Traditional Chinese medicine specimen collection and informatization processing method based on image machine learning

Abstract

The invention belongs to the technical field of image recognition, and discloses a traditional Chinese medicine specimen collection and informatization processing method based on image machine learning, and the method comprises the steps: carrying out the multispectral collection of a traditional Chinese medicine specimen through a special collection platform, and guaranteeing the data accuracy through a black and white bottom plate and a graduated scale; carrying out geometric correction, color correction and enhancement processing on the acquired image by using an image processing technology; extracting multi-dimensional features based on an improved SIFT algorithm and a deep convolutional neural network, and constructing a specimen feature map; performing specimen attribute identification in combination with a traditional Chinese medicine knowledge base, and establishing a traditional Chinese medicine specimen informatization model; specimen recognition is realized through multi-dimensional feature matching; evaluating the growth state of the specimen in the collection environment based on the ecological adaptability model, and verifying the accuracy of the recognition result; and finally generating a traditional Chinese medicine specimen digital file containing comprehensive information. According to the invention, digitization, intelligentization and standardization of collection, identification and management of the traditional Chinese medicine specimens are realized, and the accuracy and efficiency of identification of the traditional Chinese medicine specimens are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and more specifically, to a method for collecting and informationizing traditional Chinese medicine specimens based on image machine learning. Background Art

[0002] As an important part of the traditional Chinese medicine system, traditional Chinese medicine has a history of thousands of years of application and a rich theoretical system; accurately identifying and standardizing the management of traditional Chinese medicine specimens is the basis for traditional Chinese medicine research, development, and clinical applications. However, the traditional collection and identification of traditional Chinese medicine specimens mainly rely on expert experience, with limitations such as strong subjectivity, inconsistent standards, and simple recording methods, making it difficult to meet the needs of modern traditional Chinese medicine research and industrial development.

[0003] The traditional collection and informationization processing of traditional Chinese medicine specimens mainly have the following problems:

[0004] First, the traditional collection method of traditional Chinese medicine specimens lacks standardized collection procedures and equipment; the environmental parameters during the collection process are not comprehensively recorded, and the lighting conditions are unstable, resulting in uneven quality of the obtained specimen images and making it difficult to use them for subsequent precise analysis.

[0005] Second, the existing digital technologies for traditional Chinese medicine specimens mostly use simple photo archiving methods and lack the ability to collect multi-dimensional and multi-spectral images, which results in the obtained images being unable to comprehensively reflect the key information such as the morphological characteristics, texture characteristics, and color characteristics of traditional Chinese medicine specimens, especially the characteristic differences under different spectral conditions.

[0006] Third, the traditional identification of traditional Chinese medicine specimens mainly relies on expert experience and lacks objective and quantitative feature extraction and pattern recognition methods; with the increase in the types of traditional Chinese medicine and the lack of professional talents, purely relying on manual identification has become difficult to meet the needs of large-scale and high-precision classification of traditional Chinese medicine specimens.

[0007] Fourth, the traditional management method of traditional Chinese medicine specimens lacks an objective verification mechanism for the authenticity and reliability of specimens; the identification results of specimens lack credibility assessment, and misjudgment and confusion are likely to occur, affecting the accuracy of subsequent research and applications.

[0008] In view of this, the present invention proposes a method for collecting and informationizing traditional Chinese medicine specimens based on image machine learning to solve the above problems. Summary of the Invention

[0009] In order to overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solutions:

[0010] A method for collecting and informationizing traditional Chinese medicine specimens based on image machine learning, comprising:

[0011] Step 1: Based on a pre - constructed specimen collection platform, multi - dimensional data of traditional Chinese medicine specimens are collected to obtain corresponding specimen sample images, and the collected specimen metadata are recorded; they are stored in a pre - constructed specimen database.

[0012] Step 2: Feature extraction is performed on the collected specimen sample images, and a corresponding specimen feature map is constructed in combination with the constructed specimen database.

[0013] Step 3: Specimen attribute recognition is performed on the obtained specimen feature map, and a corresponding informatization model of traditional Chinese medicine specimens is constructed based on the specimen attribute recognition results.

[0014] Step 4: Multi - dimensional feature matching processing is performed on the constructed informatization model of traditional Chinese medicine specimens to obtain corresponding specimen recognition results.

[0015] Step 5: The growth environment of corresponding traditional Chinese medicine specimens is simulated respectively based on the specimen recognition results and the collected environmental information, and the accuracy of the corresponding specimen recognition results is evaluated based on the growth environment simulation results to obtain corresponding credibility indicators.

[0016] Step 6: Multi - dimensional comprehensive evaluation is performed on the credibility indicators of the specimen recognition results, and the evaluation results are integrated with the traditional Chinese medicine specimen information to obtain corresponding informatized traditional Chinese medicine specimen data files.

[0017] Furthermore, the process of multi - dimensional data collection of traditional Chinese medicine specimens includes:

[0018] Construct a specimen collection platform; based on pre - set collection requirements, the image acquisition terminal uses a black - and - white bottom plate to collect images of the traditional Chinese medicine specimens to be collected in different light - supplementing modes, obtaining corresponding specimen sample images and specimen sample videos.

