A kind of intelligent production supporting process of Cordyceps militaris

Abstract

The present invention relates to the field of intelligent production technology, and more particularly to a kind of intelligent production supporting process of Cordyceps militaris, obtaining the first entity image sequence of the culture medium and the first content curve of multiple index components in the culture medium during the process of culturing Cordyceps militaris spawn to the end of the aerial mycelium stage in the culture medium; Based on the first entity image sequence and multiple first content curves, the first environmental condition required for the primordium induction stage and the first index component increase and decrease of the culture medium are determined; By the first environmental condition and the first index component increase and decrease, the environment and culture medium in the primordium induction stage are controlled until the cultivation of the primordium induction stage is completed, and the fruiting body primordium is obtained; After the fruiting body primordium is cultivated to meet the harvesting standard, the Cordyceps militaris fruiting body meeting the harvesting standard is harvested. The present invention can improve production efficiency and ensure the quality of the final product.

Description

Technical Field

[0001] The present application relates to the field of intelligent production technology, and in particular to a supporting process for the intelligent production of Cordyceps militaris. Background Art

[0002] Cordyceps militaris is a kind of insect-borne fungus, also known as northern cordyceps, northern cordyceps or pupa grass. Cordyceps militaris has high medicinal value. Modern research shows that Cordyceps militaris contains cordycepin, cordyceps polysaccharides and other biologically active substances, which play an important role in anti-inflammatory, anti-cancer, antibacterial, and immune enhancement. It is similar to Cordyceps sinensis and is one of the ideal substitutes for Cordyceps sinensis.

[0003] In the traditional production process of Cordyceps militaris, it generally includes the stages of inoculation, spawning, bud induction, culture and harvesting. The inoculation stage is mainly to place the Cordyceps militaris strains in the culture medium; the spawning stage is mainly to make the Cordyceps militaris strains grow in the culture medium to produce hyphae and aerial hyphae in the culture medium; bud induction is mainly to induce the formation of fruiting body primordia; culture is to make the fruiting body primordia grow to meet the harvesting standards; harvesting is to harvest the Cordyceps militaris fruiting bodies that meet the harvesting standards. Among them, the process of inducing fruiting body primordia mainly uses lightless culture and light culture to induce the production of fruiting body primordia. The quality of the fruiting body primordia determines the quality of subsequent culture. However, in the existing production process, the production of induced fruiting body primordia often relies on artificial experience to control the environment, prepare the culture medium and monitor the growth. This is not only inefficient, but also difficult to ensure the stability and consistency of the quality of the fruiting body primordia.

[0004] The problems and defects of the existing technology are as follows: the existing process of inducing the production of fruiting body primordia in Cordyceps militaris relies on artificial experience for environmental control, culture medium preparation and growth monitoring, which is inefficient and difficult to ensure the stability and consistency of product quality, and it is impossible to efficiently produce Cordyceps militaris with a high degree of standardization. Summary of the invention

[0005] To achieve the above objectives, this application provides the following technical solutions:

[0006] According to the first aspect of the present invention, the present invention seeks protection for a Cordyceps militaris intelligent production supporting process, comprising:

[0007] Acquire a first entity image sequence of the culture medium and a first content curve of a plurality of indicator components in the culture medium during the process of culturing Cordyceps militaris strains in the culture medium until the aerial mycelium stage ends, wherein the first entity image sequence includes a first entity image distributed in a first time dimension, each indicator component corresponds to a first content curve, and the first content curve includes a first content value of the corresponding indicator component distributed in a second time dimension, and the indicator component is a main component of a component required when the Cordyceps militaris strains are cultivated to meet a harvesting standard;

[0008] Determining the first environmental conditions required for the primordium induction stage and the increase or decrease of the first indicator component of the culture medium based on the first entity image sequence and the plurality of the first content curves;

[0009] The environment and the culture medium in the primordium induction stage are controlled by the first environmental condition and the increase or decrease of the first indicator component until the culture in the primordium induction stage is completed to obtain the fruiting body primordium;

[0010] After the fruiting body primordium is cultured until it meets the harvesting standard, the Cordyceps militaris fruiting bodies that meet the harvesting standard are harvested.

[0011] Optionally, determining the first environmental condition required for the primordium induction stage and the increase or decrease of the first indicator component of the culture medium based on the first entity image sequence and the plurality of the first content curves includes:

[0012] Performing image segmentation processing on the first entity image sequence to obtain a target image sequence of the culture medium, wherein the target image sequence includes target images distributed in the first time dimension;

[0013] Performing morphological processing on the target image sequence to obtain a morphological feature sequence of the culture medium, wherein the morphological feature sequence includes the target images distributed in the first time dimension;

[0014] performing normalization processing on the plurality of the first content curves respectively to obtain a plurality of normalized content curves after normalization processing, wherein the normalized content curves include normalized content values ​​distributed in the second time dimension;

[0015] fusing the plurality of normalized content curves with the morphological feature sequence respectively to obtain a plurality of fused curves, each of the fused curves corresponding to one normalized content curve, the fused curves including fused values ​​distributed in a third time dimension;

[0016] Performing time series feature extraction processing on the plurality of fused curves respectively to obtain a plurality of time series features, wherein each fused curve corresponds to a time series feature;

[0017] Performing environmental condition prediction processing and index component prediction processing on the plurality of time series features respectively to obtain environmental condition prediction results and index component prediction results;

[0018] The first environmental condition required for the primordium induction stage is determined based on the environmental condition prediction result, and the increase or decrease amount of the first indicator component of the culture medium is determined based on the indicator component prediction result.

[0019] Optionally, performing image segmentation processing on the first entity image sequence to obtain a target image sequence of the culture medium further comprises:

[0020] Performing image segmentation on each first entity image in the first entity image sequence to obtain an image segmentation result of each first entity image;

[0021] Classify the image segmentation results of each of the first entity images to determine a plurality of target images of the culture medium, where each culture medium corresponds to one target image;

[0022] Arranging the target images of the culture medium according to the real space position of the culture medium to obtain a plurality of target image arrays of the culture medium, each of the target image arrays corresponding to one of the first entity images;

[0023] All the target image arrays are sorted according to the first time dimension to obtain a target image sequence of the culture medium.

