An intelligent control method, system and device for edible mushroom production and a medium

Abstract

The embodiment of the specification provides an edible mushroom production intelligent control method, system, device and medium, the method comprises: obtaining light and temperature information of the growth period of edible mushrooms, combining preset target output result information, determining a first sequence and a second sequence through a decision model, wherein the first sequence corresponds to the humidity change of the growth period, and the second sequence corresponds to the carbon dioxide concentration change of the growth period, and the light and temperature information is determined based on natural environment information and energy consumption control information; based on the preset target output result information, the image information of the growth period of the edible mushrooms, the actual light information, the actual temperature information and the second parameter information, the first sequence and the second sequence are updated to determine the updated first sequence and the updated second sequence, wherein the second parameter information comprises planting information of the edible mushrooms; based on the updated first sequence and the updated second sequence, the growth environment parameters of the edible mushrooms are controlled.

Description

[0001] Case Analysis

[0002] This application is a divisional application of Chinese application filed on May 23, 2022, with application number 202210561437.7, entitled "A method, system, apparatus and medium for obtaining edible fungi decision schemes". Technical Field

[0003] This manual relates to the field of agricultural informatization, and in particular to an intelligent control method, system, device and medium for edible fungi production. Background Technology

[0004] Edible fungi are highly sensitive to their growth environment; even minor changes can significantly impact their quality. Generally, a database of edible fungi cultivation techniques is established by combining expert research findings with practical cultivation experience to optimize production management. While this allows for a degree of automatic adjustment of the growth environment, its regulatory function is limited in the face of complex and ever-changing real-world conditions, making it difficult to achieve stable, large-scale production of edible fungi.

[0005] Therefore, it is necessary to propose an intelligent control method for edible fungi production to provide a suitable growth environment for edible fungi in real time and reduce the difficulty of mass-producing edible fungi products with stable quality. Summary of the Invention

[0006] This specification provides one or more embodiments of an intelligent control method for edible fungi production. The intelligent control method for edible fungi production includes: acquiring first parameter information of the edible fungi's growth period, the first parameter information including light information and temperature information; inputting the preset target output result information of the edible fungi and the first parameter information into a decision model, the decision model being a machine learning model; the decision model including at least an environmental feature extraction layer, a target feature extraction layer, and an output layer, the output layer including a first output layer and a second output layer; processing the first parameter information through the environmental feature extraction layer, the output temperature feature vector and light feature vector being used as inputs to the first output layer and the second output layer, respectively; processing the preset target output result information through the target feature extraction layer, the output target feature vector being used as inputs to the first output layer and the second output layer. The system describes the inputs to the first output layer and the second output layer; it outputs a first sequence and a second sequence through the first output layer and the second output layer, respectively; the first sequence reflects the humidity change information during the growth period of the edible fungus, and the second sequence reflects the carbon dioxide concentration change information during the growth period of the edible fungus; based on the preset target output result information, the image information of the edible fungus growth period, the actual light information, the actual temperature information, and the second parameter information, it updates the first sequence and the second sequence to determine the updated first sequence and the updated second sequence, wherein the second parameter information includes the planting information of the edible fungus; and it controls the growth environment parameters of the edible fungus based on the updated first sequence and the updated second sequence.

[0007] This specification provides one or more embodiments of an intelligent control system for edible fungi production. The system includes an acquisition module, an output module, and a control module. The acquisition module acquires first parameter information related to the growth period of the edible fungi, including light and temperature information. The output module inputs the preset target output result information of the edible fungi and the first parameter information into a decision model, which is a machine learning model. The decision model includes at least an environmental feature extraction layer, a target feature extraction layer, and an output layer. The output layer includes a first output layer and a second output layer. The environmental feature extraction layer processes the first parameter information, and the output temperature feature vector and light feature vector are used as inputs to the first output layer and the second output layer, respectively. The target feature extraction layer processes the preset target output result information. The target output result information and the output target feature vector are used as inputs to the first output layer and the second output layer; the first output layer and the second output layer respectively output a first sequence and a second sequence; the first sequence reflects the humidity change information during the growth period of the edible fungus, and the second sequence reflects the carbon dioxide concentration change information during the growth period of the edible fungus; the control module is used to update the first sequence and the second sequence based on the preset target output result information, the image information of the edible fungus growth period, the actual light information, the actual temperature information, and the second parameter information, to determine the updated first sequence and the updated second sequence, the second parameter information including the planting information of the edible fungus; and to control the growth environment parameters of the edible fungus based on the updated first sequence and the updated second sequence.

[0008] This specification provides one or more embodiments of an intelligent control device for edible fungi production, the device comprising at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute an intelligent control method for edible fungi production.

[0009] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes an intelligent control method for edible fungi production. Attached Figure Description

[0010] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0011] Figure 1 This is a schematic diagram illustrating the application scenario of a system for obtaining edible fungi decision-making schemes, as shown in some embodiments of this specification.

[0012] Figure 2 This is an exemplary block diagram of a system for obtaining edible fungi decision-making schemes according to some embodiments of this specification;

[0013] Figure 3 This is an exemplary flowchart of a method for obtaining edible fungi decision-making schemes according to some embodiments of this specification;

[0014] Figure 4 This is an exemplary flowchart illustrating a method for updating a first sequence and a second sequence according to some embodiments of this specification;

[0015] Figure 5 This is an exemplary model structure diagram of obtaining a decision-making scheme model for edible fungi, as shown in some embodiments of this specification. Detailed Implementation

[0016] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0017] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0018] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0020] Figure 1 This is a schematic diagram illustrating the application scenario of a system for obtaining edible fungi decision-making schemes, based on some embodiments of this specification.

