Rice processing control method and system based on artificial intelligence

Abstract

The invention provides a rice processing control method and system based on artificial intelligence, and relates to the technical field of rice processing.The method comprises the steps that a rice image in a winnowing machine is obtained; based on the rice image in the winnowing machine, a convolutional neural network model is used for processing to determine the initial wind power of the winnowing machine; performing winnowing cleaning on the rice based on the initial wind power of the winnowing machine, and obtaining a monitoring video of the rice under the initial wind power; based on the monitoring video of the rice under the initial wind power, using an information processing model to determine a plurality of candidate wind power of the rice; determining target wind power of the rice based on the monitoring video of the rice under the initial wind power and the multiple candidate wind power of the rice; and performing blowing screening on the rice based on the target wind power of the rice. The method can quickly and accurately determine the wind power in the winnowing machine.

Description

Technical Field

[0001] The present invention relates to the technical field of rice processing control, and particularly relates to an artificial intelligence-based rice processing control method and system. Background Art

[0002] As one of the main grains globally, the quality of rice directly affects food safety and consumer health. In the production process of rice, winnowing is an important step to remove impurities, stones, empty husks and other bad particles. By cleaning rice with a winnower, the quality of rice can be effectively improved to ensure that the final product meets market standards. However, traditional winnowing methods often rely on experienced operators to manually adjust the wind force. Due to the lack of precise measurement means, operators usually need to try many times to find the appropriate wind force setting, which is not only time-consuming and laborious, but may also lead to unstable quality of some batches of rice.

[0003] Therefore, how to quickly and accurately determine the wind force in the winnower is an urgent problem to be solved currently. Summary of the Invention

[0004] The main technical problem to be solved by the present invention is how to quickly and accurately determine the wind force in the winnower.

[0005] According to a first aspect, the present invention provides an artificial intelligence-based rice processing control method, including: acquiring a rice image in the winnower; using a convolutional neural network model to process the rice image in the winnower to determine an initial wind force of the winnower; performing winnowing cleaning on the rice based on the initial wind force of the winnower, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining a target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and performing blowing and screening on the rice based on the target wind force of the rice.

[0006] In a possible implementation, determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and multiple candidate wind forces of the rice includes: using a diffusion model to generate a simulated video of the rice under each candidate wind force based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice, and the similarity between the simulated video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; constructing a graph structure, where the graph structure includes multiple nodes and multiple edges between the nodes, the multiple nodes include an initial wind force node and multiple candidate wind force nodes, where the initial wind force node is the central node, each candidate wind force node is respectively connected to the initial wind force node to establish an edge, the node feature of the initial wind force node is the monitoring video of the rice under the initial wind force, the node feature of each candidate wind force node includes the simulated video of the rice under each candidate wind force, and the edge between the nodes is the similarity between the simulated video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on processing the graph structure by a graph autoencoder.

[0007] In a possible implementation, the information processing model is a Transformer model, the input of the information processing model is the monitoring video of the rice under the initial wind force, and the output of the information processing model is multiple candidate wind forces of the rice.

[0008] In a possible implementation, the input of the convolutional neural network model is the image of the rice in the air separator, and the output of the convolutional neural network model is the magnitude of the initial wind force of the air separator.

[0009] According to a second aspect, the present invention provides an artificial intelligence-based rice processing control system, including: A first acquisition module for acquiring an image of the rice in the air separator; An initial wind force magnitude determination module for determining the magnitude of the initial wind force of the air separator based on processing the image of the rice in the air separator using a convolutional neural network model; A second acquisition module for performing air separation cleaning on the rice based on the magnitude of the initial wind force of the air separator and acquiring a monitoring video of the rice under the initial wind force; An information processing module for determining multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force using an information processing model; A target wind force determination module for determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; A setting module for performing blowing and screening on the rice based on the target wind force of the rice.