[0024] Optionally, performing morphological processing on the target image sequence to obtain a morphological feature sequence of the culture medium includes:

[0025] For each target image array in the target image sequence, a local morphological feature extraction process is performed on each target image in the target image array to obtain a morphological feature array corresponding to the target image array, wherein each array unit in the morphological feature array corresponds to a local morphological feature value;

[0026] Performing global morphological feature extraction processing based on the morphological feature array to obtain global morphological feature values ​​of the target image array, each of the target image arrays corresponds to one global morphological feature value;

[0027] All the global morphological feature values ​​are sorted according to the first time dimension to obtain a morphological feature sequence of the culture medium.

[0028] Optionally, for each target image array in the target image sequence, performing local morphological feature extraction processing on each target image in the target image array to obtain a morphological feature array corresponding to the target image array includes:

[0029] For each target image array in the target image sequence, extract the corresponding target image in each array unit;

[0030] Determining a total area of ​​the target image and an edge feathering area of ​​the target image;

[0031] Calculating a ratio between the edge feathering area and the total area, and determining the ratio as a morphological feature value of the array unit;

[0032] The morphological feature values ​​of all the array elements in the target image array are calculated to obtain a morphological feature array corresponding to the target image array.

[0033] Optionally, the step of respectively normalizing the plurality of first content curves to obtain a plurality of normalized content curves after normalization includes:

[0034] For each of the first content curves, determining the maximum content value, the minimum content value and the average content value of the corresponding indicator component in each of the first content curves, and determining the maximum slope value and the minimum slope value of each of the first content curves;

[0035] Normalizing each of the first content curves based on the maximum content value, the minimum content value, the average content value, the maximum slope value, and the minimum slope value of each of the first content curves;

[0036] The normalization process is as follows: ;

[0037] in, Indicates i The first content curve j The content normalized value of the first content value, Indicates i the maximum content value of the first content curve, Indicates i the minimum content value of the first content curve, Indicates i The first content curve j The first content value, Indicates i The maximum slope value of the first content curve, Indicates i The minimum slope value of the first content curve, Indicates i The first content curve j The slope at the first content value, Through the i The first content curve j Taking the derivative at the first content value, we get: Indicates i The average content value in the first content curve.

[0038] Optionally, the step of fusing the plurality of normalized content curves with the morphological feature sequence to obtain a plurality of fused curves comprises:

[0039] Determining the starting time point of the plurality of normalized content curves as the alignment starting time point and determining the ending time point of the plurality of normalized content curves as the alignment ending time point;

[0040] Fitting the morphological feature sequence within the time interval between the alignment start time point and the alignment end time point to obtain a morphological feature curve;

[0041] The morphological characteristic curve is fitted with each of the normalized content curves to obtain multiple fusion curves.

[0042] Optionally, the primordium induction stage includes a lightless culture stage and a light culture stage, the lightless culture stage is in front, and the light culture stage is in the back, and the environmental condition prediction processing and the indicator component prediction processing are performed on the multiple time series features respectively, and the environmental condition prediction results and the indicator component prediction results are obtained, including:

[0043] Performing environmental condition prediction processing based on the time series characteristics to obtain a first environmental condition prediction result corresponding to the light-free culture stage and a second environmental condition prediction result corresponding to the light culture stage, wherein the first environmental condition prediction result does not include light prediction, and the second environmental condition prediction result includes light prediction;

[0044] Performing indicator component prediction processing based on the time series characteristics to obtain a first indicator component prediction result for the light-free cultivation stage and a second indicator component prediction result for the light-irradiated cultivation stage;

[0045] The step of determining the first environmental condition required for the primordium induction stage based on the environmental condition prediction result, and determining the increase or decrease amount of the first indicator component of the culture medium based on the indicator component prediction result comprises:

[0046] The first environmental condition prediction result and the second environmental condition prediction result are spliced ​​in time to obtain the first environmental condition required for the primordium induction stage;

[0047] The prediction result of the first indicator component and the prediction result of the second indicator component are spliced ​​in time to obtain the increase or decrease of the first indicator component of the culture medium.

[0048] Optionally, controlling the environment and the culture medium in the primordium induction stage by the first environmental condition and the increase or decrease of the first indicator component includes:

[0049] In the lightless culture stage, the environment is controlled according to the first environmental condition prediction result, and the content of the indicator component of the culture medium is controlled according to the first indicator component prediction result;

[0050] During a preset time period starting from the illumination culture stage, the environment is controlled according to the second environmental condition prediction result, and a second entity image sequence of the culture medium and a second content curve of a plurality of indicator components in the culture medium are obtained during the preset time period;

[0051] Based on the second entity image sequence and the plurality of the second content curves, determining the second environmental conditions required for the remaining stages of the illumination culture stage and the increase or decrease of the second indicator component of the culture medium;

[0052] The environment and the culture medium of the remaining stage in the light culture stage are controlled by the second environmental condition and the increase or decrease of the second indicator component.

[0053] The present invention relates to the field of intelligent production technology, and in particular to a supporting process for intelligent production of Cordyceps militaris. In the process of culturing Cordyceps militaris strains in a culture medium to an aerial mycelium stage, a first entity image sequence of a culture medium and a first content curve of a plurality of index components in the culture medium are obtained, wherein the first entity image sequence includes a first entity image distributed in a first time dimension, each index component corresponds to a first content curve, the first content curve includes a first content value of a corresponding index component distributed in a second time dimension, and the index component is a main component of a component required when the Cordyceps militaris strains are cultivated to meet a harvesting standard; based on the first entity image sequence and the plurality of first content curves, a first environmental condition required in a primordium induction stage and an increase or decrease in a first index component of the culture medium are determined; the environment and the culture medium in the primordium induction stage are controlled by the first environmental condition and the increase or decrease in the first index component until the cultivation in the primordium induction stage is completed to obtain a fruiting body primordium; after the fruiting body primordium is cultivated to meet the harvesting standard, the Cordyceps militaris fruiting body meeting the harvesting standard is harvested. The present invention adopts a highly automated production process, and adjusts the culture conditions in the primordium induction stage by monitoring the image data in the previous stage and the content data of the index components in the culture medium, and accurately controls the key components in the culture medium, so as to produce fruiting body primordia of quality standards without the need for human intervention, thereby improving production efficiency and ensuring the quality of the final product. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flowchart of a Cordyceps militaris intelligent production supporting process claimed in an embodiment of the present invention;