[0021] like Figure 1 As shown, in some embodiments, the application scenario 100 of the system for obtaining edible fungi decision-making schemes may include an edible fungi cultivation room 110, a data acquisition device 120, a network 130, a processor 140, and a storage device 150. The system for obtaining edible fungi decision-making schemes can be used to determine suitable environmental parameters for edible fungi growth.

[0022] The edible mushroom cultivation room 110 can be used for the cultivation of edible mushrooms. The edible mushroom cultivation room 110 also includes, but is not limited to, edible mushrooms 111 and an environmental control device 112. In some embodiments, the environmental control device may include, but is not limited to, heating / cooling equipment, shading / supplementary lighting equipment, and gas regulation devices. For example, the edible mushroom cultivation room 110 can receive adjustment parameter instructions from the processor 140 to adjust the parameters of the environmental control device 112. For example, the edible mushroom cultivation room 110 can receive data output from the environmental control device 112 and send the data to the processor 140 and the storage device 150. In some embodiments, the edible mushroom cultivation room 110 is also referred to as a mushroom greenhouse, an edible mushroom production greenhouse, etc.

[0023] The data acquisition device 120 can be used to collect data. The data acquisition device 120 may include, but is not limited to, a camera 120-1, a video camera 120-2, a temperature acquisition device 120-3, a light acquisition device 120-4, a gas detection device, a humidity detection device, etc. In some embodiments, the camera 120-1 and the video camera 120-2 can acquire images containing edible fungi 111. In some embodiments, the camera 120-1 and the video camera 120-2 can acquire images in various feasible ways, including but not limited to continuous acquisition, timed acquisition, etc. In some embodiments, there may be multiple cameras 120-1 and video cameras 120-2, and they can be placed at different locations around the same target object to simultaneously acquire information about the target object from different angles. In some embodiments, the temperature acquisition device 120-3 may include devices such as a thermometer, a temperature sensor, a resistance temperature detector, or a thermocouple. In some embodiments, the light acquisition device 120-4 may include a light sensor, a light detector, a light detector, a photodiode, etc. In some embodiments, the data acquisition device 120 can collect data such as images, temperature, and light within the edible fungi cultivation room 110.

[0024] Network 130 can connect various components of the system and / or connect the system to external resources. Network 130 enables communication between the components and with other parts outside the system. For example, processor 140 obtains information and / or instructions from storage device 150 and mushroom cultivation room 110 via network 130.

[0025] Processor 140 can process data and / or information from at least one component of the system or an external data source. For example, processor 140 can acquire data collected by data acquisition device 120, process the acquired data, and extract information from the data. In some embodiments, processor 140 can be local or remote. For example, processor 140 can acquire information and / or data from storage device 150, edible mushroom cultivation room 110, and data acquisition device 120 via wired or wireless means. In some embodiments, processor 140 can be implemented on a cloud platform.

[0026] Storage device 150 can be used to store data and / or instructions. For example, storage device 150 can store data output from edible mushroom cultivation room 110. As another example, storage device 150 can store data acquired by data acquisition device 120. Storage device 150 may include one or more storage components, each of which may be a separate device or part of other devices. In some embodiments, storage device 150 may be implemented on a cloud platform.

[0027] It should be noted that the above description of the system and its components is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various components or construct subsystems connected to other components without departing from these principles. For example, the school service platform and the school management platform can be integrated into one component. As another example, the various components can share a storage device, or each component can have its own separate storage device. Such variations are all within the scope of this specification.

[0028] Figure 2 This is a schematic diagram of the modules of a system 200 for obtaining edible fungi decision-making schemes according to some embodiments of this specification.

[0029] In some embodiments, such as Figure 2 As shown, the system 200 for obtaining edible fungi decision-making schemes may include an acquisition module 210, an output module 220, and a determination module 230.

[0030] The acquisition module 210 is used to acquire first parameter information about the growth period of edible fungi, including light and temperature information. For definitions and explanations of relevant terms, as well as the acquisition method, please refer to [link to documentation]. Figure 3 and Figure 4 And its related descriptions.

[0031] The output module 220 is used to input the preset target output result information of edible fungi and the first parameter information into the decision model, and output a first sequence and a second sequence. The first sequence reflects the humidity change information during the growth period of edible fungi, and the second sequence reflects the carbon dioxide concentration change information during the growth period of edible fungi. The first and second sequences are updated through the following methods: acquiring image information of the edible fungi growth period; determining the process quality information of edible fungi based on the image information; acquiring actual light and actual temperature information; determining the updated first parameter information based on the actual light and actual temperature information; inputting the preset target output result information, process quality information, updated first parameter information, and second parameter information into the decision model, and outputting the updated first and second sequences, wherein the second parameter information includes the cultivation information of edible fungi. The update frequency of the first and second sequences is determined through the following method: inputting third parameter information into the quality model, outputting the predicted appearance information and predicted harvestability information of edible fungi, wherein the third parameter information includes at least: the updated first and second sequences, and the second parameter information; performing weighted fusion based on the predicted appearance information and predicted harvestability information to determine a quality factor; and determining the update frequency of the first and second sequences based on the quality factor. For further explanation of the relevant terms, please see Figure 3 For a description of the decision-making model and related information, please refer to [link / reference]. Figure 5 And its related description. For the methods and update frequencies for updating the first and second sequences, see [link to relevant documentation]. Figure 4 And its related descriptions.