[0010] In a possible implementation, the target wind force determination module is further configured to: Based on the monitoring video of the rice under the initial wind force, use a diffusion model for multiple candidate wind forces of the rice to generate simulation videos of the rice under each candidate wind force, and calculate the similarity between the simulation videos of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; Construct a graph structure, which includes multiple nodes and multiple edges between the nodes. The multiple nodes include an initial wind force node and multiple candidate wind force nodes. Among them, the initial wind force node is the central node, and each candidate wind force node is respectively connected to the initial wind force node by an edge. The node feature of the initial wind force node is the monitoring video of the rice under the initial wind force, and the node feature of each candidate wind force node includes the simulation video of the rice under each candidate wind force. The edge between the nodes is the similarity between the simulation video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; Based on the graph autoencoder, process the graph structure to determine the target wind force of the rice.

[0011] In a possible implementation, the information processing model is a Transformer model. The input of the information processing model is the monitoring video of the rice under the initial wind force, and the output of the information processing model is multiple candidate wind forces of the rice.

[0012] In a possible implementation, the input of the convolutional neural network model is the image of the rice in the air classifier, and the output of the convolutional neural network model is the magnitude of the initial wind force of the air classifier.

[0013] According to a third aspect, an embodiment of the present invention provides an electronic device, including: a processor; a memory; and a computer program; wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the method as described above. The method includes: acquiring an image of the rice in the air classifier; using a convolutional neural network model to process the image of the rice in the air classifier to determine the magnitude of the initial wind force of the air classifier; performing air separation and cleaning on the rice based on the magnitude of the initial wind force of the air classifier, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and performing blowing and screening on the rice based on the target wind force of the rice.

[0014] According to a fourth aspect, the present embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the aforementioned rice processing control method based on artificial intelligence. The method includes: acquiring a rice image in a winnower; using a convolutional neural network model to process the rice image in the winnower to determine the initial wind force magnitude of the winnower; performing winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and performing blowing and screening on the rice based on the target wind force of the rice.

[0015] A rice processing control method and system based on artificial intelligence provided by the present invention. The method includes acquiring a rice image in a winnower; using a convolutional neural network model to process the rice image in the winnower to determine the initial wind force magnitude of the winnower; performing winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and performing blowing and screening on the rice based on the target wind force of the rice. This method can quickly and accurately determine the wind force magnitude in the winnower. Description of the Drawings

[0016] Figure 1 It is a schematic diagram of an application scenario of a rice processing control method based on artificial intelligence provided by an embodiment of the present invention; Figure 2 It is a schematic flowchart of a rice processing control method based on artificial intelligence provided by an embodiment of the present invention; Figure 3 It is a schematic flowchart of determining the target wind force of rice provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of a rice processing control system based on artificial intelligence provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of an electronic device provided by an embodiment of the present invention; Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed Embodiments

[0017] The present invention will be further described in detail below in conjunction with the accompanying drawings through specific embodiments. Similar elements in different embodiments are denoted by related similar element numbers. In the following embodiments, many detailed descriptions are provided to enable a better understanding of the present invention. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other elements, materials, or methods. In some cases, some operations related to the present invention are not shown or described in the specification, in order to avoid overwhelming the core part of the present invention with excessive descriptions. For those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations based on the descriptions in the specification and the general technical knowledge in the art.

[0018] Figure 1 It is a schematic diagram of an application scenario of an artificial intelligence-based rice processing control method provided by an embodiment of the present invention. Figure 1 The application scenario of the artificial intelligence-based rice processing control method can include a server 11, a network 12, a terminal 13, and a storage device 14.

[0019] In some embodiments, the server 11 can be a single server or a server group. The server 11 can access the information and / or data stored in the terminal 13 or the storage device 14 through the network 12. In some embodiments, the server 11 can be used to execute Figure 2 the artificial intelligence-based rice processing control method shown in

[0020] The network 12 can facilitate the exchange of information and / or data. In some embodiments, the network 12 can be any form of wired or wireless network, or any combination thereof.

[0021] The terminal 13 can refer to one or more terminal devices used by a user. In some embodiments, the terminal 13 can include one or more combinations of mobile devices, tablet computers, laptop computers, etc.

[0022] The storage device 14 can store data and / or instructions. For example, the storage device 14 can store data instructions of the artificial intelligence-based rice processing control method.

