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-11-15
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

Continuous cropping of sweet potato seedlings leads to imbalance in soil nutrients, affecting yield and quality. It is difficult for traditional cultivation methods to deal with growth needs and soil environment changes in time.

Method used

Using deep learning-based image processing technology, sweet potato seedlings and soil state images are collected through cameras, growth state characteristics and soil fertility characteristics are extracted, intelligent fertilizer application amount is generated, ensuring the accuracy of fertilization and avoiding fertilizer damage.

Benefits of technology

Accurate fertilization according to the growth needs of sweet potato seedlings is achieved, the problem of soil nutrient imbalance is solved, and the cultivation effect and yield of sweet potato seedlings are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118749384B_ABST
    Figure CN118749384B_ABST
Patent Text Reader

Abstract

The present application relates to a facility cultivation method for overcoming the continuous cropping obstacles of sweet potato seedlings. The method comprises: using deep learning-based image processing technology to extract features from the growth state image of sweet potato seedlings, mining the growth state features of the seedlings, and inferring the semantic feature description of the seedlings' adaptive fertility, and then intelligently generating the recommended amount of fertilizer based on the feature difference between the actual soil fertility state and the inferred seedlings' adaptive fertility features in the semantic feature space. In this way, precise fertilization can be achieved according to the growth needs of the sweet potato seedlings, effectively solving the problem of soil nutrient imbalance caused by continuous cropping obstacles, and improving the cultivation effect and yield of sweet potato seedlings.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of seedling cultivation, and in particular, to a facility cultivation method for overcoming obstacles to continuous cropping of sweet potato seedlings. Background Art

[0002] As an important food crop, sweet potato is widely planted around the world and is deeply loved by growers and consumers. However, in the process of continuous planting of sweet potato, the problem of continuous cropping barrier is often faced. Continuous cropping barrier refers to the phenomenon that after the continuous planting of the same crop or related crops, due to factors such as the accumulation of pathogens in the soil, nutrient imbalance and changes in soil structure, the crop yield decreases, the quality deteriorates, and the development condition deteriorates. Continuous cropping barrier of sweet potato has become one of the key issues restricting the sustainable development of the sweet potato industry. In severe cases, it can even lead to a 20% to 80% reduction in crop yield, which has a serious impact on agricultural production.

[0003] Traditional sweet potato cultivation methods often rely on the grower's experience and intuition for soil management and fertilization, which makes it difficult to respond in a timely manner to the complex growth needs and soil environmental changes during sweet potato seedling cultivation, which can easily lead to poor growth of sweet potato seedlings, reduced yields, and frequent occurrence of diseases and insect pests.

[0004] Therefore, a facility cultivation method that overcomes the obstacles of continuous cropping of sweet potato seedlings is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides a facility cultivation method for overcoming the obstacles of continuous cropping of sweet potato seedlings, the method comprising:

[0007] Build plastic greenhouses;

[0008] Carrying out a treatment to eliminate the soil in the plastic greenhouse;

[0009] Providing a nursery bed in the plastic greenhouse, and sowing sweet potato seedlings on the nursery bed;

[0010] Based on the growth requirements of the sweet potato, fertilizing the sweet potato seedlings to avoid fertilizer damage caused by excessive fertilization;

[0011] Wherein, based on the growth requirements of the sweet potato, fertilizing the sweet potato seedlings to avoid fertilizer damage caused by excessive fertilization includes:

[0012] Acquire the growth status image of the sweet potato seedlings and the soil status image collected by the camera;

[0013] Extracting the growth state features of the sweet potato seedlings and performing feature gradient mask enhancement processing on the growth state image of the sweet potato seedlings to obtain a significant feature map of the growth state of the sweet potato seedlings;

[0014] Based on the sweet potato seedling growth state significant feature map, soil fertility reasoning is performed to obtain a soil adaptive fertility reasoning semantic feature vector;

[0015] Extracting soil fertility state features of the soil state image to obtain a real-time soil fertility state semantic feature vector;

[0016] Based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector, a decoded value of the recommended fertilization amount is generated.

[0017] Optionally, the sweet potato seedling growth status image is subjected to seedling growth status feature extraction and feature gradient mask enhancement processing to obtain a sweet potato seedling growth status salient feature map, including: extracting seedling growth status features of the sweet potato seedling growth status image to obtain a sweet potato seedling growth status feature map; inputting the sweet potato seedling growth status feature map into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth status salient feature map.