[0019] Simultaneously, geographic information is collected based on a positioning terminal; and the metadata information corresponding to each specimen sample image is recorded, and it is associated with the collected sample image set to generate specimen metadata; the collected specimen metadata are uploaded and stored in a pre - constructed specimen database.

[0020] Furthermore, the light - supplementing mode refers to the natural light mode and the fluorescence mode provided by the light - supplementing terminal. The natural light mode refers to guiding the environmental light source to irradiate the specimen through the natural light guiding structure built in the light - supplementing terminal; the fluorescence mode refers to using an ultraviolet LED array with adjustable wavelength and a multi - band infrared light source built in, so that the corresponding image acquisition terminal performs image acquisition on the traditional Chinese medicine specimens to be collected in multi - spectral bands based on ultraviolet light and infrared light respectively.

[0021] Furthermore, the process of feature extraction of the obtained specimen sample images includes:

[0022] Obtain the images of the collected specimen samples. Based on the scale ruler on the corresponding black-and-white base plate during the image acquisition process, perform geometric correction on the corresponding specimen sample images, and perform polynomial color correction and white balance adjustment on the geometrically corrected specimen sample images; obtain the corresponding initial sample images;

[0023] Perform local mean filtering on the obtained initial sample images to remove Gaussian noise in the images and simultaneously retain edges and texture details. Among them, the formula for performing non-local mean filtering is: In the formula, BF[I] p represents the pixel value at pixel point p after local mean filtering; S represents the spatial domain, that is, the neighborhood range of the corresponding pixel point p; q represents the pixel point index within the neighborhood range; and respectively represent the spatial domain kernel function and the range domain kernel function; I p and I q respectively represent the pixel values at the corresponding pixel point p and pixel point q, and p≠q; W p represents the normalization coefficient;

[0024] After the local mean filtering is completed, obtain the specimen type corresponding to the corresponding traditional Chinese medicine specimen, and perform adaptive enhancement on the corresponding traditional Chinese medicine specimen based on it to obtain the corresponding enhanced specimen image, and obtain the corresponding standard specimen image based on it;

[0025] Extract the feature points in the corresponding standard specimen image based on the improved SIFT algorithm to obtain the corresponding image feature points;

[0026] Mark the feature points on the adaptively enhanced initial sample image based on the obtained image feature points and generate the corresponding feature distribution map;

[0027] Extract the morphological features, texture features, and color features from the corresponding standard specimen image based on the obtained feature distribution map to obtain the corresponding morphological feature vectors, texture feature vectors, and color feature vectors; at the same time, use a 3D convolutional neural network to process the collected specimen sample videos to capture the change features of traditional Chinese medicine specimens in the time dimension and obtain the corresponding video dynamic features; for example, the fluorescence decay process or the appearance changes of specimens at different angles;

[0028] Fuse the obtained morphological feature vectors, texture feature vectors, color feature vectors, and video dynamic features in different spectral bands with the collected geographical location information to obtain the corresponding comprehensive feature vectors.

[0029] Furthermore, the process of extracting the feature points in the corresponding standard specimen image based on the improved SIFT algorithm to obtain the corresponding image feature points includes:

[0030] Perform Gaussian convolution processing on the standard specimen image based on Gaussian convolution kernels at different scales respectively to obtain Gaussian convolution images at different scales; perform image difference calculation on the Gaussian convolution images at adjacent scales to obtain corresponding difference images; obtain the pixel values corresponding to the pixel points in the corresponding difference images, and compare them with the pixel values corresponding to all pixel points in the neighborhood pixel set. Based on the comparison results, determine whether the pixel value corresponding to the corresponding pixel point is a local extreme point. If not, do not perform any other operations. If so, mark the corresponding pixel point as a candidate key point;

[0031] Furthermore, obtain the curvature coefficients corresponding to each candidate key point, and compare them with a preset curvature threshold. If the curvature coefficient is not greater than the curvature threshold, mark the corresponding candidate key point as an image feature point. If the curvature coefficient is greater than the curvature threshold, discard the corresponding candidate key point.

[0032] Further, the process of identifying the specimen attributes of the obtained specimen feature map and constructing a corresponding information model of traditional Chinese medicine specimens based on the specimen attribute identification results includes:

[0033] Based on the existing knowledge base of traditional Chinese medicine, obtain a dataset of traditional Chinese medicine attributes related to traditional Chinese medicine; extract attribute features of the corresponding traditional Chinese medicine specimens based on the obtained specimen feature map to obtain corresponding specimen attribute features

[0034] Associate the obtained specimen attribute features with the dataset of traditional Chinese medicine attributes, and construct a corresponding classification framework based on them; the process of constructing the classification framework is as follows: based on the association relationship between the key attribute information in the dataset of traditional Chinese medicine attributes and the specimen attribute features, visualize it as an attribute association network, and mark the association strength and direction between each key attribute information and the specimen attribute features; build a corresponding classification framework based on the attribute association network;

[0035] Integrate the classification framework with the obtained specimen feature map to obtain a corresponding information model of traditional Chinese medicine specimens. The information model of traditional Chinese medicine specimens includes the feature data, attribute associations and classification structures of traditional Chinese medicine specimens, as well as relevant traditional Chinese medicine knowledge.