[0023] In an embodiment of the present invention, there is provided an artificial intelligence-based rice processing control method as shown in Figure 2 The artificial intelligence-based rice processing control method includes steps S1 to S6: Step S1, obtaining a rice image in a winnowing machine; A winnowing machine is a mechanical device for removing impurities from rice. The winnowing machine can blow away the lighter impurities through air flow and leave the heavier rice grains.

[0024] The air classifier can include a vibrating screen and an air classifier.

[0025] The rice image refers to an image of rice grains and their surrounding environment captured by a camera or other imaging device installed on the air classifier.

[0026] Step S2: Based on the rice image in the air classifier, use a convolutional neural network model for processing to determine the initial wind force magnitude of the air classifier. The convolutional neural network model is a deep learning model that can be used for image recognition and classification tasks. The convolutional neural network model extracts image features through multiple convolutional layers and uses fully connected layers for classification or regression prediction.

[0027] The initial wind force magnitude is the basic wind force intensity set when the air classifier starts running, which is output by the convolutional neural network model.

[0028] The input of the convolutional neural network model is the rice image in the air classifier, and the output of the convolutional neural network model is the initial wind force magnitude of the air classifier.

[0029] Rice grains of different sizes have different requirements for wind force. Larger grains usually require stronger wind force to effectively separate impurities.

[0030] The rice image in the air classifier can show the types of impurities and their ratio to the rice. For example, light impurities (such as dust and empty husks) do not require high wind force to be effectively removed, while heavy impurities require too high wind force. As an example, if the image shows that the rice grains are relatively dense, higher wind force may be required to ensure sufficient air circulation for effective screening.

[0031] Step S3: Based on the initial wind force magnitude of the air classifier, perform air classification and cleaning on the rice, and obtain a monitoring video of the rice under the initial wind force. When the initial wind force magnitude of the air classifier is determined, a monitoring video of the rice under the initial wind force can be obtained.

[0032] Step S4: Based on the monitoring video of the rice under the initial wind force, use an information processing model to determine multiple candidate wind forces for the rice. The information processing model is a Transformer model. The input of the information processing model is the monitoring video of the rice under the initial wind force, and the output of the information processing model is multiple candidate wind forces for the rice. The Transformer model is a way of implementing artificial intelligence.

[0033] The multiple candidate wind forces for the rice are multiple alternative wind forces output by the information processing model.

[0034] The monitoring video of rice under the initial wind force provides dynamic information on the behavior of rice under specific wind force conditions. This information is crucial for the information processing model to evaluate the effectiveness of the current wind force setting and can help the information processing model identify more suitable wind force parameters.

[0035] As an example, the movement trajectories of rice grains of different sizes and weights in the air stream can reflect whether the current wind force is sufficient or too strong. For example, if heavier grains are also blown away, it may indicate that the wind force is too large; conversely, if impurities are not effectively removed, the wind force may need to be increased. The monitoring video of rice under the initial wind force can show the separation situation between rice and impurities. Good separation means that the wind force setting is close to the ideal value; while poor separation indicates that the wind force needs to be adjusted.

[0036] Transformer can efficiently process long sequence data, and the Transformer model can be used to capture the complex interaction relationships between rice grains at different moments and between the grains and the environment.

[0037] Through the self-attention mechanism, Transformer allows the model to focus on the relationships between different parts of the input sequence. In the analysis of the monitoring video, the Transformer model can be used to capture the complex interaction relationships between rice grains at different moments and between the grains and the environment.

[0038] The information processing model includes a rice information determination layer and a wind force determination layer. Both the rice information determination layer and the wind force determination layer include a Transformer structure. The input of the rice information determination layer is the monitoring video of the rice under the initial wind force, and the outputs of the rice information determination layer are the density change sequence of impurity particles, the density change sequence of rice grains, and the distribution uniformity sequence of rice grains. The input of the wind force determination layer is the density change sequence of impurity particles, the density change sequence of rice grains, and the distribution uniformity sequence of rice grains, and the output of the wind force determination layer is multiple candidate wind forces for the rice.