[0018] Optionally, extracting the seedling growth status features of the sweet potato seedling growth status image to obtain a sweet potato seedling growth status feature map, including: inputting the sweet potato seedling growth status image into a seedling growth status feature extractor based on a first hole convolutional neural network model to obtain the sweet potato seedling growth status feature map.

[0019] Optionally, the sweet potato seedling growth state characteristic map is input into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth state significant feature map, including: calculating the multidirectional gradient value distribution of each position in the sweet potato seedling growth state characteristic map, and determining the gradient amplitude value of each position in the sweet potato seedling growth state characteristic map based on the multidirectional gradient value distribution of each position to obtain the sweet potato seedling growth state characteristic gradient amplitude distribution map; calculating the gradient amplitude local description operator of each position in the sweet potato seedling growth state characteristic gradient amplitude distribution map to obtain the sweet potato seedling growth state characteristic gradient amplitude local significant distribution map; inputting the sweet potato seedling growth state characteristic gradient amplitude local significant distribution map into a gated masker based on a GELU function to obtain a gradient amplitude local significant gated mask map; calculating the position point multiplication between the gradient amplitude local significant gated mask map and the sweet potato seedling growth state characteristic map to obtain the sweet potato seedling growth state significant feature map.

[0020] Optionally, the local description operator of the gradient amplitude at each position in the characteristic gradient amplitude distribution map of the sweet potato seedling growth state is calculated to obtain a local significant distribution map of the characteristic gradient amplitude of the sweet potato seedling growth state, including: determining the scale of the local neighborhood, calculating the mean of the difference between the gradient amplitude value at a predetermined position in the characteristic gradient amplitude distribution map of the sweet potato seedling growth state and the gradient amplitude values ​​at other positions in the local neighborhood to obtain the local description operator of the gradient amplitude corresponding to the predetermined position.

[0021] Optionally, soil fertility inference is performed based on the significant feature map of the sweet potato seedling growth status to obtain the soil adaptive fertility inference semantic feature vector, including: inputting the significant feature map of the sweet potato seedling growth status into a decoder-based soil fertility inference device to obtain the soil adaptive fertility inference semantic feature vector.

[0022] Optionally, extracting soil fertility status features of the soil status image to obtain a real-time soil fertility status semantic feature vector includes: inputting the soil status image into a soil fertility status feature extractor based on a second atrous convolutional neural network model to obtain the real-time soil fertility status semantic feature vector.

[0023] Optionally, based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector, a decoded value of the recommended fertilization amount is generated, including: calculating the positional difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector to obtain a soil fertility deficiency status semantic feature vector; inputting the soil fertility deficiency status semantic feature vector into a decoder-based fertilization intelligent controller to obtain a decoded value of the recommended fertilization amount.

[0024] By adopting the above technical solution, the image of the growth state of sweet potato seedlings is extracted by using deep learning-based image processing technology, the growth state characteristics of the seedlings are mined, and the semantic feature description of the seedlings' adaptive fertility is inferred based on this, and then the recommended amount of fertilizer is intelligently generated based on the feature difference between the actual soil fertility state and the inferred seedlings' adaptive fertility characteristics in the semantic feature space. In this way, precise fertilization can be achieved according to the growth needs of sweet potato seedlings, effectively solving the problem of soil nutrient imbalance caused by continuous cropping obstacles, and improving the cultivation effect and yield of sweet potato seedlings.

[0025] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0027] Figure 1 It is a flow chart of a facility cultivation method for overcoming obstacles to continuous cropping of sweet potato seedlings according to an exemplary embodiment.

[0028] Figure 2 It is a block diagram of a facility cultivation system for overcoming obstacles to continuous cropping of sweet potato seedlings according to an exemplary embodiment.

[0029] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0030] Figure 4 It is an application scenario diagram of a facility cultivation method for overcoming obstacles to continuous cropping of sweet potato seedlings according to an exemplary embodiment. DETAILED DESCRIPTION

[0031] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0032] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0033] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0034] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0035] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0036] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0037] Sweet potatoes, as an important food crop widely grown around the world, are not only rich in nutrition, but also have won the favor of growers and consumers around the world with their unique flavor and diverse ways of eating. However, in the process of sweet potato cultivation, especially when planted continuously, growers often encounter a thorny problem - continuous cropping disorder. This phenomenon refers to the fact that when the same crop or its related crops are planted continuously on the same piece of land, due to the accumulation of pathogens in the soil, nutrient imbalance, and changes in soil structure, etc., the crop yield decreases, the quality deteriorates, and even the growth and development conditions deteriorate.