[0036] Further, the process of obtaining the specimen identification result includes:

[0037] Match the feature of the information database of traditional Chinese medicine specimens corresponding to the corresponding traditional Chinese medicine specimen with the known traditional Chinese medicine specimens to generate a corresponding initial matching result. The initial matching result includes multiple sample matching items;

[0038] Sort the known traditional Chinese medicine specimens based on the comprehensive similarity coefficient to obtain the corresponding similarity ranking sequence; pre-screen the known traditional Chinese medicine specimens in the similarity ranking sequence based on a preset similarity threshold, and use the known traditional Chinese medicine specimens with a comprehensive similarity coefficient higher than the similarity threshold as candidate specimens; summarize all the obtained candidate specimens to obtain the corresponding specimen identification results, and the specimen identification results include information such as the specimen ID, name, and similarity data of the known traditional Chinese medicine specimens; the similarity data includes the similarity calculation results and the comprehensive similarity coefficient in multiple dimensions.

[0039] Further, the process of obtaining the credibility index includes:

[0040] Extract parameters from the traditional Chinese medicine specimen information database to obtain the ecological environment data related to the corresponding traditional Chinese medicine specimens; and based on this, extract the range of environmental parameter requirements for the suitable growth of the traditional Chinese medicine specimens to obtain the corresponding specimen growth environment requirement parameter set;

[0041] Based on a pre-constructed ecological adaptability model, obtain the adaptability of the corresponding traditional Chinese medicine specimens to different environmental factors to obtain the corresponding ecological adaptability index;

[0042] Define the mathematical formula of the ecological adaptability model as:

[0043] In the formula, μ x (i) represents the ecological adaptability index corresponding to the i-th environmental factor; x i represents the specific value of the i-th environmental factor; T i,min and T i,max respectively represent the left boundary and the right boundary of the parameters corresponding to the i-th environmental factor; obtained from the specimen growth environment requirement parameter set; δ represents the width of the transition interval;

[0044] Furthermore, based on the ecological adaptability index and combined with the specimen growth environment requirement parameter set, simulate the growth state and performance of the corresponding traditional Chinese medicine specimens under different environmental factors, and record the key data points during the simulation process. The key data points include growth rate, morphological changes, physiological indicators, etc., and integrate the simulation results to generate a complete specimen ecological adaptability data set;

[0045] Obtain the metadata of the collected specimens, and based on this, obtain the environmental data recorded during the collection process of the corresponding traditional Chinese medicine specimens. The environmental data includes environmental parameters such as temperature, humidity, light, altitude, etc., and perform standardization processing on it to obtain the corresponding environmental condition parameter set;

[0046] Obtain the ecological adaptability indicators corresponding to each environmental parameter within the corresponding environmental data based on the ecological adaptability model, and simulate the growth state and performance of traditional Chinese medicine specimens under the current environmental data based on them to obtain the corresponding ecological data set of environmental specimens;

[0047] Obtain the ecological data set of environmental specimens and the data set of specimen ecological adaptability obtained, and evaluate the matching degree between the two to obtain the corresponding environmental matching degree indicator; the process of obtaining the corresponding environmental matching degree indicator is as follows: correspond the environmental factors and environmental parameters in the corresponding ecological data set of environmental specimens and the data set of specimen ecological adaptability one by one, and obtain the parameter difference value and the difference value of the ecological adaptability indicator between the two, and calculate the adaptability score in combination with the pre-set matching function to obtain the corresponding environmental matching degree indicator;

[0048] Integrate the obtained environmental matching degree indicators to obtain the corresponding specimen ecological environment evaluation data, and the specimen ecological environment evaluation data includes ecological adaptability indicators, environmental matching degree indicators, and the best growth interval, etc.;

[0049] Furthermore, obtain the ecological growth requirements corresponding to each candidate specimen based on the sample recognition result, and conduct a consistency comparison with the obtained specimen ecological environment evaluation data;

[0050] If the two are inconsistent, discard the corresponding candidate specimen;

[0051] If the two are consistent, obtain the matching degree between the environmental parameters and the ecological growth requirements corresponding to each candidate specimen obtained from the corresponding sample recognition result to obtain the corresponding environmental verification result;

[0052] Furthermore, correct the sample recognition result corresponding to the corresponding candidate specimen based on the environmental verification result to adjust the error in the sample recognition result and improve the accuracy;

[0053] Furthermore, obtain the classification rules and identification key points of each type of traditional Chinese medicine based on the knowledge base of traditional Chinese medicine experts, and obtain the evaluation factors of the corresponding specimen recognition result based on them. The evaluation factors include the morphological feature compliance, ecological environment consistency, distribution area overlap rate, etc.;

[0054] Perform weighted summation on the obtained evaluation factors to obtain the corresponding credibility indicator.