[0039] Step S5, determine the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; In some embodiments, Figure 3 It is a schematic flow diagram for determining the target wind force of rice provided by an embodiment of the present invention. The determination of the target wind force of the rice includes steps S21 to S23: Step S21, use a diffusion model based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice to generate a simulation video of the rice under each candidate wind force and the similarity between the simulation video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force.

[0040] Diffusion models are a type of generative model. Diffusion models generate new data samples by gradually adding noise to the data and then learning how to reverse this process. Diffusion models can be used to generate simulated videos based on surveillance videos under the initial wind force and multiple candidate wind forces.

[0041] The surveillance video provides the actual dynamic behavior of the rice under specific wind force conditions, and the diffusion model can infer the possible performance of the rice under other wind force conditions based on these behavior patterns. Due to the powerful generative ability of the diffusion model, it can generate simulated videos that are very close to the actual situation, thus improving the accuracy of predicting the effects of different wind force settings.

[0042] The surveillance video under the initial wind force contains the behavioral characteristics of the rice grains under the specific initial wind force conditions, including their movement trajectories, distribution densities, separation effects, etc. This information is the basis for the diffusion model to understand and simulate the behavior under different wind force conditions.

[0043] Different candidate wind forces represent possible adjustment options, each corresponding to a different air flow intensity, which will result in different movement patterns and separation effects of the rice grains.

[0044] In step S22, a graph structure is constructed. The graph structure includes multiple nodes and multiple edges between the nodes. The multiple nodes include an initial wind force node and multiple candidate wind force nodes, where the initial wind force node is the central node. Each candidate wind force node is respectively connected to the initial wind force node by an edge. The node feature of the initial wind force node is the surveillance video of the rice under the initial wind force. The node feature of each candidate wind force node includes the simulated video of the rice under each candidate wind force. The edge between the nodes is the similarity between the simulated video of the rice under each candidate wind force and the surveillance video of the rice under the initial wind force.

[0045] The graph structure is a form of data representation. The graph structure consists of nodes (Vertices or Nodes) and edges (Edges). Nodes represent entities or states, and edges represent the relationships between these entities. The graph structure can be used to represent the rice processing effects under different wind force conditions. The initial wind force node is the central node, and other nodes represent different candidate wind force settings.

[0046] The node feature of the initial wind force node is the surveillance video of the rice under the initial wind force. The surveillance video of the rice under the initial wind force records information such as the movement trajectory and separation effect of the rice grains under this wind force condition.

[0047] The candidate wind force node represents a specific candidate wind force setting. The node feature of the candidate wind force node includes the simulated video under the corresponding candidate wind force condition.

[0048] The edges between nodes represent the similarity between the simulated videos of rice under each candidate wind force and the monitored video of rice under the initial wind force.

[0049] Step S23: Process the graph structure based on the graph autoencoder to determine the target wind force of the rice.

[0050] The graph autoencoder can process based on the graph structure to determine the target wind force of the rice.

[0051] By organizing different wind force settings and their corresponding effects into a graph structure, it is possible to better capture the complex relationships between various wind force settings. For example, which wind force settings lead to similar separation effects and which settings have significant differences. The graph structure not only considers the effects of individual wind force settings but also comprehensively analyzes the interactions between multiple settings. Each node (the initial wind force node and candidate wind force nodes) contains the behavioral characteristics of rice grains under specific conditions (monitored video or simulated video), and the edges represent the similarity scores between these nodes. This method enables the graph autoencoder to quantitatively evaluate the similarities and differences between different wind force settings. By analyzing the relationships between nodes and edges through the graph autoencoder, it can help the graph autoencoder identify which wind force settings are closest to the ideal state, thus providing a scientific basis for selecting the optimal target wind force.

[0052] The graph autoencoder (GAE) is a deep learning model specifically designed to process graph-structured data. The graph autoencoder can automatically encode the information of nodes and edges in the graph structure, extract useful features from it, and then help make optimal decisions.