[0038] The problem of continuous cropping of sweet potatoes has become one of the key issues restricting the sustainable development of the sweet potato industry, bringing extremely serious impacts on agricultural production. This reduction in production not only reduces farmers' economic benefits, but also may affect food security and market supply.

[0039] Traditional sweet potato cultivation methods often rely on the grower's experience and intuition to manage the soil and apply fertilizers. This method often seems to be inadequate when faced with the complex growth requirements and soil environmental changes during the sweet potato seedling cultivation process. Due to the lack of scientific data support and precise analysis tools, this cultivation method that relies on personal experience is difficult to adjust management strategies in a timely and accurate manner to meet the actual needs of sweet potato growth. As a result, this not only easily leads to poor growth of sweet potato seedlings, but may also cause a decrease in yield and increase the frequency of pests and diseases, thus affecting the healthy development of the entire sweet potato industry.

[0040] In order to overcome the obstacles of continuous cropping in the process of sweet potato seedling cultivation, a more scientific and systematic method of facility cultivation is urgently needed. This method should be based on a deep understanding of soil and crop growth conditions, and adopt modern technical means to achieve real-time monitoring and intelligent analysis of soil fertility, crop growth status and environmental factors. Through this intelligent monitoring and analysis, the growth needs of sweet potatoes can be more accurately grasped, and soil management and fertilization strategies can be adjusted in time, thereby effectively improving the yield and quality of sweet potatoes.

[0041] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0042] Figure 1is a flow chart of a facility cultivation method for overcoming obstacles to continuous cropping of sweet potato seedlings according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0043] Step S101, building a plastic greenhouse;

[0044] Step S102, performing a treatment on the soil in the plastic greenhouse;

[0045] Step S103, setting a nursery bed in the plastic greenhouse, and sowing sweet potato seedlings on the nursery bed;

[0046] Step S104: fertilizing the sweet potato seedlings based on the sweet potato growth requirements to avoid fertilizer damage caused by excessive fertilization;

[0047] Wherein, step S104, fertilizing the sweet potato seedlings based on the sweet potato growth requirements to avoid fertilizer damage caused by excessive fertilization, comprises:

[0048] Step S1041, acquiring a sweet potato seedling growth state image and a soil state image captured by a camera;

[0049] Step S1042, extracting the seedling growth state features and performing feature gradient mask enhancement processing on the sweet potato seedling growth state image to obtain a significant feature map of the sweet potato seedling growth state;

[0050] Step S1043, performing soil fertility reasoning based on the sweet potato seedling growth state significant feature map to obtain a soil adaptive fertility reasoning semantic feature vector;

[0051] Step S1044, extracting soil fertility state features of the soil state image to obtain a real-time soil fertility state semantic feature vector;

[0052] Step S1045: generating a decoded value of the recommended fertilization amount based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector.

[0053] In response to the above technical problems, the technical concept of this application is to use deep learning-based image processing technology to extract features from the sweet potato seedling growth state image, dig out the growth state characteristics of the seedlings, and infer the semantic feature description of the seedlings' adaptive fertility based on this, and then intelligently generate the recommended amount of fertilizer based on the feature difference between the actual soil fertility state and the inferred seedlings' adaptive fertility characteristics in the semantic feature space. In this way, precise fertilization can be achieved according to the growth needs of the sweet potato seedlings, effectively solving the problem of soil nutrient imbalance caused by continuous cropping obstacles, and improving the cultivation effect and yield of sweet potato seedlings.

[0054] Based on this, the present application proposes a facility cultivation method for overcoming the obstacles of continuous cropping of sweet potato seedlings, which includes: building a plastic greenhouse; eliminating the soil in the plastic greenhouse; setting up a seedling bed in the plastic greenhouse, and sowing sweet potato seedlings on the seedling bed; based on the growth requirements of sweet potatoes, fertilizing the sweet potato seedlings to avoid fertilizer damage caused by excessive fertilization.

[0055] Specifically, in the process of fertilizing the sweet potato seedlings, first, the growth status image of the sweet potato seedlings and the soil status image captured by the camera are obtained. It should be understood that using a camera for image acquisition is a non-destructive detection method that does not cause any physical damage to the sweet potato seedlings and the soil, allowing multiple detections and evaluations to be performed without affecting the normal growth of the crop, which has significant advantages for the sustainability and safety of agricultural production.