[0055] Further, the construction process of the informatized traditional Chinese medicine specimen data file includes:

[0056] Define the recognition criteria for the credibility index, and obtain the corresponding credibility level of the specimen recognition result based on it; obtain the information-based model of traditional Chinese medicine specimens corresponding to the candidate specimens in the corresponding sample recognition result, and compare its data with the information-based model of traditional Chinese medicine specimens corresponding to the traditional Chinese medicine specimens to be collected, obtain the corresponding data coverage rate and accuracy rate, and generate the corresponding data integrity evaluation value based on it;

[0057] Integrate the obtained credibility level and data integrity evaluation value, and synchronously integrate the relevant information involved in the corresponding traditional Chinese medicine specimens to generate a digital file of information-based traditional Chinese medicine specimens.

[0058] The technical effects and advantages of a method for collecting and information-based processing of traditional Chinese medicine specimens based on image machine learning according to the present invention:

[0059] In the collection link, the standardized specimen collection platform and process overcome the drawbacks of the lack of norms in traditional collection. The scalable box adapts to different specimens, and the stable light supplement mode and comprehensive environmental parameter recording ensure high-quality images and comprehensive and accurate data collection, providing a reliable basis for subsequent precise analysis; the multi-dimensional and multi-spectral image collection ability makes up for the deficiencies of traditional simple photo archiving, and can comprehensively reflect the key features of the specimen such as morphology, texture, color, etc. and their differences under different spectra; in the processing link, the feature extraction and pattern recognition methods based on image machine learning change the traditional subjective identification method that relies on expert experience, and achieve high-precision specimen classification through objective quantification means, meeting the needs of large-scale identification of traditional Chinese medicine specimens. At the same time, the constructed specimen feature map, information-based model and credibility evaluation mechanism not only help to deeply study the characteristics of traditional Chinese medicine specimens, but also can objectively verify the specimen identification results, avoid misjudgment and confusion, and improve the authenticity and reliability of specimen management; in addition, the finally generated digital file of information-based traditional Chinese medicine specimens is convenient for storage, retrieval and sharing, strongly promoting the development of modern traditional Chinese medicine research and industry, and enabling traditional Chinese medicine to better serve clinical applications and other fields in inheritance and innovation. Brief Description of the Drawings

[0060] Figure 1 It is a schematic diagram of a method for collecting and information-based processing of traditional Chinese medicine specimens based on image machine learning according to the present invention;

[0061] Figure 2 It is a schematic diagram of a system for collecting and information-based processing of traditional Chinese medicine specimens based on image machine learning according to the present invention. Detailed Embodiments

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0063] Embodiment 1

[0064] Please refer to Figure 1 As shown, a method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning in this embodiment includes:

[0065] Step 1: Collect multi-dimensional data of traditional Chinese medicine specimens based on a pre-constructed specimen collection platform, obtain corresponding specimen sample images, and record the collected specimen metadata; store them in a pre-constructed specimen database.

[0066] It should be further noted that in the specific implementation process, the process of obtaining corresponding specimen sample images and recording the collected specimen metadata includes:

[0067] Construct a specimen collection platform, which consists of a telescopic box body, an image acquisition terminal, a positioning terminal, a supplementary light terminal, and a replaceable black-and-white bottom plate; among them, high-precision scale rulers are installed in the black-and-white bottom plates for scale reference.

[0068] Estimate the size of the traditional Chinese medicine specimen to be collected, and adjust the size parameters of the telescopic box body in the corresponding specimen collection platform based on it to make it adapt to traditional Chinese medicine specimens of different specifications.

[0069] After the size adjustment is completed, initialize the acquisition parameters in the image acquisition terminal. After the initialization is completed, the image acquisition terminal uses the black-and-white bottom plate to collect images of the traditional Chinese medicine specimen to be collected in different supplementary light modes based on the preset acquisition requirements, and obtain corresponding specimen sample images and specimen sample videos.

[0070] Among them, the supplementary light mode refers to the natural light mode and the fluorescence mode provided by the supplementary light terminal. The natural light mode guides the ambient light to irradiate the specimen through the natural light guiding structure built in the supplementary light terminal; the natural light introduction structure realizes the precise control of natural light through an adjustable-angle window, equipped with a light intensity sensor and an automatic adjustment light-shielding plate; the fluorescence mode refers to using the built-in ultraviolet LED array with adjustable wavelength and a multi-band infrared light source, so that the corresponding image acquisition terminal respectively performs multi-spectral band image acquisition of the traditional Chinese medicine specimen to be collected based on ultraviolet light and infrared light.