[0053] Through learning the graph structure, the graph autoencoder can identify which candidate wind force nodes are most similar to the initial wind force node (i.e., have a high similarity score). This indicates that these candidate wind force settings are closer to the current best practice in terms of effects. The graph autoencoder can not only consider the features of a single node but also combine the information of all nodes and edges in the entire graph structure to make a more comprehensive evaluation, thus recommending the target wind force setting most suitable for the current situation.

[0054] Step S6: Blow and screen the rice based on the target wind force of the rice.

[0055] The initial wind force of the rice set at the beginning is relatively small to facilitate determining the target wind force in subsequent processes. After determining the target wind force of the rice, the rice is then blown and screened based on the target wind force of the rice.

[0056] Based on the same inventive concept, Figure 4Schematic diagram of a rice processing control system based on artificial intelligence provided by an embodiment of the present invention. The rice processing control system based on artificial intelligence includes: A first acquisition module 41, configured to acquire a rice image in a winnower. An initial wind force magnitude determination module 42, configured to process the rice image in the winnower using a convolutional neural network model to determine the initial wind force magnitude of the winnower. A second acquisition module 43, configured to perform winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and acquire a monitoring video of the rice under the initial wind force. An information processing module 44, configured to use an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force. A target wind force determination module 45, configured to determine the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice. A setting module 46, configured to perform blowing and screening on the rice based on the target wind force of the rice.

[0057] Based on the same inventive concept, an embodiment of the present invention provides an electronic device, as Figure 5 shown, including: Including: a processor 51; a memory 52; and a computer program. Wherein, the computer program is stored in the memory 52 and is configured to be executed by the processor 51 to implement the rice processing control method based on artificial intelligence provided as described above. The method includes: acquiring a rice image in a winnower; processing the rice image in the winnower using a convolutional neural network model to determine the initial wind force magnitude of the winnower; performing winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; performing blowing and screening on the rice based on the target wind force of the rice.

[0058] Based on the same inventive concept, this embodiment provides a computer-readable storage medium with a computer program stored thereon. When the program is executed by a processor 51, it implements the aforementioned rice processing control method based on artificial intelligence. The method includes: acquiring a rice image in a winnower; using a convolutional neural network model to process the rice image in the winnower to determine the initial wind force magnitude of the winnower; performing winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and acquiring a monitoring video of the rice under the initial wind force; using an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and performing blowing and screening on the rice based on the target wind force of the rice.

[0059] The rice processing control method based on artificial intelligence provided in the embodiments of this application can be applied to electronic devices such as terminal devices (such as mobile phones), tablet computers, laptop computers, ultra-mobile personal computers (UMPCs), handheld computers, netbooks, personal digital assistants (PDAs), wearable devices (such as smart watches, smart glasses, or smart helmets, etc.), augmented reality (AR) / virtual reality (VR) devices, smart home devices, in-vehicle computers, etc. The embodiments of this application do not impose any restrictions on this.

[0060] Taking the mobile phone 100 as an example of the above-mentioned electronic device, Figure 6 shows a schematic structural diagram of the mobile phone 100.

[0061] As Figure 6 shown, the mobile phone 100 may include a processing module 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc.

[0062] The processing module 110 can be used to: obtain an image of rice in the winnower; use a convolutional neural network model to process the image of rice in the winnower to determine the initial wind force magnitude of the winnower; perform winnowing and cleaning on the rice based on the initial wind force magnitude of the winnower, and obtain a monitoring video of the rice under the initial wind force; use an information processing model to determine multiple candidate wind forces of the rice based on the monitoring video of the rice under the initial wind force; determine the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; and perform blowing and screening on the rice based on the target wind force of the rice.

[0063] Similarly, it should be noted that, in order to simplify the presentation disclosed in this specification and thus assist in the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of this specification, multiple features are sometimes grouped together in one embodiment, drawing, or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than those recited in the claims. In fact, the features of the embodiments are less than all the features of the individual embodiments disclosed above.