[0056] In one embodiment of the present application, the sweet potato seedling growth status image is subjected to seedling growth status feature extraction and feature gradient mask enhancement processing to obtain a sweet potato seedling growth status salient feature map, including: extracting seedling growth status features of the sweet potato seedling growth status image to obtain a sweet potato seedling growth status feature map; inputting the sweet potato seedling growth status feature map into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth status salient feature map.

[0057] Next, in order to effectively extract the overall morphological information and minute detail features such as leaf size, shape, and signs of pests and diseases from the sweet potato seedling growth status image, the present application uses a seedling growth status feature extractor based on the first atrous convolutional neural network model to extract image features from the sweet potato seedling growth status image. By utilizing the atrous convolution characteristics, without significantly increasing the number of parameters and the amount of calculation, the receptive field is increased by introducing holes (i.e., the dilation rate), so that the network can capture a wider range of spatial contextual information, thereby effectively capturing the overall growth morphological characteristics of the seedlings while maintaining sensitivity to minute detail features, thereby generating a sweet potato seedling growth status feature map.

[0058] In one embodiment of the present application, the seedling growth status features of the sweet potato seedling growth status image are extracted to obtain a sweet potato seedling growth status feature map, including: inputting the sweet potato seedling growth status image into a seedling growth status feature extractor based on a first hole convolutional neural network model to obtain the sweet potato seedling growth status feature map.

[0059] Furthermore, in order to highlight the key features of the sweet potato seedlings, such as the leaf shape, edge contour, and lesions, the present application further introduces a gradient mask feature significance enhancement module to process the sweet potato seedling growth state feature map, based on the gradient distribution of the feature map, to highlight the significant feature area therein. Specifically, the gradient mask feature significance enhancement module first quantifies the local change intensity of the pixels at each position by calculating the gradient amplitude value at each position in the sweet potato seedling growth state feature map. Then, the gradient amplitude local description operator is calculated by measuring the significance of the gradient amplitude value at each position relative to the gradient amplitude value at other positions in the local neighborhood, and the feature importance of each position is evaluated. Then, the GELU function is used to perform nonlinear activation on the gradient amplitude local description operator at each position to generate a gated mask for guiding the enhancement of key information in the sweet potato seedling growth state feature map. Finally, the sweet potato seedling growth state feature map is element-wise weighted based on the generated gated mask, so as to focus on the key seedling growth state features and obtain a significant feature map of the sweet potato seedling growth state.

[0060] In one embodiment of the present application, the sweet potato seedling growth state characteristic map is input into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth state significant feature map, including: calculating the multidirectional gradient value distribution of each position in the sweet potato seedling growth state characteristic map, and determining the gradient amplitude value of each position in the sweet potato seedling growth state characteristic map based on the multidirectional gradient value distribution of each position to obtain the sweet potato seedling growth state characteristic gradient amplitude distribution map; calculating the gradient amplitude local description operator of each position in the sweet potato seedling growth state characteristic gradient amplitude distribution map to obtain the sweet potato seedling growth state characteristic gradient amplitude local significant distribution map; inputting the sweet potato seedling growth state characteristic gradient amplitude local significant distribution map into a gated masker based on a GELU function to obtain a gradient amplitude local significant gated mask map; calculating the position point multiplication between the gradient amplitude local significant gated mask map and the sweet potato seedling growth state characteristic map to obtain the sweet potato seedling growth state significant feature map.

[0061] Furthermore, in one embodiment of the present application, the local description operator of the gradient amplitude of each position in the characteristic gradient amplitude distribution map of the sweet potato seedling growth state is calculated to obtain a local significant distribution map of the characteristic gradient amplitude of the sweet potato seedling growth state, including: determining the scale of the local neighborhood, calculating the mean of the difference between the gradient amplitude value of a predetermined position in the characteristic gradient amplitude distribution map of the sweet potato seedling growth state and the gradient amplitude values ​​of other positions in the local neighborhood to obtain the local description operator of the gradient amplitude corresponding to the predetermined position.

