Extraction method of earthworm polypeptide

Abstract

The invention relates to the technical field of intelligent control, and particularly discloses an earthworm polypeptide extraction method, which is characterized in that a soaking state monitoring video of an earthworm soaked in a citric acid solution is analyzed based on a deep learning artificial intelligence technology, and surface state characteristics and surface state time sequence change characteristics of the earthworm are excavated from the soaking state monitoring video; therefore, whether the earthworms are fully soaked or not is judged. In this way, the key time point when soaking is finished can be intelligently judged, so that the extraction process of earthworm polypeptide is optimized.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and more specifically, to a method for extracting earthworm polypeptides. Background Art

[0002] Earthworm peptides have a variety of biological activities, including antibacterial, antioxidant, anti-inflammatory, immunomodulatory, etc. They are produced by the decomposition of proteins in earthworms and play an important physiological function in the digestive tract of earthworms.

[0003] Earthworm peptides are believed to have a wide range of application potentials, and can be used in the fields of medicine, cosmetics, food, etc. However, there are some problems with the current extraction schemes of earthworm peptides. For example, they are low in efficiency, complex in operation, high in cost, and easy to cause peptide degradation. Therefore, an optimized extraction method of earthworm peptides is expected. Summary of the invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method for extracting earthworm polypeptides, which uses deep learning artificial intelligence technology to analyze the monitoring video of the earthworms soaking in a citric acid solution, and mines the surface state characteristics of the earthworms and the time-series change characteristics of the surface state, so as to judge whether the earthworms are fully soaked. In this way, the key time point of the end of soaking can be intelligently judged to optimize the extraction process of earthworm polypeptides.

[0005] Accordingly, according to one aspect of the present application, a method for extracting earthworm polypeptides is provided, comprising: cleaning the earthworms, cutting the earthworms into small pieces, and soaking them in a solution containing 0.5% sodium chloride to obtain earthworms after impurities are removed; filtering and rinsing the earthworms after impurities are removed with clean water, and then soaking them in a solution containing 0.1% citric acid to obtain earthworms after alkaline neutralization; filtering and rinsing the earthworms after alkaline neutralization with clean water, and then soaking them in a solution containing 0.5% acetic acid to obtain earthworms after tissue destruction; The earthworms after tissue destruction are filtered and rinsed with clean water, and then immersed in a solution containing 0.1% pancreatic enzyme to obtain hydrolyzed earthworms; the hydrolyzed earthworms are filtered, the supernatant is collected, and concentrated and separated with an ultrafiltration membrane to obtain a filtrate containing polypeptides; and the filtrate containing polypeptides is purified by ethanol precipitation to obtain earthworm polypeptides, wherein the earthworms after impurity removal are filtered and rinsed with clean water, and then immersed in a solution containing 0.1% citric acid to obtain alkaline neutralized earthworms, including:

[0006] Acquire a monitoring video of the soaking state of the cleaned earthworms soaked in the citric acid solution for a predetermined period of time, which is captured by a camera;

[0007] Extracting the surface state features of the earthworms in the immersion state monitoring video to obtain a sequence of earthworm surface state time series feature graphs;

[0008] Constructing the time series variation characteristics of the sequence of the earthworm surface state time series characteristic graphs to obtain a time series representation feature vector of the earthworm surface state variation;

[0009] Based on the characteristic vector representing the time series of the earthworm surface state change, it is determined whether the soaking is sufficient.

[0010] In the above-mentioned earthworm polypeptide extraction method, the earthworm surface state characteristics of the immersion state monitoring video are extracted to obtain a sequence of earthworm surface state time-series feature graphs, including: performing data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips; using a deep learning network model to perform feature extraction on the sequence of immersion state monitoring video clips to obtain a sequence of earthworm surface state time-series feature graphs.

[0011] In the above-mentioned earthworm polypeptide extraction method, data preprocessing is performed on the immersion state monitoring video to obtain a sequence of immersion state monitoring video segments, including: video segmentation is performed on the immersion state monitoring video to obtain a sequence of immersion state monitoring video segments.