[0064] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example rather than a limitation, alternative configurations of the embodiments of this specification can be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An artificial intelligence-based rice processing control method, characterized in that, Including: Obtaining a rice image in an air separator; Processing the rice image in the air separator using a convolutional neural network model to determine the initial wind force magnitude of the air separator; Performing air separation cleaning on the rice based on the initial wind force magnitude of the air separator, and obtaining a monitoring video of the rice under the initial wind force; Determining multiple candidate wind forces of the rice using an information processing model based on the monitoring video of the rice under the initial wind force, where the information processing model includes a rice information determination layer and a wind force determination layer; Determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; Performing blowing and screening on the rice based on the target wind force of the rice.

2. The rice processing control method based on artificial intelligence according to claim 1, wherein The determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice includes: Generating a simulation video of the rice under each candidate wind force and the similarity between the simulation video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; Constructing a graph structure, where the graph structure includes multiple nodes and multiple edges between the nodes. The multiple nodes include an initial wind force node and multiple candidate wind force nodes. The initial wind force node is the central node, and each candidate wind force node is respectively connected to the initial wind force node to establish an edge. The node feature of the initial wind force node is the monitoring video of the rice under the initial wind force, and the node feature of each candidate wind force node includes the simulation video of the rice under each candidate wind force. The edge between the nodes is the similarity between the simulation video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; Processing the graph structure based on a graph autoencoder to determine the target wind force of the rice.

3. The artificial intelligence-based rice processing control method according to claim 1, wherein The information processing model is a Transformer model. The input of the information processing model is the monitoring video of the rice under the initial wind force, and the output of the information processing model is multiple candidate wind forces of the rice.

4. The rice processing control method based on artificial intelligence according to claim 1, wherein The input of the convolutional neural network model is the rice image in the air separator, and the output of the convolutional neural network model is the initial wind force magnitude of the air separator.

5. An artificial intelligence-based rice processing control system, characterized in that, Including: A first obtaining module for obtaining a rice image in an air separator; An initial wind force magnitude determination module for processing the rice image in the air separator using a convolutional neural network model to determine the initial wind force magnitude of the air separator; A second obtaining module for performing air separation cleaning on the rice based on the initial wind force magnitude of the air separator and obtaining a monitoring video of the rice under the initial wind force; An information processing module for determining multiple candidate wind forces of the rice using an information processing model based on the monitoring video of the rice under the initial wind force, where the information processing model includes a rice information determination layer and a wind force determination layer; A target wind force determination module for determining the target wind force of the rice based on the monitoring video of the rice under the initial wind force and the multiple candidate wind forces of the rice; A setting module for performing blowing and screening on the rice based on the target wind force of the rice.

6. The rice processing control system based on artificial intelligence according to claim 5, wherein The target wind force determination module is further used for: Generate simulation videos of the rice under each candidate wind force using a diffusion model based on the monitoring video of the rice under the initial wind force, and calculate the similarity between the simulation videos of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; Construct a graph structure, which includes multiple nodes and multiple edges between the nodes. The multiple nodes include an initial wind force node and multiple candidate wind force nodes. The initial wind force node is the central node, and each candidate wind force node is connected to the initial wind force node by an edge. The node feature of the initial wind force node is the monitoring video of the rice under the initial wind force, and the node feature of each candidate wind force node includes the simulation video of the rice under each candidate wind force. The edge between the nodes is the similarity between the simulation video of the rice under each candidate wind force and the monitoring video of the rice under the initial wind force; Determine the target wind force of the rice by processing the graph structure based on a graph autoencoder.

7. The rice processing control system based on artificial intelligence according to claim 5, characterized in that, The information processing model is a Transformer model. The input of the information processing model is the monitoring video of the rice under the initial wind force, and the output of the information processing model is multiple candidate wind forces of the rice.

8. The rice processing control system based on artificial intelligence according to claim 5, characterized in that, The input of the convolutional neural network model is the rice image in the winnowing machine, and the output of the convolutional neural network model is the magnitude of the initial wind force of the winnowing machine.

9. An electronic device, characterized in that, Comprising: A processor; A memory; And a computer program; wherein, the computer program is stored in the memory and configured to be executed by the processor to implement the artificial intelligence-based rice processing control method according to any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the artificial intelligence-based rice processing control method according to any one of claims 1 to 4.

Images