[0055] Figure 2A second flow chart of a supporting process for intelligent production of Cordyceps militaris claimed in an embodiment of the present invention;

[0056] Figure 3 A third flow chart of a supporting process for intelligent production of Cordyceps militaris claimed in an embodiment of the present invention;

[0057] Figure 4 A fourth flow chart of a supporting process for intelligent production of Cordyceps militaris claimed in an embodiment of the present invention;

[0058] Figure 5 A fifth flow chart of a supporting process for intelligent production of Cordyceps militaris claimed in an embodiment of the present invention;

[0059] Figure 6 A sixth flow chart of a Cordyceps militaris intelligent production supporting process claimed in an embodiment of the present invention;

[0060] Figure 7 The seventh flow chart of a Cordyceps militaris intelligent production supporting process claimed in the embodiment of the present invention;

[0061] Figure 8 The eighth flow chart of a supporting process for intelligent production of Cordyceps militaris claimed in an embodiment of the present invention;

[0062] Fig. 9 This is the ninth flow chart of a supporting process for the intelligent production of Cordyceps militaris as claimed in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments collected by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0064] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" are any variations thereof, and are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0065] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0066] According to one aspect of the present disclosure, a process for intelligent production of Cordyceps militaris is provided. Figure 1 According to some embodiments of the present disclosure, a supporting process for intelligent production of Cordyceps militaris includes:

[0067] Step S110: obtaining a first entity image sequence of the culture medium and a first content curve of a plurality of indicator components in the culture medium during the process of culturing the Cordyceps militaris strains in the culture medium until the aerial mycelium stage ends, wherein the first entity image sequence includes a first entity image distributed in a first time dimension, each indicator component corresponds to a first content curve, and the first content curve includes a first content value of the corresponding indicator component distributed in a second time dimension, and the indicator component is a main component of a component required when the Cordyceps militaris strains are cultivated to meet the harvesting standard;

[0068] Step S120: determining the first environmental conditions required for the primordium induction stage and the increase or decrease of the first indicator component of the culture medium based on the first entity image sequence and the plurality of first content curves;

[0069] Step S130: controlling the environment and culture medium in the primordium induction stage by means of the first environmental condition and the increase or decrease of the first indicator component until the culture in the primordium induction stage is completed to obtain the fruiting body primordium;

[0070] Step S140: After the fruiting body primordium is cultivated to meet the harvesting standard, the Cordyceps militaris fruiting bodies meeting the harvesting standard are harvested.

[0071] The method comprises obtaining a first entity image sequence of a culture medium and first content curves of a plurality of index components in the culture medium during the process of culturing Cordyceps militaris strains to the aerial mycelium stage in the culture medium, wherein the first entity image sequence includes a first entity image distributed in a first time dimension, each index component corresponds to a first content curve, the first content curve includes a first content value of a corresponding index component distributed in a second time dimension, and the index component is a main component of a component required when the Cordyceps militaris strains are cultivated to meet the harvesting standard; based on the first entity image sequence and the plurality of first content curves, determining the first environmental condition required for the primordium induction stage and the increase or decrease of the first index component of the culture medium; controlling the environment and the culture medium in the primordium induction stage by the first environmental condition and the increase or decrease of the first index component until the cultivation in the primordium induction stage is completed to obtain the primordium of the fruiting body; after the primordium of the fruiting body is cultivated to meet the harvesting standard, harvesting the Cordyceps militaris fruiting body that meets the harvesting standard. A highly automated production process is adopted. By monitoring the image data in the previous stage and the content data of the indicator components in the culture medium, the culture conditions in the primordium induction stage are adjusted, and the key components in the culture medium are precisely controlled. This can produce fruiting body primordia of quality standards without the need for human intervention, improve production efficiency, and ensure the quality of the final product.

[0072] In some embodiments, in step S110, the Cordyceps militaris strain is cultured in a culture medium, that is, cultured from the inoculation stage to the end of the germination stage. The culture from the inoculation stage to the end of the germination stage can be performed manually or by automated equipment. The end of the germination stage can be the growth of hyphae and aerial hyphae in the culture medium. During the two steps from the inoculation stage to the end of the germination stage, the culture machine is photographed by a camera at regular intervals to obtain a first entity image sequence A. N ={a 1 ,a 2 ,…a N-1 , a N}, N is the number of first entity images in the first entity image sequence.

[0073] At the same time, in the two processes from the inoculation stage to the end of the spawning stage, the components in each culture medium are measured by the component content determination method and recorded as discrete content values. The discrete content values ​​are fitted to obtain the content curve F(r), where r is the content value and different components correspond to different content curves. By performing principal component analysis on the components required to cultivate the Cordyceps militaris to meet the harvesting standards, the components with the greatest impact are analyzed as the principal components, and these principal components are determined as indicator components, and the content curve corresponding to the indicator components is determined as the first content curve R ( ), Indicates i The first content curve j The first content value. There can be one or more indicator components. The embodiment of the present invention is described by taking multiple as examples.

[0074] The first time dimension and the second time dimension can be understood as time dimensions with different time granularities. For the first content curve R( ), since the curve is obtained by fitting, the second time dimension can be continuous, and the time granularity can be understood as infinitely close to 0. The time granularity of the first time dimension is the acquisition frequency of the first entity image.

[0075] The first entity image may include a plurality of culture media, each of which corresponds to a first content curve R( ). The first entity image can be obtained by taking a picture of one side of the culture medium placement cabinet by a camera, thereby obtaining a first entity image including multiple culture media. The culture media in the first entity image are referred to as entities.