[0012] In the above-mentioned earthworm polypeptide extraction method, the deep learning network model is an immersion state temporal association feature extractor based on a three-dimensional convolutional neural network model; wherein, the immersion state temporal association feature extractor based on a three-dimensional convolutional neural network model includes an input layer, a three-dimensional convolutional layer, an activation function layer, a three-dimensional pooling layer and an output layer.

[0013] In the above-mentioned earthworm polypeptide extraction method, a deep learning network model is used to perform feature extraction on the sequence of the immersion state monitoring video clips to obtain a sequence of the earthworm surface state time series feature graphs, including: passing the sequence of the immersion state monitoring video clips through the immersion state time series association feature extractor based on the three-dimensional convolutional neural network model to obtain a sequence of the earthworm surface state time series feature graphs.

[0014] In the above-mentioned method for extracting earthworm polypeptides, the temporal change characteristics of the sequence of earthworm surface state temporal feature graphs are constructed to obtain a temporal characterization feature vector of earthworm surface state changes, including: performing feature distribution optimization on the sequence of earthworm surface state temporal feature graphs to obtain a sequence of optimized earthworm surface state temporal feature graphs; calculating the state change semantic measurement coefficient between each adjacent two optimized earthworm surface state temporal feature graphs in the sequence of optimized earthworm surface state temporal feature graphs to obtain the temporal characterization feature vector of earthworm surface state changes composed of multiple state change semantic measurement coefficients.

[0015] In the above-mentioned earthworm polypeptide extraction method, the state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs is calculated to obtain the earthworm surface state change time series characterization feature vector composed of multiple state change semantic measurement coefficients, including: calculating the state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs by the following state change semantic measurement formula;

[0016] The state change semantic measurement formula is:

[0017]

[0018] Among them, p (x,y,z) is the characteristic value of the position (x, y, z) in the earthworm surface state time series characteristic diagram after the previous optimization, q (x,y,z) is the characteristic value of the position (x, y, z) in the latter optimized earthworm surface state time series characteristic graph, H is the height of each optimized earthworm surface state time series characteristic graph, W is the width of each optimized earthworm surface state time series characteristic graph, C is the number of channels of each optimized earthworm surface state time series characteristic graph, Hp i is the semantic measurement coefficient of the i-th state change, and log represents the logarithmic function operation with base 2.

[0019] In the above-mentioned earthworm polypeptide extraction method, whether the soaking is sufficient is determined based on the characteristic vector representing the time series of changes in the earthworm surface state, including: passing the characteristic vector representing the time series of changes in the earthworm surface state through a classifier to obtain a classification result, and the classification result is used to indicate whether the soaking is sufficient.

[0020] In the above-mentioned method for extracting earthworm polypeptides, the time series characterization feature vector of the earthworm surface state change is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the earthworm is sufficiently soaked, including: using the fully connected layer of the classifier to fully connect the feature vector of the earthworm surface state change to obtain a fully connected encoded feature vector; inputting the fully connected encoded feature vector into the Softmax classification function of the classifier to obtain the probability value of the earthworm surface state change time series characterization feature vector belonging to each classification label, and the classification labels include sufficiently soaked and not sufficiently soaked; and determining the classification label corresponding to the largest probability value as the classification result.

[0021] Compared with the prior art, the earthworm polypeptide extraction method provided in the present application uses deep learning artificial intelligence technology to analyze the monitoring video of the earthworms soaking in a citric acid solution, thereby mining the surface state characteristics of the earthworms and the time-series change characteristics of the surface state, so as to judge whether the earthworms are fully soaked. In this way, the key time point of the end of soaking can be intelligently judged to optimize the extraction process of earthworm polypeptides. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0023] Figure 1 The present invention is a flowchart of a method for extracting earthworm polypeptides according to an embodiment of the present application.

[0024] Figure 2 Schematic diagram of the structure of the method for extracting earthworm polypeptides according to an embodiment of the present application.

[0025] Figure 3 The present invention is a flowchart of extracting the earthworm surface state characteristics of the immersion state monitoring video in the earthworm polypeptide extraction method according to an embodiment of the present application to obtain a sequence of earthworm surface state time series characteristic diagrams.

