Poria cocos strain activity detection method and deviceTechnical Field
The invention relates to the field of biological monitoring, in particular to a poria cocos strain activity detection method and device.
Background
Poria cocos is fungus widely applied to traditional Chinese medicine production, and quality and yield of the fungus are closely related to activity of Poria cocos strains. In actual production, the activity of the Poria cocos strain directly affects the growth rate, biomass and final yield and medicinal value of Poria cocos. Therefore, how to quickly and accurately evaluate the activity of the tuckahoe strain is of great importance. The traditional poria cocos activity detection method generally depends on manual observation and laboratory analysis, and has the problems of long detection period, low accuracy and complex operation.
With the development of machine vision and intelligent analysis technology, detection methods based on image processing and feature extraction are becoming research hotspots. By capturing the growth state of the tuckahoe strain and combining the hypha growth time sequence characteristics under different temperature conditions, the quick and non-contact monitoring of the strain activity can be realized. By utilizing machine vision and intelligent algorithm, image data in the growth process of the poria cocos can be automatically collected, characteristics of hypha growth speed, biomass and the like are extracted, and finally the vitality of the strain is judged.
The prior art does not provide an effective solution for the real-time detection and monitoring of the activity of the poria cocos strains under different temperature conditions.
Disclosure of Invention
The invention provides a method and a device for detecting the activity of a poria cocos strain, which aim to realize the real-time monitoring of the growth state of the poria cocos strain at different temperatures.
The invention provides the following technical scheme:
in one aspect, the application provides a method for detecting the activity of a poria cocos strain, which comprises the following steps:
Step S1, tuckahoe strain image acquisition;
S2, preprocessing the image acquired in the step S1;
step S3, extracting characteristics of the preprocessed image, and extracting key growth characteristics of hyphae, including growth speed, biomass and morphological characteristics;
and S4, quantitatively evaluating the activity of the poria cocos strain by combining the extracted growth speed, biomass and morphological characteristics.
In a possible implementation manner, in the step S1, the tuckahoe strain is inoculated on a plurality of culture mediums, the different culture mediums are placed in different temperature environments, the temperature environments are constant, and the fixed shooting angle and the fixed illumination condition are maintained in the image acquisition process.
In a possible implementation manner, in the step S2, the color image is converted into a gray-scale image, and the graying formula is:
Igray=0.299×R+0.587×G+0.114×B
Wherein R, G, B is the red, green, blue channel pixel value of the image, and Igray is the generated gray value.
In one possible implementation, in the step S1, a gaussian filtering algorithm is used to eliminate noise, and a kernel function formula of the gaussian filtering is:
where σ is the standard deviation of the gaussian kernel.
In a possible implementation manner, in the step S2, a histogram equalization technology is used to enhance the contrast of the image, including:
Wherein I is the gray value of the original image, Imin and Imax are the minimum and maximum gray values in the image, respectively, and I' is the enhanced gray value.
In one possible embodiment, in step S3, the growth rate of the mycelium is calculated by analysis of successive time-series images:
Wherein Vgrowth is the growth rate of hyphae, at+1 and a represent the hyphae coverage area at time points t+1 and t, respectively, and Δt is the time interval.
In one possible embodiment, the biomass of the mycelium is estimated by a biomass model:
B=ρ×A
wherein B is biomass of hyphae, A is coverage area of the hyphae, and ρ is a hyphae density coefficient determined by experiments.
In one possible embodiment, the morphological features include length, width, shape factor of the mycelium, which can be calculated by the following formula:
wherein A is the coverage area of hyphae, P is the perimeter of the hyphae, morphological characteristics reflect the structure and growth health condition of the hyphae, and the hyphae with irregular morphology usually indicate abnormal growth.