[0076] In some embodiments, in step S120, the first entity image sequence A N ={a 1 ,a 2 ,…a N-1 , a N}, the growth of the Cordyceps militaris from the inoculation stage to the end of the germination stage can be extracted, and the change of the content value of the index component of the culture medium can be extracted through the first content curve. N ={a 1 ,a 2 ,…a N-1 , a N} and a plurality of first content curves R( ) can be used to predict the environmental conditions and the increase or decrease of each indicator component required for the subsequent primordium induction stage, and then the culture of the primordium induction stage can be precisely controlled.

[0077] In some embodiments, in step S120, the first environmental condition may include temperature, humidity, light, time and other conditions, and the increase or decrease of the indicator component may be an increase in the content of the indicator component or a decrease in the content of the indicator component.

[0078] In some embodiments, in step S130, for the primordium induction stage, the environment is controlled by the first environmental condition, and the content of the index component of the culture medium is controlled by the increase or decrease of the first index component. According to the changes in the previous two stages, the primordium induction stage can be controlled more specifically to produce fruiting body primordia with higher stability and consistency. The fruiting body primordium is a rice-shaped specialized aerial mycelium, and an image recognition algorithm can be used to identify whether the rice-shaped specialized aerial mycelium appears. If it appears, it can be determined that the primordium induction stage is over. If it does not appear, the culture of the primordium induction stage can continue.

[0079] In some embodiments, in step S140, after the primordium induction stage is completed, the fruiting body primordium is obtained, and the cultivation stage is entered. In the cultivation stage, the fruiting body primordium is cultivated to make the fruiting body mature, and the harvesting standard is that the size of the fruiting body of Cordyceps militaris basically does not grow any more. After the fruiting body matures, it can be determined that the cultivation stage is over and the harvesting stage is entered.

[0080] It should be noted that the cultivation stage can adopt the existing cultivation method, or the third entity image sequence of the culture medium and the third content curves of multiple index components in the culture medium can be obtained during the process from the inoculation stage to the end of the primordium induction stage. Based on the third entity image sequence and the multiple third content curves, the third environmental conditions required for the cultivation stage and the increase or decrease of the third index component of the culture medium are determined, and the cultivation stage is refined and intelligently controlled, so that more standardized mature fruiting bodies can be produced in the cultivation stage, further improving the production efficiency and quality of Cordyceps militaris.

[0081] In some embodiments, in step S120, based on the first entity image sequence and the plurality of first content curves, determining the first environmental conditions required for the primordium induction stage and the increase or decrease of the first indicator component of the culture medium includes:

[0082] like Figure 2 As shown, step S120 specifically includes:

[0083] Step S210: performing image segmentation processing on the first entity image sequence to obtain a target image sequence of the culture medium, wherein the target image sequence includes target images distributed in a first time dimension;

[0084] Step S220: performing morphological processing on the target image sequence to obtain a morphological feature sequence of the culture medium, wherein the morphological feature sequence includes morphological feature values ​​distributed in the first time dimension;

[0085] Step S230: performing normalization processing on the plurality of first content curves respectively to obtain a plurality of normalized content curves after normalization processing, wherein the normalized content curves include normalized content values ​​distributed in the second time dimension;

[0086] Step S240: fusing the multiple normalized content curves with the morphological feature sequence respectively to obtain multiple fused curves, each fused curve corresponding to a normalized content curve, and the fused curve includes fused values ​​distributed in the third time dimension;

[0087] Step S250: performing time series feature extraction processing on the multiple fusion curves respectively to obtain multiple time series features, where each fusion curve corresponds to one time series feature;

[0088] Step S260: performing environmental condition prediction processing and index component prediction processing on multiple time series features respectively to obtain environmental condition prediction results and index component prediction results;

[0089] Step S270: Determine the first environmental condition required for the primordium induction stage based on the environmental condition prediction result, and determine the increase or decrease amount of the first indicator component of the culture medium based on the indicator component prediction result.

[0090] By forming a first entity image sequence A comprising a plurality of culture media N ={a 1 ,a 2 ,…a N-1 , a N Each first entity image is segmented to obtain a target image in each first entity image, and then a target image sequence B containing only the target culture medium is obtained. N ={b 1 ,b 2 ,…b N-1 ,b N The first entity image sequence A can be processed by image processing techniques (such as edge detection, threshold segmentation, region growing, etc.) N ={a 1 ,a 2 ,…a N-1 , a N}, remove the background, noise and other non-target areas, and only retain the image part of the target culture medium, thereby obtaining the target image sequence B N ={b 1 ,b 2 ,…b N-1 ,bN}, target image sequence B N ={b 1 ,b 2 ,…b N-1 ,b N Each image in} contains the target culture medium and is continuously distributed in the first time dimension. Extract the morphological features in the target image for subsequent analysis. Apply morphological operations (such as corrosion, expansion, opening operation, closing operation, etc.) to the target image sequence B N ={b 1 ,b 2 ,…b N-1 ,b N} to highlight or eliminate specific structures in the image, thereby extracting the morphological features of the culture medium. Generate morphological feature sequence C N ={c 1 ,c 2 ,…c N-1 ,c N}, morphological feature sequence C N ={c 1 ,c 2 ,…c N-1 ,c N} contains the morphological characteristic values ​​of the target culture medium and is continuously distributed in the first time dimension.

[0091] The first content curve R( ) is converted into a unified, dimensionless numerical form for easy comparison and analysis. ) Apply the normalization formula to convert the first content curve R( ) is mapped to a uniform interval (such as [0,1]), and multiple normalized content curves S( ). The first content curve R( ) All were converted into content normalized values , content normalized value Continuously distributed in the second time dimension.

[0092] The normalized content information is combined with the morphological features to generate the fusion curve M( ) to facilitate the subsequent time series feature extraction. Each normalized content curve S( ) and the corresponding morphological feature sequence C N ={c 1 ,c 2 ,…c N-1 ,c N} can be fused by weighting, superposition or other fusion algorithms to generate multiple fusion curves M( ). Each fusion curve M( ) contain normalized content information and morphological feature information, and are continuously distributed in the third time dimension.

[0093] From the fusion curve M( ) to extract time series features that are useful for prediction. Apply time series feature extraction algorithms (such as Fourier transform, wavelet transform, sliding window statistics, etc.) to each fusion curve to extract representative time series features. Multiple time series features are obtained, and each time series feature corresponds to a fusion curve M( ), and can be used for subsequent predictive analysis.