[0032] The determining module 230 is used to control the growth environment parameters of edible fungi based on the first and second sequences. For further explanation of the control method, see [link to relevant documentation]. Figure 3 And its related descriptions.

[0033] It should be noted that the above description of the system and its modules for obtaining edible fungi decision-making schemes is for convenience only and should not limit this specification to the scope of the illustrated embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2 The acquisition module, output module, and determination module disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0034] Figure 3This is an exemplary flowchart of a method for obtaining a decision-making scheme for edible fungi, according to some embodiments of this specification. In some embodiments, process 300 may be executed by processor 140. Figure 3 As shown, process 300 includes the following steps.

[0035] Step 310: Obtain the first parameter information for the growth period of edible fungi. The first parameter information includes light information and temperature information, which are determined based on the natural environment information and energy consumption control information of edible fungi.

[0036] The growth period refers to the time it takes for edible fungi to grow from sprouting to maturity. The primary parameter information refers to the parameters required for the growth of edible fungi during this growth period. These parameters may include one or more of the following: temperature, humidity, light, and carbon dioxide concentration.

[0037] In some embodiments, the first parameter information may include light information and temperature information. The light information may be a sequence of light data combining light data from multiple consecutive time points during the growth period of the edible fungus. The temperature information may be a sequence of temperature data combining temperature data from multiple consecutive time points during the growth period of the edible fungus. In some embodiments, the first parameter information may also include other parameter information.

[0038] In some embodiments, the first parameter information can be obtained by those skilled in the art based on experience and actual conditions. For example, if planting experience tells us that the ideal conditions for the growth of a certain mycelium are a temperature of 23°C to 25°C, a humidity of 65% to 75%, an optimal light intensity of 300 lx, and a carbon dioxide concentration below 0.1%, then the relevant data in the first parameter information for the growth of this mycelium will be set in the corresponding ranges.

[0039] In some embodiments, the first parameter information may also be updated first parameter information. For details on how to obtain the updated first parameter information, please refer to the following text. Figure 4 Step 440 and its description.

[0040] Step 320: Input the preset target output result information of edible fungi and the first parameter information into the decision model, and output a first sequence and a second sequence. The first sequence reflects the humidity change information during the growth period of edible fungi, and the second sequence reflects the carbon dioxide concentration change information during the growth period of edible fungi.

[0041] The preset target output result information refers to the preset target output result information that edible fungi need to achieve. In some embodiments, the preset target output result information may include one or more of the following: target quality of edible fungi, target cycle of edible fungi, and target yield rate of edible fungi. Quality may include one or more of the following: size, height, color, and presence of rotten or moldy parts of edible fungi. Cycle may include the time spent on one or more processes from cultivation to harvesting of edible fungi. Yield rate may refer to the yield of edible fungi per unit area.

[0042] The first sequence can reflect humidity changes during the growth period of edible fungi. For example, the first sequence can be a time-based humidity data sequence during the growth period of edible fungi.

[0043] The second sequence reflects information on changes in carbon dioxide concentration during the growth period of edible fungi. For example, the second sequence can be a time-based carbon dioxide concentration data sequence during the growth period of edible fungi.

[0044] In some embodiments, the processor 140 can input the preset target output result information of edible fungi and the first parameter information into the decision model, and output a first sequence and a second sequence. For the definition of the decision model and how to obtain it, please refer to the following text. Figure 5 And its related descriptions.

[0045] Step 330: Control the growth environment parameters of edible fungi based on the first sequence and the second sequence.

[0046] Growth environment parameters refer to the various environmental parameters required for the cultivation of edible fungi. For example, growth environment parameters may include humidity and carbon dioxide concentration during the growth of edible fungi.

[0047] In some embodiments, based on the first sequence and the second sequence, if the current actual humidity data is detected to be different from the humidity data at the corresponding time point in the first sequence and / or the current actual carbon dioxide concentration is detected to be different from the carbon dioxide concentration data at the corresponding time point in the second sequence, the processor 140 can use an adjustment device to adjust the current humidity and carbon dioxide concentration to the values ​​in the corresponding sequences. The adjustment device may include a humidifier, a dryer, a ventilation device, etc.

[0048] In some embodiments, if the current actual humidity data is detected to be different from the humidity data at the corresponding time point in the first sequence and / or the current actual carbon dioxide concentration is detected to be different from the carbon dioxide concentration data at the corresponding time point in the second sequence, the processor 140 may also send a warning message to the user terminal to remind the user whether it is necessary to adjust the growth environment parameters of the edible fungi.

[0049] In some embodiments of this specification, edible fungi that meet the requirements are obtained by controlling and adjusting the growth environment parameters during the growth period of edible fungi; and the management optimization of edible fungi production is achieved by intelligently monitoring the cultivation and production of edible fungi.

[0050] Figure 4 This is an exemplary flowchart illustrating a method for updating a first sequence and a second sequence according to some embodiments of this specification. In some embodiments, process 400 may be executed by processor 140. Figure 4 As shown, process 400 includes the following steps.

[0051] Step 410: Obtain image information of the edible fungi during their growth period.

[0052] Image information can refer to images that reflect the growth process of edible fungi. For example, image information can include images or videos taken during the growth period of edible fungi.

[0053] In some embodiments, images or videos of the edible fungi's growth period can be acquired by the data acquisition device 120.

[0054] Step 420: Determine the process quality information of edible fungi based on image information.