[0026] Figure 4 The present invention is a flowchart for constructing the time series change characteristics of the sequence of the earthworm surface state time series characteristic diagram in the earthworm polypeptide extraction method according to the embodiment of the present application to obtain the earthworm surface state change time series characterization feature vector.

[0027] Figure 5 This is a flow chart of obtaining a classification result by passing a feature vector representing the time series of changes in the earthworm surface state through a classifier in a method for extracting earthworm polypeptides according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0029] The present application provides a method for extracting earthworm polypeptides, and the specific steps thereof include: S1, washing the earthworms, cutting the earthworms into small segments, and soaking the segments in a solution containing 0.5% sodium chloride to obtain decontaminated earthworms; S2, filtering and rinsing the decontaminated earthworms with clean water, and then soaking the segments in a solution containing 0.1% citric acid to obtain alkaline neutralized earthworms; S3, filtering and rinsing the alkaline neutralized earthworms with clean water, and then soaking the segments in a solution containing 0.5% acetic acid to obtain tissue-destroyed earthworms; S4, filtering and rinsing the tissue-destroyed earthworms with clean water, and then soaking the segments in a solution containing 0.1% pancreatic enzyme to obtain hydrolyzed earthworms; S5, filtering the hydrolyzed earthworms, collecting the supernatant, and concentrating and separating the segments using an ultrafiltration membrane to obtain a filtrate containing polypeptides; and, S6, purifying the filtrate containing polypeptides using an ethanol precipitation method to obtain earthworm polypeptides.

[0030] In particular, in step S2, the role of the citric acid solution is to neutralize the alkaline substances in the earthworm body. If the soaking is not sufficient, alkaline substances will remain in the earthworm body, affecting the subsequent tissue destruction and hydrolysis process. Specifically, the residual alkaline substances will reduce the acidity of the acetic acid solution, resulting in incomplete tissue destruction, resulting in a lower extraction rate of earthworm polypeptides; the residual alkaline substances will also affect the activity of pancreatic enzymes, resulting in a lower hydrolysis efficiency, resulting in an uneven molecular weight distribution of earthworm polypeptides, etc. At the same time, if the earthworms are soaked in the citric acid solution for too long, the acidic environment may cause irreversible degradation of the polypeptides in the earthworm body. This will cause structural destruction and loss of activity of the polypeptides, reducing the content and quality of the extracted polypeptides. In actual application scenarios, the soaking time is usually specified based on experience, but this method is very dependent on the feelings and judgments of individuals or teams. This subjectivity may lead to differences in the judgment of soaking time by different people or different teams, lacking objectivity and consistency. In addition, experience may be affected by factors such as individual differences, experimental conditions, and sample characteristics, making the soaking time specified by experience have a certain degree of uncertainty. Therefore, an optimized solution is expected.

[0031] Considering that when the alkaline substances in the earthworms are neutralized, the surface state of the earthworms will change. In the technical concept of this application, it is expected to utilize this characteristic and combine it with an intelligent algorithm to analyze the monitoring video of the earthworms soaking in citric acid solution, from which the surface state characteristics of the earthworms and the time-series change characteristics of the surface state are mined, so as to judge whether the soaking is sufficient. In this way, the key time point of the end of soaking can be intelligently judged to optimize the extraction process of earthworm polypeptides.

[0032] Figure 1 The present invention is a flowchart of a method for extracting earthworm polypeptides according to an embodiment of the present application. Figure 2FIG. 1 is a schematic diagram of the structure of the method for extracting earthworm polypeptides according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the method for extracting earthworm polypeptides according to the embodiment of the present application includes the steps of: S110, obtaining a monitoring video of the soaking state of the earthworms after impurity removal soaked in the citric acid solution during a predetermined time period, which is collected by a camera; S120, extracting the surface state characteristics of the earthworms in the soaking state monitoring video to obtain a sequence of earthworm surface state time series feature diagrams; S130, constructing the time series change characteristics of the sequence of earthworm surface state time series feature diagrams to obtain a time series characterization feature vector of earthworm surface state changes; S140, determining whether the soaking is sufficient based on the time series characterization feature vector of earthworm surface state changes.