[0062] Specifically, the sweet potato seedling growth state characteristic map is processed by the following characteristic distribution gradient mask formula to obtain the sweet potato seedling growth state significant characteristic map, wherein the characteristic distribution gradient mask formula is: in, , , , , and Respectively represent the grayscale values ​​of the corresponding positions of the sweet potato seedling growth state characteristic diagram, The sweet potato seedling growth state characteristic diagram The gradient value in the horizontal direction of the pixel point. The sweet potato seedling growth state characteristic diagram The gradient value in the ordinate direction of the pixel point. The sweet potato seedling growth state characteristic diagram is shown in The channel direction gradient value of the position pixel point, The first characterization diagram of the growth state of the sweet potato seedlings is shown in FIG. The gradient amplitude value of the position, A represents the gradient amplitude distribution map of the sweet potato seedling growth state characteristics The set of gradient magnitude values ​​in the local neighborhood centered at position, represents the number of gradient magnitude values ​​in the set of gradient magnitude values ​​in the local neighborhood, and Respectively represent the offset in the horizontal axis direction, the vertical axis direction and the channel direction, The first characterization diagram of the growth state of the sweet potato seedlings is shown in FIG. The local description operator of the gradient magnitude of the position, represents the GELU function,

[0063] The first part of the gradient magnitude local significant gated mask map is represented by The local description operator of the gradient magnitude of the position, represents the local significant gated mask map of the gradient magnitude, It means point multiplication by position. A diagram showing significant characteristics of the growth status of the sweet potato seedlings.

[0064] The growth state of sweet potato seedlings is a direct reflection of their growth environment, especially the soil fertility state. According to the characteristics of sweet potato seedlings such as leaf color, stem thickness, and root development, their demand for nutrients can be identified. Based on this, in the technical solution of the present application, a decoder-based soil fertility reasoner is used to decode and reconstruct the significant feature map of the sweet potato seedling growth state. By modeling the complex relationship between the seedling growth state and soil fertility, the seedling growth state features are mapped to the semantic space of soil fertility, and the ideal demand of sweet potato seedlings for soil fertility is expressed in the form of a vector in a high-dimensional semantic feature space, thereby generating a soil-adapted fertility reasoning semantic feature vector, which provides an important reference benchmark for the regulation of soil fertility.

[0065] In one embodiment of the present application, soil fertility inference is performed based on the significant feature map of the sweet potato seedling growth status to obtain the soil adaptive fertility inference semantic feature vector, including: inputting the significant feature map of the sweet potato seedling growth status into a decoder-based soil fertility reasoner to obtain the soil adaptive fertility inference semantic feature vector.

[0066] Secondly, in order to extract the current soil state information, the soil state image is further input into the soil fertility state feature extractor based on the second atrous convolutional neural network model for feature extraction. The second atrous convolutional neural network model adopts a deep network structure and extracts semantic features of the soil state image through layer-by-layer convolution processing to mine relevant features such as soil moisture, fine texture, color distribution, and pests and diseases, obtain the characteristic representation of the current soil fertility state, and obtain the real-time soil fertility state semantic feature vector.

[0067] In one embodiment of the present application, extracting soil fertility status features of the soil status image to obtain a real-time soil fertility status semantic feature vector includes: inputting the soil status image into a soil fertility status feature extractor based on a second atrous convolutional neural network model to obtain the real-time soil fertility status semantic feature vector. Specifically, the second atrous convolutional neural network model includes a plurality of atrous convolutional layers, an activation layer based on a ReLU function, and a pooling layer. The plurality of atrous convolutional layers have different atrous rates, and as the network level deepens, the atrous rate gradually increases, thereby achieving image feature extraction at different levels; the activation layer enhances the expression capability of the model by introducing nonlinear transformations; and the pooling layer is used to reduce feature dimensions and retain important information.

[0068] More specifically, in the training process of the second hole convolutional neural network model, first, a large number of soil state image data sets are collected, which cover various situations such as different humidity, texture, color distribution, and whether there are pests and diseases. Then, for each image, corresponding annotations are performed, such as humidity level (such as high, medium, low), texture features (such as rough, smooth), color distribution (such as RGB value) and pest and disease conditions (such as presence, absence). After the annotation is completed, these image data sets are divided into training set, verification set and test set. The second hole convolutional neural network model is trained for multiple rounds on the training set, and the network parameters are adjusted using the back propagation algorithm and the gradient descent optimization method so that it can effectively learn the feature representation of different soil state images. The model is verified on the verification set to ensure that the model has good generalization ability. Finally, the final performance evaluation is performed on the test set to ensure that the second hole convolutional neural network model can accurately extract the key features in the soil state image. After the training is completed, the model is used to extract features from the new soil state image, and the semantic feature vector of the real-time fertility state of the soil is obtained through layer-by-layer feature transfer.