[0094] The extracted time series features are used to predict environmental conditions and indicator components. The time series features are input into the prediction model (such as machine learning model, deep learning model, etc.) to perform environmental condition prediction processing and indicator component prediction processing respectively. The environmental condition prediction results and indicator component prediction results are obtained, and the environmental condition prediction results and indicator component prediction results will be used to guide the subsequent primordium induction stage.

[0095] Determine the specific control parameters of the primordium induction stage based on the prediction results. Determine the first environmental conditions (such as temperature, humidity, light, etc.) required for the primordium induction stage based on the environmental condition prediction results; at the same time, determine the increase or decrease of the first indicator component of the culture medium based on the indicator component prediction results. Provide precise control guidance for the subsequent primordium induction stage.

[0096] In some embodiments, Figure 3 As shown, step S210 specifically includes:

[0097] Step S310: performing image segmentation on each first entity image in the first entity image sequence to obtain an image segmentation result of each first entity image;

[0098] Step S320: performing classification processing on the image segmentation results of each first entity image to determine a plurality of target images of culture media, where each culture media corresponds to one target image;

[0099] Step S330: Arranging the target images of the culture medium according to the real space position of the culture medium to obtain a plurality of target image arrays of the culture medium, each target image array corresponding to a first entity image;

[0100] Step S340: Sort all target image arrays according to the first time dimension to obtain a target image sequence of the culture medium.

[0101] For the first entity image sequence AN ={a 1 ,a 2 ,…a N-1 , a N}, perform image segmentation on each first entity image in the image segmentation image to obtain the image segmentation result of each first entity image. Perform image segmentation on each first entity image input. Divide the first entity image into multiple regions to obtain different parts of the first entity image. Classify the image segmentation result of each first entity image to determine multiple target images of culture media, and each culture media corresponds to a target image. By using image recognition or machine learning algorithms, it can be identified which segmented regions are culture media. For each first entity image, several target images of culture media can be determined. Each identified culture media corresponds to a target image. Arrange the target images of the culture media according to the real space position of the culture media to obtain a target image array P×Q of multiple culture media, where P represents the number of rows and Q represents the number of columns, and each target image array P×Q corresponds to a first entity image. Arrange the identified target images of the culture media according to the real space position of the culture media in the real space. In this way, the spatial relationship of objects in the image can be maintained, and each arranged target image array P×Q corresponds to a first entity image, reflecting the spatial distribution of the culture media in the first entity image. All target image arrays are sorted according to the first time dimension to obtain a target image sequence N×P×Q of the culture medium. All the arranged target image arrays are sorted according to the time order of the first entity image to which they belong in the first entity image sequence, thereby obtaining a target image sequence N×P×Q of multiple culture media arranged in time order. The target image sequence N×P×Q not only retains the spatial distribution information of the culture medium in each first entity image, but also reflects the temporal changes in the spatial distribution information of the culture medium.

[0102] In some embodiments, Figure 4 As shown, step S220 specifically includes:

[0103] Step S410: for each target image array in the target image sequence, perform local morphological feature extraction processing on each target image in the target image array to obtain a morphological feature array corresponding to the target image array, wherein each array unit in the morphological feature array corresponds to a local morphological feature value;

[0104] Step S420: performing global morphological feature extraction processing based on the morphological feature array to obtain global morphological feature values ​​of the target image array, where each target image array corresponds to a global morphological feature value;

[0105] Step S430: sorting all global morphological feature values ​​according to the first time dimension to obtain a morphological feature sequence of the culture medium.

[0106] For each target image array P×Q in the target image sequence N×P×Q, local morphological feature extraction is performed on each target image in the target image array P×Q to obtain a morphological feature array P`×Q` corresponding to the target image array P×Q, where P` represents the number of rows, Q` represents the number of columns, and each array unit in the morphological feature array P`×Q` corresponds to a local morphological feature value. Each target image array P×Q in the target image sequence can be traversed. For each target image in the target image array (i.e., each target image of the culture medium), local morphological features are extracted. Local morphological features may include shape, edge, texture, etc., and can describe the local visual characteristics of the culture medium. After the extraction is completed, each target image will correspond to a local morphological feature value, and the array composed of local morphological feature values ​​is the morphological feature array.

[0107] Based on the morphological feature array P`×Q`, global morphological feature extraction processing is performed to obtain the global morphological feature value of the target image array, and each target image array corresponds to a global morphological feature value. The global morphological feature can be extracted based on the obtained morphological feature array P`×Q`. The global morphological feature is an overall description of the entire target image array (i.e., the target image of all culture media in a first entity image), and the global morphological feature can include comprehensive or statistical information of all local features. Each target image array will correspond to a global morphological feature value, and this global morphological feature value comprehensively reflects the visual characteristics of all culture media images in a first entity image. Specifically, the global morphological feature value can be the average value of the local morphological feature values.

[0108] All global morphological feature values ​​are sorted according to the first time dimension to obtain the morphological feature sequence C of the culture medium. N ={c 1 ,c 2 ,…c N-1 ,c N}. Sort all extracted global morphological feature values ​​according to the time order of the first entity image to which they belong in the original sequence. Retain the temporal change information of the morphological features, thereby forming a morphological feature sequence C arranged in time. N ={c 1 ,c 2 ,…c N-1 ,c N}. Morphological feature sequence C N ={c 1 ,c 2 ,…cN-1 ,c N}For analyzing morphological changes in culture over time.

[0109] In some embodiments, Figure 5 As shown, step S410 specifically includes:

[0110] Step S510: for each target image array in the target image sequence, extract the target image corresponding to each array unit;

[0111] Step S520: determining the total area of ​​the target image and the edge feathering area of ​​the target image;

[0112] Step S530: Calculate the ratio between the edge feathering area and the total area, and determine the ratio as the local morphological feature value of the array unit;

[0113] Step S540: Calculate the local morphological feature values ​​of all array elements in the target image array to obtain a morphological feature array corresponding to the target image array.