[0033] In the above-mentioned method for extracting earthworm polypeptides, the step S110 is to obtain a monitoring video of the soaking state of the earthworms after impurity removal soaked in the citric acid solution during a predetermined time period, which is captured by a camera. It should be understood that the camera can capture the soaking state of the earthworms in the citric acid solution in real time, so as to timely observe and record the changes in the surface state of the earthworms, and perform subsequent analysis and judgment based on real-time monitoring. The soaking state monitoring video captured by the camera contains information such as the color and morphological changes of the earthworms. By analyzing and processing the soaking state monitoring video using an intelligent algorithm, the characteristics of the surface state of the earthworms, such as color, texture, shape and other characteristics, can be extracted. Based on these characteristic information, it can help to judge whether the citric acid solution has fully neutralized the alkaline substances in the earthworms, thereby determining the key time point for the end of soaking.

[0034] In the above-mentioned method for extracting earthworm polypeptides, the step S120 extracts the surface state characteristics of the earthworms in the immersion state monitoring video to obtain a sequence of earthworm surface state time series characteristic diagrams. It should be understood that by extracting the surface state characteristics of the earthworms in the immersion state monitoring video, the surface state changes of the earthworms during the immersion process can be understood, such as surface wetness, color changes, surface activity, etc. These characteristic information reflects the physiological changes and reactions that may occur in the earthworms during the immersion process. Based on these characteristic information, it is possible to determine whether the alkaline substances in the earthworms are completely neutralized. That is, by quantitatively analyzing the surface state characteristics of the earthworms, the end time of the earthworm immersion can be determined.

[0035] Figure 3 The present invention is a flowchart of extracting the earthworm surface state characteristics of the immersion state monitoring video in the earthworm polypeptide extraction method according to the embodiment of the present application to obtain a sequence of earthworm surface state time series characteristic diagrams. Figure 3As shown, the step S120 includes: S121, performing data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips; S122, using a deep learning network model to perform feature extraction on the sequence of immersion state monitoring video clips to obtain a sequence of earthworm surface state time series feature graphs.

[0036] Specifically, the step S121 performs data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips. In a specific example of the present application, the processing method for performing data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips is to perform video segmentation on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips. It should be understood that by decomposing the entire immersion process into video clips of multiple time periods, the changes in the immersion state of earthworms at different stages can be observed more finely, which is conducive to the time series analysis of subtle changes in the immersion process. In other words, by cutting the video into clips, it is possible to focus on the immersion state of earthworms at specific stages in a targeted manner, which helps to more accurately capture the detailed features of the immersion state of earthworms, thereby more accurately judging the adequacy of the immersion. At the same time, by decomposing a large video into multiple small clips through the video segmentation operation, each video clip can be analyzed and understood separately, thereby reducing the complexity of data processing and making data processing more efficient.

[0037] Specifically, in step S122, the sequence of the soaking state monitoring video clips is feature extracted using a deep learning network model to obtain a sequence of the earthworm surface state time series feature graphs. In a specific example of the present application, the deep learning network model is a soaking state time series correlation feature extractor based on a three-dimensional convolutional neural network model, that is, a soaking state time series correlation feature extractor is constructed using a three-dimensional convolutional neural network model to capture the time series feature distribution of the earthworm soaking state contained in each soaking state monitoring video clip, so as to obtain a sequence of the earthworm surface state time series feature graphs. A person of ordinary skill in the art should know that a three-dimensional convolutional neural network (3D CNN) is a deep learning model suitable for processing video or other three-dimensional data, which can effectively capture the spatial correlation and temporal correlation in three-dimensional data, thereby learning the time series features of the earthworm soaking state, such as color, shape, texture and other information. The working principle of a three-dimensional convolutional neural network is similar to that of a traditional two-dimensional convolutional neural network, both of which use convolution operations to capture feature information in input data, but the three-dimensional convolutional neural network takes the time dimension into account when processing video data. In a three-dimensional convolutional neural network, the convolution kernel moves not only in the width and height of the image, but also in the depth (time). The convolution kernel captures the temporal and spatial features implicit in the input data by performing convolution operations in a sliding window manner in three dimensions, so that the network can understand the movement and changes in the video data, thereby better capturing the surface state characteristics of the earthworms during the immersion process. At the same time, the three-dimensional convolutional neural network also uses pooling operations to reduce the size of the feature map, retain the key features therein and filter out the interference of redundant information, further reduce the number of parameters of the model, improve computational efficiency and more accurately describe the surface state of the earthworms. More specifically, in an embodiment of the present application, the immersion state temporal correlation feature extractor based on the three-dimensional convolutional neural network model includes an input layer, a three-dimensional convolution layer, an activation function layer, a three-dimensional pooling layer and an output layer.