[0069] Then, in order to further quantify whether the current soil fertility deviates from the optimal conditions for seedling growth, in the technical solution of the present application, by calculating the positional difference between the soil adaptive fertility inference semantic feature vector and the soil real-time fertility status semantic feature vector, the difference between the actual soil fertility and the ideal adaptive fertility in the semantic feature space is revealed, and a soil fertility deficiency status semantic feature vector is generated to provide a quantitative basis for precise fertilization.

[0070] Furthermore, the soil fertility deficiency state semantic feature vector is input into a decoder-based intelligent fertilization controller to obtain a decoded value of the recommended fertilization amount. In the technical solution of the present application, the decoder-based intelligent fertilization controller is trained and optimized to establish a mapping relationship between the fertilization amount and the soil fertility state difference, so that the amount of fertilizer currently required to be applied can be calculated according to the fertility deficiency information contained in the soil fertility deficiency state semantic feature vector, and the intelligent control of the fertilization amount can be achieved to ensure that while meeting the growth needs of the sweet potato seedlings, soil nutrient imbalance and environmental problems caused by excessive fertilization are avoided.

[0071] In one embodiment of the present application, based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector, a decoded value of the recommended fertilization amount is generated, including: calculating the positional difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector to obtain a soil fertility deficiency status semantic feature vector; inputting the soil fertility deficiency status semantic feature vector into a decoder-based fertilization intelligent controller to obtain a decoded value of the recommended fertilization amount.

[0072] In a preferred example, the semantic feature vector of the soil fertility deficiency state is passed through a decoder-based intelligent fertilization controller to obtain a decoded value of the recommended fertilization amount, which includes:

[0073] Determine the maximum eigenvalue of the soil fertility state semantics and the minimum eigenvalue of the soil fertility state semantics feature vector, and calculate the mean value of the soil fertility state semantics and the standard deviation of the soil fertility state semantics of the feature set of the soil fertility state semantics feature vector;

[0074] Calculating the quotient of the semantic mean of the soil fertility deficiency state and the semantic standard deviation of the soil fertility deficiency state to obtain a semantic statistical standardized value of the soil fertility deficiency state;

[0075] Calculate the inverse of each eigenvalue of the soil fertility deficiency state semantic feature vector, multiply it by the difference between the maximum eigenvalue of the soil fertility deficiency state semantics and the minimum eigenvalue of the soil fertility deficiency state semantics, and then perform point subtraction from the soil fertility deficiency state semantic statistical normalized value to obtain a soil fertility deficiency state semantic distribution approximation vector;

[0076] Calculate an exponential function with a natural constant as the base, with each eigenvalue of the soil fertility deficiency state semantic distribution approximation vector as an exponent, and perform point addition with the soil fertility deficiency state semantic statistical normalized value to obtain a soil fertility deficiency state semantic probability approximation vector;

[0077] Calculating the base-two logarithm of the absolute value of each eigenvalue of the soil fertility deficiency state semantic probability approximation vector to obtain an optimized soil fertility deficiency state semantic feature vector; and

[0078] The optimized semantic feature vector of soil fertility deficiency state is passed through the decoder-based intelligent fertilization controller to obtain a decoded value of the recommended fertilization amount.

[0079] Among them, the semantic feature vector of soil fertility deficiency state is recorded as The optimization expression is:

[0080] in, and are the feature sets of the semantic feature vector of the soil fertility deficiency state The mean, standard deviation, minimum and maximum values ​​of is the base 2 logarithm. represents the inverse of each eigenvalue of the semantic feature vector of the soil fertility deficiency state, To calculate the exponential function of each eigenvalue of a vector with a natural constant as the base, is the logarithm with base 2, is the optimized semantic feature vector of soil fertility deficiency state, It is added by location point. It is subtracted by position point. It is the point product by position.

[0081] Here, in the preferred example, the sweet potato seedling growth state feature map and the soil real-time fertility state semantic feature vector respectively represent the seedling growth state image semantic features and soil fertility semantic image features determined by the sweet potato seedling growth state image and the soil state image through dilated convolution coding. The soil adaptive fertility reasoning semantic feature vector represents the soil fertility decoding semantic features corresponding to the seedling growth state image semantic features determined by the decoder-based soil fertility reasoner. Therefore, when calculating the positional difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility status semantic feature vector to obtain the soil fertility deficiency status semantic feature vector, the obtained soil fertility deficiency status semantic feature vector not only has the image semantic difference feature between the sweet potato seedling growth status image and the soil status image, but also has the soil fertility difference semantic feature determined by decoding reasoning, that is, the soil fertility deficiency status semantic feature vector has a diversified set expression distribution as a feature distribution as a whole. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness of the soil fertility deficiency status semantic feature vector when it is decoded and regressed by a decoder-based fertilization intelligent controller, thereby improving the accuracy of the decoded value of the recommended fertilization amount.