In one possible implementation, a time series characteristic model is constructed for growth rate and biomass data collected at different time points:
Wherein y (t) represents the growth characteristics (growth rate, biomass or morphological characteristics) at the current moment, c the mean value of the time series, phii is an autoregressive coefficient, capture the influence of historical data on the current state, y(t-i) the observed value at the ith past time point, thetaj the moving average coefficient is used for correcting the residual error in the model, epsilon(t-j) the error term at the jth past time point, epsilon t is a random error term, and p and q are the orders of autoregressive and moving average respectively.
In one possible implementation, the predicted output value of the ARIMA model is used to replace the real-time measured features to construct an improved activity function:
wherein L (T) represents the activity value of the strain at the temperature T,Is the predicted growth rate of ARIMA,Is the biomass change rate predicted by ARIMA, represents the hypha coverage area growth rate,Is the morphological characteristic value predicted by ARIMA, reflects the health state of the strain, and w1,w2,w3 is a weight parameter which represents the contribution degree of each characteristic to the activity value.
On the other hand, the application provides a tuckahoe strain activity detection device, which comprises a tuckahoe strain image acquisition module, a tuckahoe strain detection module and a tuckahoe strain detection module, wherein the tuckahoe strain image acquisition module is used for acquiring real-time growth images of tuckahoe strains under different temperature conditions;
The poria cocos strain image preprocessing module is used for processing the acquired images to improve the accuracy of subsequent analysis, and comprises denoising, graying and contrast enhancement;
The poria cocos strain image feature extraction module is used for carrying out feature extraction on the preprocessed image and extracting key growth features of hyphae, including growth speed, biomass and morphological features;
And the poria cocos strain activity analysis module is used for analyzing time sequence characteristics of hypha growth at different temperatures and judging the activity of the strain.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.
In the invention, the growth speed, the biomass change rate and the morphological characteristic value predicted by using the ARIMA model can be used for generating a vitality curve at the future moment. By carrying out time sequence analysis on the predicted vitality values, the vitality trend of the strain in a future period of time can be judged. The trend analysis can discover the risk of the activity reduction of the strain in advance, and adjust the environmental conditions in time to keep the optimal growth state. Short-term viability trend analysis can identify short-term growth state changes of the strain by predicting viability values for a few hours or days in the future. The method has guiding significance for the instant regulation and control of the strain. The long-term predictive power of the ARIMA model can provide basis for the growth plan of the Poria cocos strain in a longer time range, and help researchers optimize the culture scheme.
By analyzing the value of the vitality function, the vitality of the poria cocos strain can be judged. The strain has high activity, high growth speed, obviously increased biomass, good shape and optimal state. The medium activity indicates that the strain has stable growth state and the mycelium has better performance but no prominence. The low activity indicates slow growth of the strain, limited increase of biomass and possible abnormality of morphology. By analyzing the vitality of the data in different time periods, the change trend of the vitality of the strain can be obtained. When the temperature is gradually increased or decreased, the vitality function value shows the decrease or recovery of the vitality of the strain. This helps to select a suitable temperature environment to maintain the optimal viability of the Poria cocos strain.
Drawings
Fig. 1 is a schematic diagram of a method for detecting activity of a poria cocos strain according to an embodiment of the present invention;
FIG. 2 is a flow chart of the activity analysis of Poria cocos strains at different temperatures.
Detailed Description
Embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.
A poria cocos strain activity detection method is characterized in that a poria cocos strain image acquisition module is used for acquiring real-time growth images of poria cocos strains under different temperature conditions, and data support is provided for subsequent analysis. Firstly, inoculating poria cocos strains on a plurality of different culture mediums, placing the different culture mediums in different temperature environments for culturing, and shooting by a high-definition camera or an industrial camera at regular time.