[0055] Process quality information refers to quality information that reflects the growth process of edible fungi. For example, process quality information may include the size, height, color, and presence of rotten or moldy parts of the fungi.

[0056] In some embodiments, the processor 140 can perform image recognition on an image of edible fungi during their growth period to identify the edible fungi in the image. The image recognition method may include, but is not limited to, one or more of computer image recognition methods, structural image recognition methods, and fuzzy image recognition methods. In some embodiments, the processor 140 can extract image features from the portion of the image containing the edible fungi within a recognition box to obtain process quality information of the edible fungi. The image feature extraction method may include, but is not limited to, one or more of grayscale feature extraction methods and texture feature extraction methods.

[0057] Step 430: Obtain actual illumination information and actual temperature information.

[0058] Actual light information refers to the actual light data at various points in time during the growth period of edible fungi. For example, actual light information can be a sequence of actual light data composed of multiple consecutive time points during the growth period of edible fungi.

[0059] Actual temperature information refers to the actual environmental temperature data at various points in time during the growth period of edible fungi. For example, actual temperature information can be a sequence of actual environmental temperature data composed of multiple consecutive time points during the growth period of edible fungi.

[0060] In some embodiments, actual illumination data can be acquired by a light sensor, and actual ambient temperature data can be acquired by a temperature sensor.

[0061] Step 440: Based on the actual illumination information and actual temperature information, determine the updated first parameter information.

[0062] In some embodiments, before the start of edible fungi cultivation, the first parameter information of edible fungi throughout the entire growth period can be predicted based on experience and actual conditions.

[0063] In some embodiments, after the start of edible mushroom cultivation, when the actual light value and / or actual temperature value at a certain time point during the growth period of the edible mushrooms, measured by a sensor at preset time intervals, is different from the predicted light value and / or temperature value at the corresponding time point in the first parameter information (e.g., light data sequence and temperature data sequence), the processor 140 can replace the predicted light value and / or temperature value in the first parameter information with the corresponding actual light value and / or actual temperature value. For example, at the start of edible mushroom cultivation, all temperature data during the growth period are predicted data. If the actual temperature data is measured once a day and the predicted temperature data during the growth period of the edible mushrooms is (2,2,3,3,3,5,5,5,8,8,12,12,12), and on the 5th day, the actual temperature data for the previous 5 days is (3,3,4,4,4), then the updated temperature data sequence is (3,3,4,4,4,5,5,5,8,8,12,12,12). Since edible fungi typically require a reasonable range of light and ambient temperature data during their growth period, when the actual light and ambient temperature data for edible fungi fall outside this range, it is advisable to control energy consumption within a preset range while simultaneously controlling the actual light and ambient temperature data of the edible fungi production and cultivation environment within a reasonable range. In some embodiments, once edible fungi cultivation begins, the first parameter information can be determined based on the natural environment information and energy consumption control information of the edible fungi. Natural environment information refers to the actual natural environment information existing during the growth period of the edible fungi. In some embodiments, natural environment information may include actual light and actual temperature information.

[0064] Energy consumption control information refers to the additional energy consumption required to modify the natural environment of edible fungi to a reasonable range during their growth period. For example, energy consumption control information can be a preset energy consumption threshold.

[0065] In some embodiments, when it is detected that the replaced actual light and actual ambient temperature data at a certain point in time in the first parameter information of edible fungi are not within the reasonable range of light and ambient temperature data required during the growth period of edible fungi, the processor 140 can control the supplemental lighting / shading device and the heating / cooling device to adjust within a preset energy consumption threshold range, reasonably supplementing / shading the light and heating / cooling the temperature to preset condition values, thereby adjusting the natural environment information during the growth period of edible fungi to the light and temperature data required during the growth period of edible fungi, as the first parameter information at that point in time. The preset condition values ​​refer to preset growth environment parameter values ​​suitable for the growth period of edible fungi, such as suitable temperature values ​​or suitable light intensity values. For example, the suitable light intensity and temperature for a certain stage of the growth period of edible fungi are weak light and 5℃-12℃. However, the current natural light intensity is moderately strong and the temperature is 15℃. Therefore, it is necessary to use shading devices and cooling equipment to appropriately reduce the current natural light intensity and temperature while meeting the preset energy consumption threshold. Then, the appropriately reduced natural light intensity and temperature are used as the first parameter information for that stage of the growth period of edible fungi.

[0066] In some embodiments, if it is not possible to simultaneously adjust multiple parameter information (e.g., temperature, humidity, light intensity, and carbon dioxide concentration) to the preset condition values ​​of multiple parameters during the growth period of edible fungi within a preset energy consumption threshold range, then it is preferable to adjust the actual light intensity data and actual temperature data during the growth period of edible fungi to the preset condition values.

[0067] Step 450: Input process quality information, preset target output result information, updated first parameter information and second parameter information into the decision model, and output the updated first sequence and second sequence, wherein the second parameter information includes edible fungi cultivation information.

[0068] In some embodiments, the second parameter information may include cultivation information for edible fungi. Cultivation information refers to information related to the cultivation of edible fungi. For example, cultivation information may include weather factors, culture medium quality, culture medium moisture content, edible fungi species, and edible fungi strain quality. For example, weather factors refer to weather-related factors such as temperature, air pressure, humidity, wind, clouds, and rain.