[0082] Therefore, by performing random statistical normalization on the diversified feature set of the semantic feature vector of the soil fertility deficiency state, the confidence space constructed based on the overall eigenvalue of the semantic feature vector of the soil fertility deficiency state is approximated by the standardized continuous probability density distribution relative to the response hypothesis test of each eigenvalue of the semantic feature vector of the soil fertility deficiency state, thereby establishing the target reachability of the discrete feature distribution of the semantic feature vector of the soil fertility deficiency state to a unified regression target, so as to achieve a balanced executable performance between mapping accuracy and mapping completeness in the decoding regression process based on the discrete feature distribution of the semantic feature vector of the soil fertility deficiency state, and improve the accuracy of the decoded value of the recommended fertilization amount obtained by the semantic feature vector of the soil fertility deficiency state through the decoder-based fertilization intelligent controller.

[0083] In summary, the above scheme uses deep learning-based image processing technology to extract features from the sweet potato seedling growth state image, mines the growth state features of the seedlings, and infers the semantic feature description of the seedlings' adaptive fertility based on this, and then intelligently generates the recommended fertilizer amount based on the feature difference between the actual soil fertility state and the inferred seedlings' adaptive fertility features in the semantic feature space. In this way, precise fertilization can be achieved according to the growth needs of the sweet potato seedlings, effectively solving the problem of soil nutrient imbalance caused by continuous cropping obstacles, and improving the cultivation effect and yield of sweet potato seedlings.

[0084] Figure 2 FIG. 1 is a block diagram of a facility cultivation system for overcoming obstacles to continuous cropping of sweet potato seedlings according to an exemplary embodiment. Figure 2 As shown, the system 200 includes:

[0085] A plastic greenhouse construction module 201 is used to construct a plastic greenhouse;

[0086] The elimination processing module 202 is used to eliminate the soil in the plastic greenhouse;

[0087] The sweet potato seedling sowing module 203 is used to set a seedling bed in the plastic greenhouse and sow the sweet potato seedlings on the seedling bed;

[0088] A fertilization module 204 is used to fertilize the sweet potato seedlings based on the sweet potato growth requirements to avoid fertilizer damage caused by excessive fertilization;

[0089] Wherein, the fertilization module 204 includes:

[0090] The image acquisition unit 2041 is used to acquire the sweet potato seedling growth state image and the soil state image collected by the camera;

[0091] A seedling growth state feature extraction unit 2042 is used to extract the seedling growth state features and perform feature gradient mask enhancement processing on the sweet potato seedling growth state image to obtain a sweet potato seedling growth state significant feature map;

[0092] A soil fertility reasoning unit 2043 is used to perform soil fertility reasoning based on the sweet potato seedling growth state significant feature map to obtain a soil adaptive fertility reasoning semantic feature vector;

[0093] A soil fertility state feature extraction unit 2044 is used to extract soil fertility state features of the soil state image to obtain a real-time soil fertility state semantic feature vector;

[0094] The recommended fertilization amount generating unit 2045 is used to generate a decoded value of the recommended fertilization amount based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility state semantic feature vector.

[0095] Reference below Figure 3, which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application 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), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0096] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0097] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0098] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0099] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0100] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0101] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0102] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0103] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0104] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0105] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0106] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0107] Figure 4 : is an application scenario diagram of a facility cultivation method for overcoming the obstacles of continuous cropping of sweet potato seedlings according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, the growth status image of the sweet potato seedlings collected by the camera is obtained (for example, Figure 4 C1) and soil state images (e.g. Figure 4 Then, the acquired sweet potato seedling growth state image and soil state image are input to a server (for example, Figure 4 In S) shown in , the server is capable of processing the sweet potato seedling growth status image and the soil status image based on a facility cultivation algorithm that overcomes the obstacles of continuous cropping of sweet potato seedlings to generate a decoded value of the recommended fertilizer amount.