The Poria strain is inoculated onto culture medium, and cultured in different temperature environments such as 25 deg.C, 30 deg.C, 35 deg.C, 40 deg.C, etc. The temperature environments are kept constant to ensure the accuracy of the data. Image acquisition was performed using a high resolution industrial camera (. Gtoreq.2000X 2000 pixels) to capture details of hypha growth. The time interval for image acquisition is set according to the experimental requirement, and typically, an image is acquired every 1 hour to record the dynamic growth process of hyphae. Shorter time intervals may also be provided if higher time accuracy is required. And in the image acquisition process, a fixed shooting angle and a fixed illumination condition are kept, so that the consistency of the images of hyphae at all time points is ensured. The light sources need to be uniformly distributed, and the LED uniform light sources are adopted, so that illumination consistency is ensured, and the influence of overexposure or shadow is avoided.
The acquired images require a series of preprocessing operations to ensure the accuracy of the image analysis. After the image acquisition is completed, the image preprocessing module is responsible for carrying out a series of processing on the original image so as to improve the accuracy and the robustness of the subsequent analysis.
Because the growth of the tuckahoe mycelium is mainly reflected on morphological characteristics, the calculation complexity can be reduced by converting a color image into a gray image, and enough characteristic information is reserved. The graying formula is:
Igray=0.299×R+0.587×G+0.114×B
Wherein R, G, B is the red, green, blue channel pixel value of the image, and Igray is the generated gray value.
The acquired image may contain noise that affects subsequent analysis. A gaussian filtering algorithm is used to cancel the noise. The gaussian filter can smooth the image by convolution operation, and the kernel function formula of the gaussian filter is as follows:
where σ is the standard deviation of the gaussian kernel.
The difference between the hypha and the background is ensured to be more obvious, the contrast of the image is enhanced by adopting a histogram equalization technology, and the distribution of pixel values in the image is adjusted, so that the details of the hypha are more prominent. The pixel value after histogram equalization can be calculated by the following formula:
Wherein I is the gray value of the original image, Imin and Imax are the minimum and maximum gray values in the image, respectively, and I' is the enhanced gray value.
And extracting key growth characteristics of hyphae, including growth speed, biomass and morphological characteristics, by a characteristic extraction module.
By analyzing the continuous time series images, the growth rate of hyphae was calculated. Growth rate is defined as the rate of change of mycelium coverage over a fixed time interval. The specific formula is as follows:
Wherein Vgrowth is the growth rate of hyphae, at+1 and a represent the hyphae coverage area at time points t+1 and t, respectively, and Δt is the time interval.
The biomass of the mycelium can be calculated from the coverage area of the mycelium and the experimentally determined density coefficient. Separating the mycelium region from the background through binarization processing of the image, further estimating the coverage area of the mycelium, and estimating the biomass of the mycelium through a biomass model:
B=ρ×A
wherein B is biomass of hyphae, A is coverage area of the hyphae, and ρ is a hyphae density coefficient determined by experiments.
Morphological characteristics are another key indicator of the health condition of the mycelium, and include the length, width, shape factor, etc. of the mycelium. The shape factor can be calculated by the following formula:
wherein A is the coverage area of hyphae, P is the perimeter of the hyphae, morphological characteristics reflect the structure and growth health condition of the hyphae, and the hyphae with irregular morphology usually indicate abnormal growth.
By analyzing the time sequence characteristics (such as the change curve of growth speed and biomass) extracted in the mycelium growth process, the activity of the strain can be judged more accurately.
And constructing a time sequence characteristic model for the growth speed and biomass data acquired at different time points. And (3) fitting a growth curve by using an ARIMA model to obtain the growth trend of the poria cocos mycelia at different temperatures. The collected poria cocos mycelium growth state data (growth speed, biomass and morphological characteristics) are used as time series input for time series analysis. The ARIMA model is suitable for analyzing time series data with seasonal trend or fluctuation, predicting the characteristic change at the current moment by utilizing an Autoregressive (AR) part of the ARIMA model through hypha growth characteristics at the previous moments, eliminating trend items through differential calculation by an integral (I) part, processing nonstationary data, extracting the real growth change of hypha, reducing noise influence by calculating error items at the previous moments by a Moving Average (MA) part, and improving the prediction precision. The ARIMA model formula is:
Wherein y (t) represents the growth characteristics (growth rate, biomass or morphological characteristics) at the current moment, c the mean value of the time series, phii is an autoregressive coefficient, capture the influence of historical data on the current state, y(t-i) the observed value at the ith past time point, thetaj the moving average coefficient is used for correcting the residual error in the model, epsilon(t-j) the error term at the jth past time point, epsilon t is a random error term, and p and q are the orders of autoregressive and moving average respectively.