[0069] In some embodiments, the updated first parameter information can be input into the environmental feature extraction layer of the decision model, and the environmental feature extraction layer outputs temperature, illumination temperature, and illumination feature vector. In some embodiments, the preset target output result information can be input into the target feature extraction layer of the decision model, and the target feature extraction layer outputs a target feature vector. In some embodiments, process quality information and second parameter information can be input into the embedding layer of the decision model, and the embedding layer outputs a process feature vector. In some embodiments, temperature, illumination temperature, illumination feature vector, and process feature vector can be input into the first output layer of the decision model, and an updated first sequence can be output. In some embodiments, the target feature vector and process feature vector can be input into the second output layer of the decision model, and an updated second sequence can be output. For a more detailed description of the decision model, embedding layer, environmental feature extraction layer, target feature extraction layer, first output layer, second output layer, temperature, illumination feature vector, target feature vector, and process feature vector, please see below. Figure 5 And its related descriptions.

[0070] In some embodiments, the first sequence and the second sequence may be updated once or multiple times. Each update process is the process of repeating steps 410-450. The specific update frequency of the first sequence and the second sequence can be determined by the following steps 460-480.

[0071] Step 460: Input the third parameter information into the quality model and output the predicted appearance information and predicted harvestability information of edible fungi. The third parameter information includes at least: the updated first sequence and second sequence, and the second parameter information.

[0072] Predicted appearance information refers to the predicted appearance information of edible fungi. For example, predicted appearance information can be the predicted appearance grade of edible fungi. For example, the appearance grade of edible fungi can be premium appearance, grade one appearance, grade two appearance, etc. Predicted ease of harvesting information refers to the predicted ease of harvesting of edible fungi. For example, predicted ease of harvesting information can be easy to harvest or difficult to harvest, etc. In some embodiments, the third parameter information includes at least: the updated first sequence and second sequence, and the second parameter information.

[0073] In some embodiments, the updated first and second sequences and second parameter information can be input into the quality model, and the quality model outputs predicted appearance information and predicted harvestability information for the edible fungi. For a more detailed description of the quality model, please see below. Figure 5 And its related descriptions.

[0074] Step 470: Based on the predicted appearance information and the predicted ease of harvesting information, a weighted fusion is performed to determine the quality factor.

[0075] Quality factors can refer to factors that evaluate the quality of edible fungi.

[0076] In some embodiments, the processor 140 may determine a quality factor by weighted fusion of predicted appearance information and predicted pickability information.

[0077] In some embodiments, the predicted quality information and the predicted ease of picking information may be preset values ​​based on experience by those skilled in the art.

[0078] In some embodiments, the weight of the predicted appearance information can be preset according to the appearance grade of the edible fungi. For example, the appearance grades of edible fungi can be divided into premium grade and grade one. If the predicted appearance information indicates that the appearance grade of the edible fungi is premium grade, the weight of the predicted appearance information can be set to 0.8; if the predicted appearance information indicates that the appearance grade of the edible fungi is grade one, the weight of the predicted appearance information can be set to 0.7.

[0079] In some embodiments, the weights of predicted quality information and predicted ease of harvesting information are also related to the accuracy of the decision model. In some embodiments, the higher the prediction accuracy of the decision model for a certain parameter (e.g., humidity, carbon dioxide concentration), the greater the weight of the predicted quality information or predicted ease of harvesting information corresponding to that parameter. In some embodiments, prediction accuracy can be calculated by comparing the predicted parameter (e.g., humidity, carbon dioxide concentration) with the actual humidity or carbon dioxide concentration, and calculating the verification accuracy rate. The higher the verification accuracy rate calculated for a certain parameter, the higher the prediction accuracy. For example, if the carbon dioxide concentration and humidity predicted by the decision model are compared with the actual carbon dioxide and humidity, and the verification accuracy rate for carbon dioxide is higher than that for humidity, and since carbon dioxide concentration affects predicted quality information, then the weight of the predicted quality information corresponding to carbon dioxide concentration should be set higher than the weight of predicted ease of harvesting information.

[0080] In some embodiments of this specification, a quality factor is determined by correlating the weights of predicted quality information and predicted ease of picking information with the accuracy of the decision model. The update frequency of the first sequence and the second sequence is then determined by the quality factor, making the determination of the update frequency of the first sequence and the second sequence more based on practical evidence and more accurate.

[0081] Step 480: Determine the frequency of updating the first and second sequences based on the quality factor.

[0082] Since a higher quality factor indicates higher requirements for the production quality of edible fungi, it is even more necessary to monitor the primary parameter information of the edible fungi in real time during the production and cultivation process to ensure that the conditions are met. In some embodiments, a higher quality factor value results in a higher frequency of updating the first and second sequences. For example, if the quality factor value is 8.0, the frequency of updating the first and second sequences can be 3 times; if the quality factor value is 8.5, the frequency of updating the first and second sequences can be 4 times.

[0083] In some embodiments of this specification, quality factors are determined by utilizing the predicted appearance information and predicted harvestability information obtained from the quality model. Then, the update frequency of the first sequence and the second sequence is determined based on the magnitude of the quality factors. Thus, according to the update frequency requirements, the first parameter information is continuously adjusted based on the actual light and temperature information. This information is then input into the decision model to predict the first and second sequences, thereby making the decision model prediction more accurate and in line with the actual growth environment parameters required for edible fungi.

[0084] It should be noted that the above description of process 400 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 400 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification. For example, only steps 410-450 of process 400 may be performed.

[0085] Figure 5 This is an exemplary model structure diagram of obtaining a decision-making scheme model for edible fungi, as shown in some embodiments of this specification.

[0086] In some embodiments, obtaining the edible fungi decision-making scheme model 500 may include a decision model and a quality model.