[0071] Synchronization is based on the positioning terminal to collect geographical information, which includes the precise coordinates and altitude of the specimen collection location; and record metadata information such as the spectral parameters of each image, the specimen collection angle, the scale value, and geographical information, and associate it with the collected sample image set to generate specimen metadata; the specimen metadata includes complete information such as the collection time, location, lighting conditions, and camera parameters to ensure data traceability;

[0072] After the collection is completed, the collected specimen sample images, specimen sample videos, and the associated specimen metadata are encrypted and uploaded. During the upload process, the optimal transmission channel is adaptively selected according to the current network environment, and the relay nodes used during the transmission are increased or decreased to ensure the data upload efficiency and data integrity; after the upload is completed, the encrypted and packaged specimen sample images, specimen sample videos, and the associated specimen metadata are decrypted and subjected to integrity verification. After the verification passes, they are stored in a pre-constructed specimen database.

[0073] Step 2: Extract the features of the collected specimen sample images and construct corresponding specimen feature maps in combination with the constructed specimen database;

[0074] It should be further noted that in the specific implementation process, the process of extracting the features of the obtained specimen sample images includes:

[0075] Obtain the collected specimen sample images, and based on the scale ruler on the corresponding black and white bottom plate during the image collection process, perform geometric correction on the corresponding specimen sample images. Among them, the process of geometric correction includes: identifying the scale ruler in the sample specimen image and establishing a mapping relationship between the image scale size and the actual scale size;

[0076] Furthermore, based on the established mapping relationship, perform geometric correction, and use the perspective transformation algorithm to correct the perspective distortion in the image to ensure the accurate geometric shape of the specimen in the image;

[0077] After the geometric correction is completed, based on the black and white bottom plate in the corresponding specimen sample image as the color reference, perform polynomial color correction and white balance adjustment on the geometrically corrected sample specimen image; obtain the corresponding initial sample image; polynomial color correction is an image color consistency processing method that maps the pixel values in the color channels under different lighting conditions to the same reference color by constructing a non-linear mapping function. The color channels include three channel colors: red, green, and blue;

[0078] Perform non-local mean filtering on the obtained initial sample image to remove Gaussian noise in the image and at the same time retain the edges and texture details. Among them, the formula for performing non-local mean filtering is: In the formula, BF[I]p denotes the pixel value at pixel point p after local mean filtering; S represents the spatial domain, that is, the neighborhood range of the corresponding pixel point p; q represents the pixel point index within the neighborhood range; and respectively represent the spatial domain kernel function and the range domain kernel function; I p and I q respectively represent the pixel values at the corresponding pixel point p and pixel point q, and p≠q; W p represents the normalization coefficient;

[0079] After the local mean filtering is completed, obtain the specimen type corresponding to the corresponding traditional Chinese medicine specimen, and perform adaptive enhancement on the corresponding traditional Chinese medicine specimen based on it to obtain the corresponding enhanced specimen image; for example: for the fruit and flower specimens with distinct colors, use the non-linear contrast enhancement algorithm; for the rhizome specimens with complex morphological structures, use the adaptive histogram equalization (CLAHE) algorithm to enhance the texture details; for the leaf specimens with low contrast, use the multi-scale enhancement algorithm based on the Retinex theory to improve the detail visibility; the enhancement parameters involved in the corresponding adaptive enhancement process are dynamically adjusted according to the statistical characteristics of the image (such as brightness mean, contrast, information entropy, etc.) to avoid image distortion caused by over-enhancement;

[0080] Perform image segmentation on the specimen foreground and standard background in the initial sample image after adaptive enhancement, and extract the pure traditional Chinese medicine specimen area based on the image segmentation result to obtain the corresponding standard specimen image;

[0081] Extract the feature points in the corresponding standard specimen image based on the improved SIFT algorithm to obtain the corresponding image feature points; among them, the process of obtaining the image feature points includes:

[0082] Perform Gaussian convolution processing on the standard specimen image based on the Gaussian convolution kernels at different scales respectively to obtain the Gaussian convolution images at different scales. Furthermore, perform image difference calculation on the Gaussian convolution images at adjacent scales to obtain the corresponding difference images;

[0083] Taking a certain pixel point in the difference image as an example, obtain the pixel value corresponding to the corresponding pixel point, and compare it with the pixel values corresponding to all pixel points in the neighborhood pixel set. Based on the comparison result, determine whether the pixel value corresponding to the corresponding pixel point is a local extreme point. If not, do not perform any other operations. If so, mark the corresponding pixel point as a candidate key point; the neighborhood pixel set includes other pixel points adjacent to this pixel point in the corresponding difference image and all pixel points at the same position and adjacent positions in the corresponding difference images at adjacent scales;

[0084] Furthermore, obtain the curvature coefficients corresponding to each candidate key point, and compare them with a preset curvature threshold. If the curvature coefficient is not greater than the curvature threshold, mark the corresponding candidate key point as an image feature point; if the curvature coefficient is greater than the curvature threshold, discard the corresponding candidate key point. The process of obtaining the curvature coefficient includes: obtaining the pixel value corresponding to the corresponding candidate key point, and respectively obtaining its second-order partial derivatives in the horizontal direction, vertical direction, and the mixed horizontal and vertical directions; and constructing a corresponding curvature matrix based on them, and obtaining the ratio between the eigenvalues corresponding to the corresponding curvature matrix, which is the required curvature coefficient.