[0114] For each target image array P×Q in the target image sequence N×P×Q, the corresponding target image in each array unit is extracted. Each target image array in the target image sequence can be traversed, and each unit in the target image array P×Q, that is, the target image of each culture medium can be extracted one by one, so that the target image of each culture medium can be processed separately.

[0115] Determine the total area of ​​the target image and the edge feathering area of ​​the target image. The total area of ​​each target image (target image of the culture medium) is measured, which can be measured by pixel count or image area. At the same time, the feathering area of ​​the image edge can be detected and calculated. The above edge feathering represents the outline area of ​​the aerial hyphae on the culture medium.

[0116] The ratio between the edge feathering area and the total area is calculated, and the ratio is determined as the local morphological feature value of the array unit. The ratio between the contour area of ​​the hyphae (feathering area) and the total area of ​​the corresponding culture medium is used to represent the ratio of the contour area of ​​the hyphae to the area of ​​the entire hyphae image, thereby providing a quantitative indicator to evaluate the extension or blurring of the hyphae edge relative to the size of the entire culture medium. The ratio can be determined as the local morphological feature value of the array unit (i.e., the target image of the culture medium), which can reflect the morphological characteristics of the hyphae and the culture medium, especially the morphology of its edge.

[0117] Calculate the local morphological eigenvalues ​​of all array elements in the target image array P×Q to obtain the morphological feature array P`×Q` corresponding to the target image array P×Q. Perform the above steps for each array element in the target image array P×Q to obtain the local morphological eigenvalues ​​corresponding to each element. The local morphological eigenvalues ​​are combined into a new array, namely the morphological feature array P`×Q`. The morphological feature array P`×Q` corresponds to the original target image array P×Q, but contains morphological feature information instead of target image data.

[0118] In some embodiments, Figure 6 As shown, step S230 specifically includes:

[0119] Step S610: for each first content curve, determining the maximum content value, the minimum content value and the average content value of the corresponding indicator component in each first content curve, and determining the maximum slope value and the minimum slope value of each first content curve;

[0120] Step S620: normalizing each first content curve based on the maximum content value, the minimum content value, the average content value, the maximum slope value, and the minimum slope value of each first content curve;

[0121] The normalization process is shown in the following formula: ;

[0122] in, Indicates i The first content curve j The content normalized value of the first content value, Indicates i the maximum content value of the first content curve, Indicates i the minimum content value of the first content curve, Indicates i The first content curve j The first content value, Indicates i The maximum slope value of the first content curve, Indicates i The minimum slope value of the first content curve, Indicates i The first content curve j The slope at the first content value, Through the i The first content curve j Taking the derivative at the first content value, we get: Indicates iThe average content value in the first content curve.

[0123] For each first content curve R( ) to conduct detailed data analysis. Traverse each first content curve R( ), find each first content curve R( ) corresponds to the maximum content value of the indicator component, which represents the highest point of the indicator component during the measurement process. Determine each first content curve R( ), the minimum content value of the indicator component represents the lowest level of the indicator component during the measurement process. At the same time, each first content curve R( ) represents the average level of the index component from the inoculation stage to the end of the germination stage. In addition, each first content curve R( ) to determine its maximum and minimum slope values. The slope reflects the speed of content change and is of great significance for understanding the dynamic changes of component content.

[0124] Each first content curve R( ) is normalized. Normalization is a data preprocessing technique that aims to adjust the data to a uniform scale for fairer comparison and analysis. Data between different curves can be directly compared because they are all converted to the same numerical range. Normalization helps to eliminate the impact of dimension and order of magnitude on data analysis results, making subsequent data fusion more accurate.

[0125] In some embodiments, Figure 7 As shown, step S240 specifically includes:

[0126] Step S710: determining the start time point of the plurality of normalized content curves as the alignment start time point and determining the end time point of the plurality of normalized content curves as the alignment end time point;

[0127] Step S720: fitting the morphological feature sequence within the time interval between the alignment start time point and the alignment end time point to obtain a morphological feature curve;

[0128] Step S730: fitting the morphological characteristic curve with each of the multiple normalized content curves to obtain multiple fusion curves.

[0129] For multiple normalized content curves S( ) to align the time points. Normalized content curve S( ) may have different time spans or start / end time points, so it is necessary to select a common time start and end point for subsequent comparison and analysis. ) have the same time span or start / end time points, you can choose a common time start and end point. This ensures that all data are compared within the same time frame, which helps to fit the morphological feature sequence more accurately.

[0130] After determining the alignment start time point and the alignment end time point, a continuous morphological feature curve C is fitted by interpolation or regression based on the existing morphological feature sequence (discrete global morphological feature value) within the time interval between the alignment start time point and the alignment end time point. ). Morphological characteristic curve C ( ) can more intuitively reflect the continuous change trend of morphological characteristics over time and provide a basis for the subsequent fusion with the normalized content curve.

[0131] The morphological characteristic curve C( ) and each normalized content curve S( ) is further fitted or fused to obtain the fusion curve M( ). Combining two different types of related data (i.e., global morphological feature values ​​and content data) together enables a more comprehensive analysis and understanding of the relationship between global morphological feature values ​​and content data. Fusion curve M( ) not only reflects the change of content over time, but also incorporates the change characteristics of morphology, making the implicit feature information richer. Each fusion curve corresponds to a normalized content curve.

[0132] In some embodiments, Figure 8 As shown, the primordium induction stage includes a lightless culture stage and a light culture stage, wherein the lightless culture stage is in the front and the light culture stage is in the back. Step S260 specifically includes:

[0133] Step S810: performing environmental condition prediction processing based on the time series characteristics to obtain a first environmental condition prediction result corresponding to the lightless cultivation stage and a second environmental condition prediction result corresponding to the light cultivation stage, wherein the first environmental condition prediction result does not include the light prediction result, and the second environmental condition prediction result includes the light prediction result;

[0134] Step S820: performing indicator component prediction processing based on the time series characteristics to obtain a first indicator component prediction result in the dark culture stage and a second indicator component prediction result in the light culture stage;

[0135] Step S270 specifically includes:

[0136] Step S830: splicing the first environmental condition prediction result and the second environmental condition prediction result in time to obtain the first environmental condition required for the primordium induction stage;

[0137] Step S840: splicing the prediction results of the first indicator component and the second indicator component in time to obtain the increase or decrease of the first indicator component of the culture medium.