[0038] In the above-mentioned method for extracting earthworm polypeptides, the step S130 constructs the time-series change characteristics of the sequence of the earthworm surface state time-series characteristic diagram to obtain the time-series characterization feature vector of the earthworm surface state change. It should be understood that the earthworm contains alkaline substances, and the citric acid solution is an acidic solution. During the process of earthworms soaking in the solution, the time-series change characteristics of their surface states can reflect the process of the alkaline substances in the earthworm body being neutralized by the acidic solution. By analyzing the time-series change of the surface state characteristics, the degree of neutralization of the alkaline substances in the earthworm body can be indirectly inferred. For example, when the alkaline substances are completely neutralized, the surface state of the earthworm may show obvious changes, such as changes in color, texture or shape. That is to say, by constructing the time-series change characteristics of the sequence of the earthworm surface state time-series characteristic diagram, it can be inferred whether the alkaline substances have been completely neutralized. In addition, by analyzing the time-series change of the surface state of the earthworm during the soaking process in the solution, the neutralization process of the alkaline substances in the earthworm body can be monitored in real time. This provides an important basis for timely adjusting the soaking time, thereby ensuring that the alkaline substances in the earthworms are fully neutralized and the earthworm polypeptides are extracted to the maximum extent.

[0039] Figure 4 A flowchart of constructing the temporal change characteristics of the sequence of the temporal characteristic diagram of the earthworm surface state in the method for extracting earthworm polypeptides according to the embodiment of the present application to obtain the temporal characteristic vector of the earthworm surface state change. Figure 4 As shown, the step S130 includes: S131, performing feature distribution optimization on the sequence of earthworm surface state time series feature graphs to obtain a sequence of optimized earthworm surface state time series feature graphs; S132, calculating the state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of optimized earthworm surface state time series feature graphs to obtain a feature vector representing the earthworm surface state change time series composed of multiple state change semantic measurement coefficients.

[0040] Specifically, the step S131 performs feature distribution optimization on the sequence of earthworm surface state temporal feature graphs to obtain a sequence of optimized earthworm surface state temporal feature graphs. In the technical solution of the present application, when the sequence of immersion state monitoring video clips is passed through an immersion state temporal association feature extractor based on a three-dimensional convolutional neural network model to obtain a sequence of earthworm surface state temporal feature graphs, each eigenvalue of each earthworm surface state temporal feature graph expresses the temporal association image semantic feature of the image frame of the corresponding immersion state monitoring video clip, that is, for the earthworm surface state temporal feature graph, it has a specific spatial meaning extracted based on the temporal association image semantic feature of the three-dimensional convolutional neural network model in both the spatial distribution dimension within the feature matrix and the channel distribution dimension between the feature matrices.

[0041] Therefore, if the spatial information expression effect of the earthworm surface state time series feature map as a high-dimensional feature can be improved, the overall expression effect of the earthworm surface state time series feature map can be improved, thereby improving the expression effect of the earthworm surface state change time series characterization feature vector composed of multiple state change semantic measurement coefficients obtained by calculating the state change semantic measurement coefficient between each adjacent two earthworm surface state time series feature maps in the sequence of earthworm surface state time series feature maps, so as to improve the accuracy of the classification result obtained by the earthworm surface state change time series characterization feature vector through the classifier.

[0042] Based on this, in the technical solution of the present application, each earthworm surface state time series characteristic graph F in the sequence of earthworm surface state time series characteristic graphs is optimized by the following optimization formula;

[0043] Wherein, the optimization formula is:

[0044]

[0045]

[0046] Among them, f i,j,k ∈F and f i',j',k' ∈F, that is, f i,j,k and f i',j',k' are all characteristic values ​​of the earthworm surface state time series characteristic graph F, and α is the local space segmentation coefficient, N is the scale of the local neighborhood of the earthworm surface state time series characteristic graph F, and f' i,j,k It is the characteristic value of the (i, j, k)th position of the earthworm surface state time series characteristic diagram after optimization.