[0085] Based on the obtained image feature points, mark the initial sample image after adaptive enhancement, and generate a corresponding feature distribution map, which records the position, intensity, and direction information of each image feature point.

[0086] Furthermore, based on the obtained feature distribution map, extract morphological features, texture features, and color features from the corresponding standard specimen image to obtain corresponding morphological feature vectors, texture feature vectors, and color feature vectors. The corresponding feature extraction is implemented based on a pre-constructed feature extraction model. The feature extraction model uses a deep convolutional neural network with multiple input channels. The deep convolutional neural network adopts a multi-branch structure, and each branch corresponds to a spectral type (natural light, ultraviolet light, infrared light). Each branch contains 5-7 convolutional layers, pooling layers, and batch normalization layers. The convolutional kernel size ranges from 3×3 to 7×7 to adapt to the feature extraction requirements of different scales. Furthermore, based on the deep convolutional neural network, capture the feature differences of traditional Chinese medicine specimens under each spectral type to obtain the required morphological feature vectors, texture feature vectors, and color feature vectors.

[0087] At the same time, use a 3D convolutional neural network to process the collected specimen sample video to capture the change features of traditional Chinese medicine specimens in the time dimension and obtain corresponding video dynamic features, such as the fluorescence decay process or the appearance change of the specimen at different angles.

[0088] Fuse the obtained morphological feature vectors, texture feature vectors, color feature vectors, and video dynamic features in different spectral bands with the collected geographical location information to obtain corresponding comprehensive feature vectors. The feature fusion uses an attention mechanism and feature connection technology, and uses a self-attention module to perform weighted fusion on features from different sources to highlight important features and suppress redundant information. And during the feature fusion process, convert the geographical location information into a feature vector and connect it with the image features to form a comprehensive feature representation.

[0089] Furthermore, based on the atlas construction technology and combined with the specimen database, the specimen metadata, morphological feature vectors, texture feature vectors, color feature vectors, video dynamic features, and comprehensive feature vectors corresponding to the corresponding traditional Chinese medicine specimens are presented in the form of an atlas to obtain the corresponding specimen feature atlas; the specimen feature atlas adopts a multi-level and multi-dimensional atlas architecture, and the atlas architecture includes a basic information layer, an image resource layer, and a feature data layer;

[0090] The basic information layer includes recording the basic attributes of traditional Chinese medicine specimens, including basic information such as specimen type, collection time, collection location, collection personnel, and environmental parameters;

[0091] The image resource layer includes storing the original specimen sample images, standard specimen images, and intermediate results of each processing link; the feature data layer includes a morphological feature sub-layer, a color feature sub-layer, a texture feature sub-layer, a dynamic feature layer, and a comprehensive feature sub-layer; and stores the corresponding morphological feature vectors, texture feature vectors, color feature vectors, video dynamic features, and comprehensive feature vectors respectively.

[0092] Step 3: Identify the specimen attributes of the obtained specimen feature atlas, and construct the corresponding informatization model of traditional Chinese medicine specimens based on the specimen attribute recognition results;

[0093] It should be further noted that in the specific implementation process, the construction process of the informatization model of traditional Chinese medicine specimens includes:

[0094] Based on the existing traditional Chinese medicine knowledge base, obtain the traditional Chinese medicine attribute data set related to traditional Chinese medicine. The traditional Chinese medicine attribute data set includes key attribute information such as the properties, tastes, efficacy, medicinal parts, and chemical components of traditional Chinese medicine;

[0095] Extract the attribute features of the corresponding traditional Chinese medicine specimens based on the obtained specimen feature atlas to obtain the corresponding specimen attribute features. The specimen attribute features include multiple dimensions such as morphological features (such as leaf shape, flower shape, fruit morphology, etc.), anatomical features (such as tissue structure, cell features, etc.), and biochemical features (such as active ingredients, content, etc.);

[0096] Associate the obtained specimen attribute features with the traditional Chinese medicine attribute data set and construct the corresponding classification framework based on it; the construction process of the corresponding classification framework is as follows: Based on the association relationship between the key attribute information in the traditional Chinese medicine attribute data set and the specimen attribute features, visualize it as an attribute association network, and mark the association strength and direction between each key attribute information and the specimen attribute features; furthermore, build the corresponding classification framework based on the attribute association network;

[0097] Integrate the classification framework with the obtained specimen feature map to obtain the corresponding informatization model of traditional Chinese medicine specimens. The informatization model of traditional Chinese medicine specimens includes the feature data, attribute associations, and classification structures of traditional Chinese medicine specimens, as well as relevant traditional Chinese medicine knowledge.

[0098] Step 4: Perform multi-dimensional feature matching processing on the constructed informatization model of traditional Chinese medicine specimens to obtain the corresponding specimen recognition results.