[0138] The time series features are used to predict the environmental conditions required for different stages (lightless cultivation stage and light cultivation stage). For the lightless cultivation stage, the predicted environmental conditions do not include light, because this stage is carried out in the absence of light. For the light cultivation stage, the predicted environmental conditions include light, because light is required in this stage to promote the growth and development of Cordyceps militaris.

[0139] Time series features are used to predict the increase or decrease in the content of indicator components of the culture medium in different culture stages (lightless culture stage and light culture stage). The indicator components can be nutrients, metabolites, or other compounds that have an impact on the growth and development of Cordyceps militaris. By predicting the increase or decrease in the content of the indicator components, the state of the culture process can be better controlled and adjusted accordingly to optimize the culture conditions.

[0140] Combining the predicted results of the environmental conditions in the lightless culture stage and the light culture stage to form a continuous environmental condition setting plan can ensure that the environmental conditions can transition smoothly throughout the primordium induction stage, thereby avoiding adverse effects of sudden environmental changes on biological samples. The above-mentioned splicing process helps to achieve the consistency and consistency of culture conditions.

[0141] Combine the prediction results of the indicator components in the dark culture stage and the light culture stage to predict the overall change trend of the indicator components in the entire primordium induction stage. Through splicing, a more comprehensive perspective can be obtained to look at the increase and decrease prediction of the indicator components in the culture medium, and then adjust the culture medium formula in the dark culture stage and the light culture stage, so as to ensure that the indicator components in the culture medium maintain an appropriate level throughout the culture cycle.

[0142] In some embodiments, Fig. 9 As shown, step S130 specifically includes:

[0143] Step S910: in the lightless culture stage, the environment is controlled according to the first environmental condition prediction result, and the content of the indicator component of the culture medium is controlled according to the first indicator component prediction result;

[0144] Step S920: performing environmental control according to the second environmental condition prediction result within a preset time period starting from the illumination culture stage, and obtaining a second entity image sequence of the culture medium and a second content curve of a plurality of indicator components in the culture medium within the preset time period;

[0145] Step S930: determining the second environmental conditions required for the remaining stages of the illumination culture stage and the increase or decrease of the second indicator component of the culture medium based on the second entity image sequence and the plurality of second content curves;

[0146] Step S940: Control the environment and culture medium of the remaining stages in the light culture stage through the second environmental condition and the increase or decrease of the second indicator component.

[0147] In the lightless culture stage, according to the prediction results of the first environmental condition, the environment is adjusted and controlled accordingly, such as temperature, humidity, gas composition, etc. At the same time, according to the prediction results of the first indicator component, the content of key indicator components in the culture medium is monitored and adjusted to ensure that the composition of the culture medium is suitable for the growth and development needs of Cordyceps militaris at this stage.

[0148] After entering the illumination culture stage, a preset time period begins, and the preset time period can be understood as an adaptation period or transition period), and the environment is controlled using the second environmental condition prediction result, which may include light intensity, temperature, humidity, etc. At the same time, the entity image sequence of the culture medium within the preset time period can be collected, i.e., the second entity image sequence, which can reflect the state change of the culture medium and the fruiting body primordium of Cordyceps militaris during the illumination culture stage. In addition, the content changes of multiple key indicator components in the culture medium are monitored and recorded to form a second content curve of multiple indicator components.

[0149] By analyzing the second entity image sequence and multiple second content curves collected in the preset time period by the method and steps in the above embodiment, the initial response of the fruiting body primordium of Cordyceps militaris to the light culture stage can be evaluated. It can be calculated how to adjust the environmental conditions (such as light intensity, temperature, humidity, etc.) to achieve the best culture effect during the remaining time of the light culture stage. At the same time, it is also determined which indicator components in the culture medium need to be increased or decreased during the remaining time of the light culture stage to maintain the most suitable growth environment.

[0150] After determining the second environmental condition required for the illumination culture stage in the remaining time and the increase or decrease of the second index component of the culture medium, the corresponding control strategy can be implemented. The control strategy may include adjusting the environmental control equipment according to the second environmental condition to maintain suitable conditions such as illumination, temperature, humidity, and adjusting the composition of the culture medium according to the increase or decrease of the second index component. Ensure that the biological sample can continue its growth and development process in the best environment and improve the quality of the fruiting body primordium of Cordyceps militaris.

[0151] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0152] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A Cordyceps militaris intelligent production supporting process, characterized in that: include: Acquire a first entity image sequence of the culture medium and a first content curve of a plurality of indicator components in the culture medium during the process of culturing Cordyceps militaris strains in the culture medium until the aerial mycelium stage ends, wherein the first entity image sequence includes a first entity image distributed in a first time dimension, each indicator component corresponds to a first content curve, the first content curve includes a first content value of a corresponding indicator component distributed in a second time dimension, and the indicator component is a main component of a component required when the Cordyceps militaris strains are cultivated to meet a harvesting standard; Performing image segmentation processing on the first entity image sequence to obtain a target image sequence of the culture medium, wherein the target image sequence includes target images distributed in the first time dimension; Performing morphological processing on the target image sequence to obtain a morphological feature sequence of the culture medium, wherein the morphological feature sequence includes morphological feature values ​​distributed in the first time dimension; Normalizing the plurality of first content curves respectively to obtain a plurality of normalized content curves after normalization, wherein the normalized content curves include normalized content values ​​distributed in the second time dimension; fusing the plurality of normalized content curves with the morphological feature sequence respectively to obtain a plurality of fused curves, each of the fused curves corresponding to one normalized content curve, the fused curves including fused values ​​distributed in a third time dimension; Performing time series feature extraction processing on the plurality of fused curves respectively to obtain a plurality of time series features, wherein each fused curve corresponds to a time series feature; Performing environmental condition prediction processing and index component prediction processing on the plurality of time series features respectively to obtain environmental condition prediction results and index component prediction results; Determining the first environmental condition required for the primordium induction stage based on the environmental condition prediction result, and determining the increase or decrease amount of the first indicator component of the culture medium based on the indicator component prediction result; The environment and the culture medium in the primordium induction stage are controlled by the first environmental condition and the increase or decrease of the first indicator component until the culture in the primordium induction stage is completed to obtain the fruiting body primordium; After the fruiting body primordium is cultured until it meets the harvesting standard, the Cordyceps militaris fruiting bodies that meet the harvesting standard are harvested.