[0047] Specifically, taking the local segmentation space in the Hilbert space after the earthworm surface state time series feature graph F is expanded as a reference, the characteristic manifold of the earthworm surface state time series feature graph F in the high-dimensional feature space is locally integrated, so as to correct the phase change discontinuity of the characteristic manifold expressed by the non-stationary data sequence after the local space expansion of the earthworm surface state time series feature graph F based on the local integration processing of the integral function, so as to obtain a finer structure and geometric features of the characteristic manifold, improve the spatial information expression effect of the earthworm surface state time series feature graph F in the high-dimensional feature space, thereby improving the expression effect of the earthworm surface state time series feature graph F, and finally improve the accuracy of the classification result obtained by the classifier of the feature vector representing the earthworm surface state change time series.

[0048] Specifically, the step S132 is to calculate the state change semantic metric coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs to obtain the earthworm surface state change time series characterization feature vector composed of multiple state change semantic metric coefficients. In the above scheme, the soaking state time series association feature extractor based on the three-dimensional convolutional neural network model is used to extract the surface state time series features of the earthworms in each time period, but these feature information is not enough to judge whether the soaking is sufficient. Because each earthworm surface state time series feature graph only reflects the state information in an independent time period, and lacks the overall state semantic connection, it is impossible to distinguish the state of different stages in the soaking process. Therefore, in the technical solution of the present application, the state change semantic metric coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs is calculated to obtain the earthworm surface state change time series characterization feature vector composed of multiple state change semantic metric coefficients. That is, by calculating the state change semantic metric coefficient, the degree of change and change trend of the earthworm soaking state between two adjacent time periods are quantified, so as to judge whether the soaking is sufficient. Specifically, as earthworms soak in citric acid solution, their surface state will change to a certain extent, such as color. These changes reflect the degree of neutralization of alkaline substances in the earthworms and the degree of degradation of polypeptides. If these changes can be accurately detected and quantified, it can be judged whether the soaking has achieved the best effect based on the size and speed of the changes.

[0049] In a specific example of the present application, step S132 includes: calculating the state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs using the following state change semantic measurement formula to obtain the earthworm surface state change time series representation feature vector composed of multiple state change semantic measurement coefficients; wherein the state change semantic measurement formula is:

[0050]

[0051] Among them, p (x,y,z) is the characteristic value of the position (x, y, z) in the earthworm surface state time series characteristic diagram after the previous optimization, q (x,y,z) is the characteristic value of the position (x, y, z) in the latter optimized earthworm surface state time series characteristic graph, H is the height of each optimized earthworm surface state time series characteristic graph, W is the width of each optimized earthworm surface state time series characteristic graph, C is the number of channels of each optimized earthworm surface state time series characteristic graph, Hp i is the semantic measurement coefficient of the i-th state change, and log represents the logarithmic function operation with base 2.

[0052] In particular, by calculating the state change semantic measurement coefficient between each two adjacent earthworm surface state time-series feature maps, the cross-difference between the previous earthworm surface state time-series feature map and the next earthworm surface state time-series feature map can be reflected, that is, the difference between the next earthworm surface state time-series feature map relative to the previous earthworm surface state time-series feature map, and the difference between the previous earthworm surface state time-series feature map relative to the next earthworm surface state time-series feature map. This cross-difference can more accurately characterize and describe the difference in earthworm surface state between each two adjacent earthworm surface state time-series feature maps.

[0053] In the above-mentioned method for extracting earthworm polypeptides, the step S140 determines whether the soaking is sufficient based on the characteristic vector representing the time series of the earthworm surface state change. It should be understood that the characteristic vector representing the time series of the earthworm surface state change contains multi-dimensional information of the earthworm surface state change, such as color, texture, shape and other characteristics, which can reflect the overall changes of the earthworm during the soaking process and provide a comprehensive information basis. The surface state of the earthworm and the degree of neutralization of the alkaline substances in the earthworm body have a certain synergistic correlation. In other words, the characteristic information in the characteristic vector representing the time series of the earthworm surface state change has a comprehensive characterization capability for the degree of neutralization of the earthworm characteristic alkaline substances. Therefore, further using a machine learning algorithm to analyze and process the characteristic vector representing the time series of the earthworm surface state change can effectively determine whether the soaking of the earthworm has reached the required degree of sufficiency.