[0099] It should be further noted that in the specific implementation process, the process of obtaining the specimen recognition results includes:

[0100] Match the informatization database of traditional Chinese medicine specimens corresponding to the corresponding traditional Chinese medicine specimens with the known traditional Chinese medicine specimens to generate the corresponding initial matching results. The initial matching results include multiple sample matching items. Among them, in the feature matching process, first calculate the similarity from multiple dimensions such as morphological features, color features, and texture features, and perform weighted summation on the similarity calculation results of each dimension to obtain the corresponding comprehensive similarity coefficient. Among them, in the actual application process, due to the differences between features, the similarity calculation methods used for different specimens are different. For example, the morphological features can be evaluated for similarity by calculating the Hausdorff distance; the feature similarity calculation of color features requires collecting the histogram intersection algorithm.

[0101] Sort the known traditional Chinese medicine specimens based on the comprehensive similarity coefficient to obtain the corresponding similarity ranking sequence; pre-screen the known traditional Chinese medicine specimens in the similarity ranking sequence based on a pre-set similarity threshold, and use the known traditional Chinese medicine specimens with a comprehensive similarity coefficient higher than the similarity threshold as candidate specimens; summarize all the obtained candidate specimens to obtain the corresponding specimen recognition results. The specimen recognition results include information such as the specimen ID, name, and similarity data of the known traditional Chinese medicine specimens; the similarity data includes the similarity calculation results and the comprehensive similarity coefficient in multiple dimensions.

[0102] Step 5: Simulate the growth environment of the corresponding traditional Chinese medicine specimens respectively based on the specimen recognition results and the collected environmental information, and evaluate the accuracy of the corresponding specimen recognition results based on the growth environment simulation results to obtain the corresponding credibility index.

[0103] It should be further noted that in the specific implementation process, the process of obtaining the credibility index includes:

[0104] Extract parameters from the informatization database of traditional Chinese medicine specimens to obtain ecological environment data related to the corresponding traditional Chinese medicine specimens. The ecological environment data includes the requirements for environmental factors such as temperature, humidity, light, and soil. Based on the known professional knowledge base of traditional Chinese medicine plant ecology, extract the range of environmental parameters suitable for the growth of traditional Chinese medicine specimens to obtain the corresponding parameter set of specimen growth environment requirements. The parameter set of specimen growth environment requirements includes environmental factors such as temperature adaptation range, humidity requirement, light intensity, altitude range, and soil pH value.

[0105] Furthermore, based on the pre-constructed ecological adaptability model, obtain the adaptability of the corresponding traditional Chinese medicine specimens to different environmental factors, and obtain the corresponding ecological adaptability indicators.

[0106] Define the mathematical formula of the ecological adaptability model as:

[0107] In the formula, μ x (i) represents the ecological adaptability index corresponding to the i-th environmental factor; x i represents the specific value of the i-th environmental factor; T i,min and T i,max respectively represent the left boundary and right boundary of the parameters corresponding to the i-th environmental factor; obtained from the parameter set of specimen growth environment requirements; δ represents the width of the transition interval.

[0108] Furthermore, based on the ecological adaptability indicators and combined with the parameter set of specimen growth environment requirements, simulate the growth state and performance of the corresponding traditional Chinese medicine specimens under different environmental factors, and record the key data points during the simulation process. The key data points include growth rate, morphological changes, physiological indicators, etc. Integrate the simulation results to generate a complete specimen ecological adaptability data set.

[0109] Obtain the metadata of the collected specimens, and based on it, obtain the environmental data recorded during the collection process of the corresponding traditional Chinese medicine specimens. The environmental data includes environmental parameters such as temperature, humidity, light, and altitude, and perform standardization processing on it to obtain the corresponding set of environmental condition parameters.

[0110] Based on the ecological adaptability model, obtain the ecological adaptability indicators corresponding to each environmental parameter in the corresponding environmental data, and based on it, simulate the growth state and performance of traditional Chinese medicine specimens under the current environmental data to obtain the corresponding environmental specimen ecological data set.

[0111] Obtain the ecological dataset of the acquired environmental specimens and the specimen ecological adaptability dataset, and evaluate the matching degree between the two to obtain the corresponding environmental matching degree index; the process of obtaining the corresponding environmental matching degree index is as follows: Correspond the environmental factors and environmental parameters in the corresponding environmental specimen ecological dataset and the specimen ecological adaptability dataset one by one, and obtain the parameter difference value and the difference value of the ecological adaptability index between the two, and calculate the adaptability score in combination with the pre-set matching function to obtain the corresponding environmental matching degree index;

[0112] Integrate the obtained environmental matching degree indexes to obtain the corresponding specimen ecological environment evaluation data, and the specimen ecological environment evaluation data includes ecological adaptability indexes, environmental matching degree indexes, and the best growth range, etc.;

[0113] Furthermore, based on the sample recognition results, obtain the ecological growth requirements corresponding to each candidate specimen, and compare them with the obtained specimen ecological environment evaluation data for consistency;

[0114] If the two are inconsistent, discard the corresponding candidate specimen;

[0115] If the two are consistent, obtain the matching degree between the environmental parameters and the ecological growth requirements corresponding to the corresponding sample recognition results to obtain the corresponding environmental verification results;

[0116] Furthermore, based on the environmental verification results, correct the sample recognition results corresponding to the corresponding candidate specimens to adjust the errors in the sample recognition results and improve the accuracy;

[0117] Furthermore, based on the traditional Chinese medicine expert knowledge base, obtain the classification rules and identification key points of each type of traditional Chinese medicine, and based on this, obtain the evaluation factors of the corresponding specimen recognition results. The evaluation factors include the morphological feature compliance, ecological environment consistency, distribution area overlap rate, etc.;

[0118] Perform weighted summation on the obtained evaluation factors to obtain the corresponding credibility index.