[0087] The decision model can be used to obtain a first sequence and a second sequence. In some embodiments, the decision model may include at least an environmental feature extraction layer, a target feature extraction layer, and an output layer. For example, the decision model, the environmental feature extraction layer, the target feature extraction layer, and the output layer may include models obtained by convolutional neural networks (CNNs) or deep neural networks (DNNs) or combinations thereof.

[0088] In some embodiments, the input to the environmental feature extraction layer may include first parameter information, and its output may include a temperature feature vector and a light feature vector. In some embodiments, the temperature feature vector refers to a feature vector that reflects the temperature at various time points during the growth period of edible fungi, and the light feature vector refers to a feature vector that reflects the light intensity at various time points during the growth period of edible fungi. In some embodiments, the input to the target feature extraction layer may include preset target output result information, and its output may include a target feature vector. In some embodiments, the target feature vector refers to a feature vector that reflects the preset target output result information of edible fungi. For example, the target feature vector may include one or more of the following: a target quality feature vector, a target periodicity feature vector, and a target yield rate feature vector. In some embodiments, the target quality feature vector refers to a feature vector that reflects the target quality of edible fungi. In some embodiments, the target periodicity feature vector refers to a feature vector that reflects the target periodicity of edible fungi. In some embodiments, the target yield rate feature vector refers to a feature vector that reflects the target quality of edible fungi. In some embodiments, the decision model may also include an embedding layer. In some embodiments, the input to the embedding layer may include process quality information and second parameter information, and its output may include a process feature vector. In some embodiments, the process feature vector may represent the relationship between the process quality information of edible fungi and the preset target output result information. For example, if the process quality information of edible fungi includes the presence of rotten or moldy parts, and the preset target parameter result information includes the absence of rotten or moldy parts, then the process feature vector indicates that the process quality information of edible fungi does not match the preset target output result information.

[0089] In some embodiments, the output layer may include a first output layer and a second output layer. In some embodiments, the inputs to both the first and second output layers include a temperature feature vector, an illumination feature vector, a target feature vector, and a process feature vector. In some embodiments, the output of the first output layer may be a first sequence. In some embodiments, the output of the first output layer may also be an updated first sequence. In some embodiments, the output of the second output layer may be a second sequence. In some embodiments, the output of the second output layer may also be an updated second sequence.

[0090] In some embodiments, the decision model can be acquired through training, which can be performed by a processing device. For example, the decision model can be trained separately based on an environment feature extraction layer, a target feature extraction layer, an output layer, and an embedding layer. In some embodiments, the decision model can be obtained by acquiring multiple training samples and training them, wherein the training samples include sample first parameter information and their labels, and the labels represent a first sequence and a second sequence corresponding to the sample first parameter information.

[0091] In some embodiments, the environmental feature extraction layer, target feature extraction layer, output layer, and embedding layer can be acquired based on training. The training of the environmental feature extraction layer, target feature extraction layer, output layer, and embedding layer can be performed by a processing device, and the training can be implemented based on the following methods.

[0092] In some embodiments, at least one training sample and an initial environment feature extraction layer, an initial target feature extraction layer, an initial output layer, and an initial embedding layer are obtained. The training sample corresponding to the initial environment feature extraction layer may include sample first parameter information and its corresponding illumination feature vector and temperature feature vector, wherein the sample first parameter information may include sample illumination information and sample temperature information. The training sample corresponding to the initial target feature extraction layer may include preset target output result information and its corresponding target feature vector. The training sample corresponding to the initial embedding layer may include sample process quality information and sample second parameter information, as well as the corresponding process feature vector. The training sample corresponding to the initial output layer may include sample temperature feature vector, sample illumination feature vector, sample target feature vector, and sample process feature vector, as well as the corresponding sample first sequence and sample second sequence, or the corresponding updated first sequence and second sequence.

[0093] In some embodiments, a data acquisition device can be used to acquire light and temperature information.

[0094] In some embodiments, relevant information in the training samples can be obtained through manual annotation, or data with existing annotation information can be used as training samples to eliminate the need for manual annotation.

[0095] In some embodiments, the parameters of the initial environment feature extraction layer, the initial target feature extraction layer, the initial output layer, and the initial embedding layer are iteratively updated based on at least one training sample to obtain the environment feature extraction layer, the target feature extraction layer, the output layer, and the embedding layer.

[0096] In some embodiments, training samples can be input into the environment feature extraction layer, target feature extraction layer, output layer, and embedding layer. Their parameters are updated iteratively through training until the trained environment feature extraction layer, target feature extraction layer, output layer, and embedding layer meet preset conditions, thus obtaining the trained environment feature extraction layer, target feature extraction layer, output layer, and embedding layer. The preset conditions may be that the loss function is less than a threshold, convergence, or the training cycle reaches a threshold. In some embodiments, the method for iteratively updating the model parameters may include conventional model training methods such as stochastic gradient descent.

[0097] Quality models can be used to obtain predictive information on the appearance and harvestability of edible fungi. For example, quality models can include models derived from convolutional neural networks, deep neural networks, or combinations thereof.

[0098] In some embodiments, the input to the quality model may include third parameter information, and its output may include predicted appearance information and predicted harvestability information of edible fungi. For a description of the predicted appearance information and predicted harvestability information of edible fungi, please refer to [link to documentation / reference]. Figure 4 The details and related descriptions will not be repeated here.

[0099] In some embodiments, the quality model can be obtained based on training. Training of the quality model can be performed by a processing device. Training of the quality model can be achieved based on the following methods.