[0054] In a specific example of the present application, the coding process for determining whether the soaking is sufficient based on the time series characterization feature vector of the earthworm surface state change is to pass the time series characterization feature vector of the earthworm surface state change through a classifier to obtain a classification result, and the classification result is used to indicate whether the soaking is sufficient. A person of ordinary skill in the art should know that a classifier is a machine learning model, and its function is to classify input data into different categories or labels. First, the classifier is trained using labeled training data. The training data includes input features and their corresponding category labels. The classifier establishes classification rules by learning the patterns and features in the training data, finds a decision boundary, and separates data of different categories to complete the classification task. In the technical solution of the present application, the time series characterization feature vector of the earthworm surface state change is input into the classifier for automatic classification processing, and the classifier can convert the complex features in the time series characterization feature vector of the earthworm surface state change into an objective classification result, thereby realizing a standardized judgment on whether the soaking is sufficient or not. In this way, no human intervention is required, the influence of subjective factors is reduced, and the judgment is more objective and accurate. At the same time, the judgment based on the classifier can realize real-time monitoring of the soaking status of earthworms, and then adjust the soaking time in time to ensure that the earthworms are fully soaked while avoiding the adverse effects caused by excessive soaking.

[0055] Figure 5 This is a flow chart of obtaining a classification result by passing the time series characteristic vector of the earthworm surface state change through a classifier in the earthworm polypeptide extraction method according to an embodiment of the present application. Figure 5 As shown, the step S140 includes: S141, using the fully connected layer of the classifier to fully connect the feature vector representing the time series change of the earthworm surface state to obtain a fully connected encoded feature vector; S142, inputting the fully connected encoded feature vector into the Softmax classification function of the classifier to obtain the probability value of the feature vector representing the time series change of the earthworm surface state belonging to each classification label, and the classification labels include fully soaked and not fully soaked; S143, determining the classification label corresponding to the largest one among the probability values ​​as the classification result.

[0056] In summary, the extraction method of earthworm polypeptides according to the embodiment of the present application is explained, which uses deep learning-based artificial intelligence technology to analyze the monitoring video of the earthworms soaking in a citric acid solution, and mines the surface state characteristics of the earthworms and the time-series change characteristics of the surface state, so as to judge whether the earthworms are fully soaked. In this way, the key time point of the end of soaking can be intelligently judged to optimize the extraction process of earthworm polypeptides.

[0057] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0058] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0059] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0060] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for extracting earthworm polypeptides, comprising: After cleaning the earthworms, the earthworms are cut into small pieces, and the pieces are soaked in a 0.5% sodium chloride solution to obtain impurity-free earthworms; After filtering and washing the removed earthworms with clean water, the earthworms are immersed in a 0.1% citric acid solution to obtain alkaline neutralized earthworms; After filtering and washing the alkaline neutralized earthworms with water, the earthworms are immersed in a solution containing 0.5% acetic acid to obtain earthworms with damaged tissues; after filtering and washing the earthworms with water, the earthworms with damaged tissues are immersed in a solution containing 0.1% pancreatic enzyme to obtain earthworms with hydrolysis; The hydrolyzed earthworms are filtered, the supernatant is collected, and the supernatant is concentrated and separated by an ultrafiltration membrane to obtain a filtrate containing polypeptides; and the filtrate containing polypeptides is purified by an ethanol precipitation method to obtain earthworm polypeptides, characterized in that the earthworms after impurity removal are filtered and rinsed with clean water, and then immersed in a 0.1% citric acid solution to obtain alkaline neutralized earthworms, comprising: Acquire a monitoring video of the soaking state of the cleaned earthworms soaked in the citric acid solution for a predetermined period of time, which is captured by a camera; Extracting the surface state features of the earthworms in the immersion state monitoring video to obtain a sequence of earthworm surface state time series feature graphs; Constructing the time series variation characteristics of the sequence of the earthworm surface state time series characteristic graphs to obtain a time series representation feature vector of the earthworm surface state variation; Based on the characteristic vector representing the time series of the earthworm surface state change, it is determined whether the soaking is sufficient.