2. The intelligent production supporting process of Cordyceps militaris according to claim 1, characterized in that: The performing image segmentation processing on the first entity image sequence to obtain the target image sequence of the culture medium comprises: Performing image segmentation on each first entity image in the first entity image sequence to obtain an image segmentation result of each first entity image; Performing classification processing on the image segmentation results of each of the first entity images to determine a plurality of target images of the culture medium, wherein each culture medium corresponds to one target image; Arranging the target images of the culture medium according to the real space position of the culture medium to obtain a plurality of target image arrays of the culture medium, each of the target image arrays corresponding to one of the first entity images; All the target image arrays are sorted according to the first time dimension to obtain a target image sequence of the culture medium.

3. The intelligent production supporting process of Cordyceps militaris according to claim 2, characterized in that: The step of performing morphological processing on the target image sequence to obtain the morphological feature sequence of the culture medium comprises: For each of the target image arrays in the target image sequence, a local morphological feature extraction process is performed on each of the target images in the target image array to obtain a morphological feature array corresponding to the target image array, wherein each array unit in the morphological feature array corresponds to a local morphological feature value; Performing global morphological feature extraction processing based on the morphological feature array to obtain global morphological feature values ​​of the target image array, each of the target image arrays corresponds to one global morphological feature value; All the global morphological feature values ​​are sorted according to the first time dimension to obtain a morphological feature sequence of the culture medium.

4. The intelligent production supporting process of Cordyceps militaris according to claim 3, characterized in that: The step of performing local morphological feature extraction processing on each target image in the target image array in the target image sequence to obtain a morphological feature array corresponding to the target image array includes: For each of the target image arrays in the target image sequence, extract the target image corresponding to each array unit; Determining a total area of ​​the target image and an edge feathering area of ​​the target image; Calculating a ratio between the edge feathering area and the total area, and determining the ratio as a local morphological feature value of the array unit; The morphological feature values ​​of all the array elements in the target image array are calculated to obtain a morphological feature array corresponding to the target image array.

5. The intelligent production supporting process of Cordyceps militaris according to claim 1, characterized in that: The step of respectively normalizing the plurality of first content curves to obtain a plurality of normalized content curves after normalization includes: For each of the first content curves, determining the maximum content value, the minimum content value and the average content value of the corresponding indicator component in each of the first content curves, and determining the maximum slope value and the minimum slope value of each of the first content curves; Normalizing each of the first content curves based on the maximum content value, the minimum content value, the average content value, the maximum slope value, and the minimum slope value of each of the first content curves; The normalization process is shown in the following formula: ； in, Indicates i The first content curve j The content normalized value of the first content value, Indicates i the maximum content value of the first content curve, Indicates i the minimum content value of the first content curve, Indicates i The first content curve j The first content value, Indicates i The maximum slope value of the first content curve, Indicates i The minimum slope value of the first content curve, Indicates i The first content curve j The slope at the first content value, Through the i The first content curve j Taking the derivative at the first content value, we get: Indicates i The average content value in the first content curve.

6. The intelligent production supporting process of Cordyceps militaris according to claim 1, characterized in that: The step of fusing the plurality of normalized content curves with the morphological feature sequence to obtain a plurality of fused curves comprises: Determining the starting time point of the plurality of normalized content curves as the alignment starting time point and determining the ending time point of the plurality of normalized content curves as the alignment ending time point; Fitting the morphological feature sequence within the time interval between the alignment start time point and the alignment end time point to obtain a morphological feature curve; The morphological characteristic curve is fitted with each of the normalized content curves to obtain multiple fusion curves.

7. The intelligent production supporting process of Cordyceps militaris according to claim 1, characterized in that: The primordium induction stage includes a lightless culture stage and an illumination culture stage, wherein the lightless culture stage is in front and the illumination culture stage is in the back, and the environmental condition prediction processing and the index component prediction processing are respectively performed on the multiple time series features to obtain the environmental condition prediction results and the index component prediction results, including: Performing environmental condition prediction processing based on the time series characteristics to obtain a first environmental condition prediction result corresponding to the light-free culture stage and a second environmental condition prediction result corresponding to the light culture stage, wherein the first environmental condition prediction result does not include a light prediction result, and the second environmental condition prediction result includes a light prediction result; Performing indicator component prediction processing based on the time series characteristics to obtain a first indicator component prediction result for the light-free cultivation stage and a second indicator component prediction result for the light-irradiated cultivation stage; The step of determining the first environmental condition required for the primordium induction stage based on the environmental condition prediction result, and determining the increase or decrease amount of the first indicator component of the culture medium based on the indicator component prediction result comprises: The first environmental condition prediction result and the second environmental condition prediction result are spliced ​​in time to obtain the first environmental condition required for the primordium induction stage; The prediction result of the first indicator component and the prediction result of the second indicator component are spliced ​​in time to obtain the increase or decrease of the first indicator component of the culture medium.

8. The intelligent production supporting process of Cordyceps militaris according to claim 7, characterized in that: The controlling of the environment and the culture medium in the primordium induction stage by the first environmental condition and the increase or decrease of the first indicator component comprises: In the lightless culture stage, the environment is controlled according to the first environmental condition prediction result, and the content of the indicator component of the culture medium is controlled according to the first indicator component prediction result; During a preset time period starting from the illumination culture stage, the environment is controlled according to the second environmental condition prediction result, and a second entity image sequence of the culture medium and a second content curve of a plurality of indicator components in the culture medium are obtained during the preset time period; Based on the second entity image sequence and the plurality of the second content curves, determining the second environmental conditions required for the remaining stages of the illumination culture stage and the increase or decrease of the second indicator component of the culture medium; The environment and the culture medium of the remaining stage in the light culture stage are controlled by the second environmental condition and the increase or decrease of the second indicator component.

Images