[0100] In some embodiments, at least one training sample and an initial quality model are obtained, wherein the training sample includes sample third parameter information labeled with predicted appearance information and predicted harvestability information. The initial quality model may be a quality model whose parameters have not been adjusted or has not yet met the training requirements. The number of training samples can be determined based on factors such as the accuracy requirements of the quality model and the actual application scenario.

[0101] In some embodiments, third parameter information can be obtained based on data from the edible fungi cultivation process.

[0102] In some embodiments, the annotation of the third parameter information regarding the predicted appearance information and the predicted ease of picking information can be obtained through manual annotation, or by obtaining the third parameter information of the already annotated predicted appearance information and predicted ease of picking information as a sample, thereby eliminating the manual annotation step.

[0103] In some embodiments, the parameters of the initial quality model are iteratively updated based on the at least one training sample to obtain the quality model.

[0104] In some embodiments, training samples can be input into the quality model, and the parameters of the initial quality model can be updated through training iterations until the trained quality model meets preset conditions, thus obtaining the trained quality model. For more details on the preset conditions, please refer to the decision model and its related descriptions, which will not be repeated here.

[0105] In some embodiments, the parameters of the decision model and the quality model can be obtained through joint training. Joint training can be performed by a processing device. Joint training can be implemented based on the following methods.

[0106] In some embodiments, the output of the decision model can be the input of the quality model, and the decision model and the quality model can be obtained through joint training. For example, training sample data, namely the sample's first parameter information and the sample's preset target output result information, are input into the decision model to obtain the updated first sequence and second sequence output by the decision model; then, the updated first sequence and second sequence are used as training sample data, and the second parameter information is input into the quality model to obtain the predicted quality information and predicted harvestability information output by the quality model; the sample quality information and sample harvestability information are used to verify the output of the quality model; using the backpropagation characteristics of the neural network model, the verification data of the updated first sequence and second sequence output by the decision model are obtained, and the verification data of the updated first sequence and second sequence are used as labels to train the decision model.

[0107] For example, the first parameter information and the preset target output result information sample are input into the initial decision model, and the third parameter information sample is input into the initial quality model. A loss function is constructed based on the label and the prediction result of the initial quality model, and the parameters of the initial decision model and the initial quality model are updated simultaneously until the trained decision model and quality model meet preset conditions, thus obtaining the trained decision model and quality model. The preset conditions can be that the loss function is less than a threshold, convergence, or the training period reaches a threshold. In some embodiments, the method for iteratively updating the model parameters can include conventional model training methods such as stochastic gradient descent.

[0108] When the decision model and quality model are trained jointly, the cost functions related to predicting appearance information and predicting harvestability information have different weights. These weights are related to the weights used in determining the quality factors. By setting different weights, the model can more effectively prioritize meeting certain requirements. For example, setting larger weights for the cost functions related to predicting appearance information can make the model more focused on producing edible fungi with higher appearance grades. For more information on weights, see [link to documentation]. Figure 4 The details and related descriptions will not be repeated here.

[0109] In some embodiments, a loss function F = α*A1 + β*B1 + C can be constructed, where α and β are weight values; A1 is the loss term corresponding to the appearance, determined based on the predicted appearance and label; B1 is the loss term corresponding to the harvestability, determined based on the predicted harvestability and label; and C is the regularization term. When the appearance is more important or the appearance grade is higher in the target quality, α is greater than β. Therefore, the predicted appearance value must be closer to the actual value than the predicted harvestability value for the function F to converge or equal to 0, thereby focusing on producing edible fungi with higher appearance grades.

[0110] In some embodiments of this specification, the decision model has a multi-layered structure, capable of simultaneously inputting and outputting multiple parameters and jointly training them, thereby improving prediction efficiency. Using the decision model and the quality model, production parameters for edible fungi can be determined based on a large amount of extensive data to meet the needs of complex and ever-changing actual production processes, improving the efficiency of acquiring edible fungi production data and reducing the amount of data processing. Furthermore, prediction based on machine learning technology can be based on even more and richer data analysis, resulting in higher accuracy in the predicted edible fungi production parameters.

[0111] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0112] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0113] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0114] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0115] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0116] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0117] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for intelligent control of edible fungi production, characterized in that, include: Obtain first parameter information of the growth period of the edible fungus, the first parameter information including light information and temperature information; The preset target output result information of the edible fungi and the first parameter information are input into the decision model. The decision model is a machine learning model. The decision model includes at least an environmental feature extraction layer, a target feature extraction layer and an output layer. The output layer includes a first output layer and a second output layer. The first parameter information is processed by the environmental feature extraction layer, and the output temperature feature vector and light feature vector are used as the inputs of the first output layer and the second output layer, respectively. The target output result information is processed by the target feature extraction layer, and the output target feature vector is used as the input of the first output layer and the second output layer. The first output layer and the second output layer respectively output a first sequence and a second sequence. The first sequence reflects the humidity change information during the growth period of the edible fungus, and the second sequence reflects the carbon dioxide concentration change information during the growth period of the edible fungus. Based on the preset target output result information, the image information of the edible fungus growth period, the actual light information, the actual temperature information, and the second parameter information, the first sequence and the second sequence are updated to determine the updated first sequence and the updated second sequence. The second parameter information includes the planting information of the edible fungus. as well as The growth environment parameters of the edible fungi are controlled based on the updated first sequence and the updated second sequence; wherein the update frequency of the first sequence and the second sequence is determined by the following method: The third parameter information is input into the quality model, and the predicted appearance information and predicted harvestability information of the edible fungi are output. The third parameter information includes at least the updated first sequence, the updated second sequence, and the second parameter information. The quality model is a machine learning model. The quality factor is determined by weighted fusion of the predicted appearance information and the predicted ease of harvesting information. The weight of the predicted appearance information is determined based on the appearance grade of edible fungi and the accuracy of the decision model, and the weight of the predicted ease of harvesting information is determined based on the accuracy of the decision model. The update frequencies of the first sequence and the second sequence are determined based on the quality factor.