By fitting different characteristic data, a predicted value at each moment can be obtained, and the growth variation trend of the strain at different temperatures is reflected. The ARIMA model predicts the time sequence of three characteristics (growth speed, biomass change rate and morphological characteristics) of the growth of the Poria cocos strain. By fitting the time series data of each feature, the predicted values of the features at the future time can be obtained. These predictions are used to reflect future growth trends of the species and provide basis for viability assessment. The ARIMA model is output as a predicted growth rate, a predicted biomass change rate, and a predicted morphological feature value. These predictions can be optimized by ARIMA model to more accurately describe the growth characteristics of the mycelium.
Based on the time sequence characteristic model, constructing a vitality function for quantifying the growth condition of the strain. It is assumed that the growth rate of hyphae is highest and biomass increases most rapidly at a suitable temperature, while the growth rate and biomass changes slow or stop at an unsuitable temperature. The predicted output value of the ARIMA model is used to construct an improved activity function instead of the real-time measured feature. Based on these predictive features, the viability function is able to more accurately quantify the growth performance of the species. The activity function expression is as follows:
wherein L (T) represents the activity value of the strain at the temperature T.Is the growth rate predicted by ARIMA.Is the biomass change rate predicted by ARIMA and represents the hypha coverage area growth rate.Is the morphological characteristic value predicted by ARIMA and reflects the health state of the strain. w1,w2,w3 is a weight parameter representing the degree of contribution of each feature to the vitality value.
The growth rate, biomass change rate and morphological feature values predicted using the ARIMA model can be used to generate a viability curve at a future time. By carrying out time sequence analysis on the predicted vitality values, the vitality trend of the strain in a future period of time can be judged. The trend analysis can discover the risk of the activity reduction of the strain in advance, and adjust the environmental conditions in time to keep the optimal growth state. Short-term viability trend analysis can identify short-term growth state changes of the strain by predicting viability values for a few hours or days in the future. The method has guiding significance for the instant regulation and control of the strain. The long-term predictive power of the ARIMA model can provide basis for the growth plan of the Poria cocos strain in a longer time range, and help researchers optimize the culture scheme.
By analyzing the value of the vitality function, the vitality of the poria cocos strain can be judged. The strain has high activity, high growth speed, obviously increased biomass, good shape and optimal state. The medium activity indicates that the strain has stable growth state and the mycelium has better performance but no prominence. The low activity indicates slow growth of the strain, limited increase of biomass and possible abnormality of morphology. By analyzing the vitality of the data in different time periods, the change trend of the vitality of the strain can be obtained. When the temperature is gradually increased or decreased, the vitality function value shows the decrease or recovery of the vitality of the strain. This helps to select a suitable temperature environment to maintain the optimal viability of the Poria cocos strain.
The application also provides a poria cocos strain activity detection device, which comprises:
The poria cocos strain image acquisition module is used for acquiring real-time growth images of the poria cocos strain under different temperature conditions;
The poria cocos strain image preprocessing module is used for processing the acquired images to improve the accuracy of subsequent analysis, and comprises denoising, graying and contrast enhancement;
The poria cocos strain image feature extraction module is used for carrying out feature extraction on the preprocessed image and extracting key growth features of hyphae, including growth speed, biomass and morphological features;
And the poria cocos strain activity analysis module is used for analyzing time sequence characteristics of hypha growth at different temperatures and judging the activity of the strain.
The present invention is not limited to the above embodiments, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present invention, and the embodiments and features of the embodiments can be combined without conflict. Therefore, the protection scope of the invention is subject to the protection scope of the claims.