[0108] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0109] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0110] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A facility cultivation method for overcoming the obstacle of continuous cropping of sweet potato seedlings, characterized in that: include: Build plastic greenhouses; Carrying out a treatment to eliminate the soil in the plastic greenhouse; Setting a nursery bed in the plastic greenhouse, and sowing sweet potato seedlings on the nursery bed; Based on the growth requirements of the sweet potato, fertilizing the sweet potato seedlings to avoid fertilizer damage caused by excessive fertilization; Wherein, based on the growth requirements of the sweet potato, fertilizing the sweet potato seedlings to avoid fertilizer damage caused by excessive fertilization includes: Acquire the growth status image of the sweet potato seedlings and the soil status image collected by the camera; Extracting the growth state features of the sweet potato seedlings and performing feature gradient mask enhancement processing on the growth state image of the sweet potato seedlings to obtain a significant feature map of the growth state of the sweet potato seedlings; Based on the sweet potato seedling growth state significant feature map, soil fertility reasoning is performed to obtain a soil adaptive fertility reasoning semantic feature vector; Extracting soil fertility state features of the soil state image to obtain a real-time soil fertility state semantic feature vector; Generate a decoded value of a recommended fertilization amount based on a feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility state semantic feature vector; Extracting the seedling growth state features and performing feature gradient mask enhancement processing on the sweet potato seedling growth state image to obtain a significant feature map of the sweet potato seedling growth state, including: Extracting the seedling growth state characteristics of the sweet potato seedling growth state image to obtain a sweet potato seedling growth state characteristic graph; Inputting the sweet potato seedling growth state feature map into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth state significant feature map; Extracting the seedling growth state features of the sweet potato seedling growth state image to obtain a sweet potato seedling growth state feature map includes: Inputting the sweet potato seedling growth state image into a seedling growth state feature extractor based on a first atrous convolutional neural network model to obtain the sweet potato seedling growth state feature map; Inputting the sweet potato seedling growth state feature map into a gradient mask feature significance enhancement module to obtain the sweet potato seedling growth state significant feature map, comprising: Calculating the multidirectional gradient value distribution of each position in the sweet potato seedling growth state characteristic map, and determining the gradient amplitude value of each position in the sweet potato seedling growth state characteristic map based on the multidirectional gradient value distribution of each position to obtain a sweet potato seedling growth state characteristic gradient amplitude distribution map; Calculating the local description operator of the gradient amplitude at each position in the sweet potato seedling growth state characteristic gradient amplitude distribution map to obtain a local significant distribution map of the sweet potato seedling growth state characteristic gradient amplitude; Inputting the local significant distribution map of the characteristic gradient amplitude of the sweet potato seedling growth state into a gated masker based on the GELU function to obtain a local significant gated mask map of the gradient amplitude; Calculate the point-by-point multiplication between the gradient amplitude local significant gated mask map and the sweet potato seedling growth state characteristic map to obtain the sweet potato seedling growth state significant characteristic map; Calculating the local description operator of the gradient amplitude at each position in the sweet potato seedling growth state characteristic gradient amplitude distribution map to obtain the local significant distribution map of the sweet potato seedling growth state characteristic gradient amplitude, including: Determine the scale of the local neighborhood, calculate the mean of the difference between the gradient amplitude value at the predetermined position in the sweet potato seedling growth state characteristic gradient amplitude distribution map and the gradient amplitude values ​​at other positions in the local neighborhood to obtain a gradient amplitude local description operator corresponding to the predetermined position; Based on the sweet potato seedling growth state significant feature map, soil fertility reasoning is performed to obtain a soil adaptive fertility reasoning semantic feature vector, including: Inputting the sweet potato seedling growth state significant feature map into a decoder-based soil fertility reasoner to obtain the soil adaptive fertility reasoning semantic feature vector; Extracting soil fertility state features of the soil state image to obtain a real-time soil fertility state semantic feature vector includes: Inputting the soil state image into a soil fertility state feature extractor based on a second atrous convolutional neural network model to obtain a real-time soil fertility state semantic feature vector; Based on the feature difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility state semantic feature vector, a decoded value of the recommended fertilization amount is generated, including: Calculating the position difference between the soil adaptive fertility reasoning semantic feature vector and the soil real-time fertility state semantic feature vector to obtain a soil fertility deficiency state semantic feature vector; The semantic feature vector of the soil fertility deficiency state is input into a decoder-based intelligent fertilization controller to obtain a decoded value of the recommended fertilization amount.