2. The intelligent control method for edible fungi production as described in claim 1, characterized in that, The parameters of the decision model and the quality model are obtained through joint training, which includes: Input the first parameter information of the sample and the preset target output result information of the sample into the initial decision model to obtain the updated first sequence and the updated second sequence output by the initial decision model; The updated first sequence and the updated second sequence are used as training sample data for the initial quality model, and the sample second parameter information is input into the initial quality model to obtain the predicted quality information and the predicted ease of picking information output by the initial quality model. The predicted quality information and the predicted ease of picking information output by the initial quality model are verified using the sample quality information and the sample ease of picking information. By utilizing the backpropagation characteristics of the neural network model, validation data of the updated first sequence and the updated second sequence output by the initial decision model are obtained, and the validation data is used as labels to train the initial decision model.

3. The intelligent control method for edible fungi production as described in claim 2, characterized in that, During the joint training process: The loss function corresponding to the quality model includes a loss term corresponding to the appearance information of the edible fungi and a loss term corresponding to the ease of harvesting information of the edible fungi. The loss term corresponding to the appearance information is determined based on the prediction result of the appearance information and the label, and the loss term corresponding to the ease of harvesting information is determined based on the prediction result of the ease of harvesting information and the label. The loss term corresponding to the appearance information and the loss term corresponding to the ease of harvesting information have different weights.

4. An intelligent control system for edible fungi production, characterized in that, The system includes an acquisition module, an output module, and a control module; The acquisition module is used to acquire first parameter information of the growth period of the edible fungus, the first parameter information including light information and temperature information; The output module is used to input the preset target output result information of the edible fungus and the first parameter information into a decision model. The decision model is a machine learning model, which includes at least an environmental feature extraction layer, a target feature extraction layer, and an output layer. The output layer includes a first output layer and a second output layer. The first parameter information is processed by the environmental feature extraction layer, and the output temperature feature vector and light feature vector are used as inputs to the first output layer and the second output layer, respectively. The preset target output result information is processed by the target feature extraction layer, and the output target feature vector is used as inputs to the first output layer and the second output layer. The first output layer and the second output layer output a first sequence and a second sequence, respectively. The first sequence reflects the humidity change information during the growth period of the edible fungus, and the second sequence reflects the carbon dioxide concentration change information during the growth period of the edible fungus. The control module is used for: Based on the preset target output result information, the image information of the edible fungus growth period, the actual light information, the actual temperature information, and the second parameter information, the first sequence and the second sequence are updated to determine the updated first sequence and the updated second sequence. The second parameter information includes the planting information of the edible fungus. as well as The growth environment parameters of the edible fungi are controlled based on the updated first sequence and the updated second sequence; The control module is also used for: The third parameter information is input into the quality model, and the predicted appearance information and predicted harvestability information of the edible fungi are output. The third parameter information includes at least the updated first sequence, the updated second sequence, and the second parameter information. The quality model is a machine learning model. The quality factor is determined by weighted fusion of the predicted appearance information and the predicted ease of harvesting information. The weight of the predicted appearance information is determined based on the appearance grade of edible fungi and the accuracy of the decision model, and the weight of the predicted ease of harvesting information is determined based on the accuracy of the decision model. The update frequencies of the first sequence and the second sequence are determined based on the quality factor.

5. The intelligent control system for edible fungi production according to claim 4, characterized in that, The parameters of the decision model and the quality model are obtained through joint training, which includes: Input the first parameter information of the sample and the preset target output result information of the sample into the initial decision model to obtain the updated first sequence and the updated second sequence output by the initial decision model; The updated first sequence and the updated second sequence are used as training sample data for the initial quality model, and the sample second parameter information is input into the initial quality model to obtain the predicted quality information and the predicted ease of picking information output by the initial quality model. The predicted quality information and the predicted ease of picking information output by the initial quality model are verified using the sample quality information and the sample ease of picking information. By utilizing the backpropagation characteristics of the neural network model, validation data of the updated first sequence and the updated second sequence output by the initial decision model are obtained, and the validation data is used as labels to train the initial decision model.

6. The intelligent control system for edible fungi production according to claim 5, characterized in that, During the joint training process: The loss function corresponding to the quality model includes a loss term corresponding to the appearance information of the edible fungi and a loss term corresponding to the ease of harvesting information of the edible fungi. The loss term corresponding to the appearance information is determined based on the prediction result of the appearance information and the label, and the loss term corresponding to the ease of harvesting information is determined based on the prediction result of the ease of harvesting information and the label. The loss term corresponding to the appearance information and the loss term corresponding to the ease of harvesting information have different weights.

7. An intelligent control device for edible fungi production, characterized in that, The device includes at least one processor and at least one memory; the at least one memory is used to store computer instructions. The at least one processor is used to execute at least a portion of the computer instructions to implement the intelligent control method for edible fungi production as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the intelligent control method for edible fungi production as described in any one of claims 1 to 3.

Images