2. The method for extracting earthworm polypeptide according to claim 1, characterized in that: Extracting the earthworm surface state features of the immersion state monitoring video to obtain a sequence of earthworm surface state time series feature graphs, including: Performing data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips; A deep learning network model is used to extract features from the sequence of immersion state monitoring video clips to obtain a sequence of time-series feature graphs of the earthworm surface state.

3. The method for extracting earthworm polypeptide according to claim 2, characterized in that: Performing data preprocessing on the immersion state monitoring video to obtain a sequence of immersion state monitoring video clips includes: The immersion state monitoring video is segmented to obtain a sequence of the immersion state monitoring video segments.

4. The method for extracting earthworm polypeptide according to claim 3, characterized in that: The deep learning network model is an immersion state temporal association feature extractor based on a three-dimensional convolutional neural network model; wherein, the immersion state temporal association feature extractor based on a three-dimensional convolutional neural network model includes an input layer, a three-dimensional convolution layer, an activation function layer, a three-dimensional pooling layer and an output layer.

5. The method for extracting earthworm polypeptide according to claim 4, characterized in that: The deep learning network model is used to extract features from the sequence of the immersion state monitoring video clips to obtain a sequence of time-series feature graphs of the earthworm surface state, including: The sequence of the immersion state monitoring video clips is passed through the immersion state temporal association feature extractor based on the three-dimensional convolutional neural network model to obtain a sequence of the earthworm surface state temporal feature graphs.

6. The method for extracting earthworm polypeptide according to claim 5, characterized in that: Constructing the time series change characteristics of the sequence of the earthworm surface state time series characteristic graphs to obtain the earthworm surface state change time series characterization feature vector, including: Performing feature distribution optimization on the sequence of earthworm surface state time series feature graphs to obtain an optimized sequence of earthworm surface state time series feature graphs; The state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs is calculated to obtain the earthworm surface state change time series characterization feature vector composed of multiple state change semantic measurement coefficients.

7. The method for extracting earthworm polypeptide according to claim 6, characterized in that: Calculating the state change semantic metric coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs to obtain the earthworm surface state change time series representation feature vector composed of a plurality of state change semantic metric coefficients, including: The state change semantic measurement coefficient between each two adjacent optimized earthworm surface state time series feature graphs in the sequence of the optimized earthworm surface state time series feature graphs is calculated using the following state change semantic measurement formula; The state change semantic measurement formula is: Among them, p (x,y,z) is the characteristic value of the position (x, y, z) in the earthworm surface state time series characteristic diagram after the previous optimization, q (x,y,z) is the characteristic value of the position (x, y, z) in the latter optimized earthworm surface state time series characteristic graph, H is the height of each optimized earthworm surface state time series characteristic graph, W is the width of each optimized earthworm surface state time series characteristic graph, C is the number of channels of each optimized earthworm surface state time series characteristic graph, Hp i is the semantic measurement coefficient of the i-th state change, and log represents the logarithmic function operation with base 2.

8. The method for extracting earthworm polypeptide according to claim 7, characterized in that: Determining whether the soaking is sufficient based on the time series characterization feature vector of the earthworm surface state change includes: The time series characteristic feature vector of the earthworm surface state change is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the soaking is sufficient.

9. The method for extracting earthworm polypeptide according to claim 8, characterized in that: The time series characteristic feature vector of the earthworm surface state change is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the soaking is sufficient, including: Using the fully connected layer of the classifier to perform fully connected encoding on the time series characterization feature vector of the earthworm surface state change to obtain a fully connected encoded feature vector; Inputting the fully connected encoded feature vector into the Softmax classification function of the classifier to obtain the probability value of the feature vector representing the time series of the earthworm surface state change belonging to each classification label, wherein the classification labels include fully soaked and not fully soaked; The classification label corresponding to the largest probability value among the probability values ​​is determined as the classification result.

Images