[0119] Step Six: Conduct multi-dimensional comprehensive evaluation on the credibility index of the specimen recognition results, and integrate the evaluation results with the traditional Chinese medicine specimen information to obtain the corresponding information-based traditional Chinese medicine specimen data file;

[0120] It should be further noted that in the specific implementation process, the construction process of the information-based traditional Chinese medicine specimen data file includes:

[0121] Define the recognition standard of the credibility index, and based on this, obtain the credibility level corresponding to the corresponding specimen recognition results. The credibility levels include high credibility, medium credibility, and low credibility;

[0122] Obtain the informatization model of traditional Chinese medicine specimens corresponding to the candidate specimens in the corresponding sample recognition results, compare it with the informatization model of traditional Chinese medicine specimens corresponding to the traditional Chinese medicine specimens to be collected, obtain the corresponding data coverage rate and accuracy rate, and generate the corresponding data integrity evaluation value based on it. Among them, the higher the data coverage rate and accuracy rate, the greater the corresponding data integrity evaluation value;

[0123] Integrate the obtained credibility level and data integrity evaluation value, and synchronously integrate the relevant information involved in the corresponding traditional Chinese medicine specimens to generate a digital file of informatized traditional Chinese medicine specimens. The digital file of informatized traditional Chinese medicine specimens includes specimen basic information, morphological characteristics, classification information, ecological characteristics, value evaluation, digital images, etc. Organize the generated digital file of informatized traditional Chinese medicine specimens into a standard format for convenient storage, retrieval, and sharing.

[0124] In this embodiment, by integrating the retractable box design, multi-spectral illumination technology, and intelligent image acquisition, accurate and standardized acquisition of traditional Chinese medicine specimens of different specifications is achieved; through a deep convolutional neural network, multi-dimensional features such as morphology, texture, and color of the specimen images obtained under multi-spectral conditions are extracted, and combined with the dynamic features of the specimen video and geographical information, a complete specimen feature map is constructed. Further, the extracted features are associated with the traditional Chinese medicine knowledge base to form a mapping relationship between specimen attributes and medicinal properties and effects, and the reliability of the specimen recognition results is verified through an ecological adaptability model. Finally, a structured digital file of traditional Chinese medicine specimens is generated, which has significant advantages such as strong adaptability of the acquisition equipment, comprehensive data acquisition dimensions, intelligent feature extraction and analysis, and systematic knowledge association, providing a scientific and effective informatization solution for the digital protection, research, and utilization of traditional Chinese medicine resources, and having important value for promoting the modern development of traditional Chinese medicine.

[0125] Embodiment 2

[0126] Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. Provide a traditional Chinese medicine specimen collection and informatization processing system based on image machine learning, including:

[0127] A collection module that performs multi-dimensional data collection on traditional Chinese medicine specimens based on a pre-constructed specimen collection platform, obtains the corresponding specimen sample images, records the collection specimen metadata, and stores it in a pre-constructed specimen database;

[0128] An analysis module for extracting features from the collected specimen sample images, and constructing a corresponding specimen feature map in combination with the constructed specimen database; performing specimen attribute recognition based on the feature map, and constructing a corresponding informatization model of traditional Chinese medicine specimens;

[0129] An evaluation module is used to perform multi-dimensional feature matching processing on the constructed informatization model of traditional Chinese medicine specimens to obtain corresponding specimen recognition results; based on the specimen recognition results and the collected environmental information, simulate the growth environment of the corresponding traditional Chinese medicine specimens, and based on the growth environment simulation results, evaluate the accuracy of the corresponding specimen recognition results to obtain corresponding credibility indicators;

[0130] An archiving module is used to perform multi-dimensional comprehensive evaluation on the credibility indicators of the specimen recognition results, and perform information integration based on the evaluation results to obtain corresponding informatization traditional Chinese medicine specimen data archives;

[0131] Each module is connected by wired and / or wireless means to achieve data transmission between modules.

[0132] Embodiment 3

[0133] This embodiment publicly provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the above-provided method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning.

[0134] Since the electronic device introduced in this embodiment is the electronic device used to implement a method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning in an embodiment of the present application, based on the method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning introduced in an embodiment of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiment of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device used in a method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning in an embodiment of the present application, it belongs to the scope protected by the present application.

[0135] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by software simulation of a large amount of collected data to obtain a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0136] The above is only the preferred implementation manner of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for ordinary technical users in the technical field of the present invention, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A method for collecting and informationizing traditional Chinese medicine specimens based on image machine learning, characterized in that Including: Step 1: Perform multi-dimensional data collection on traditional Chinese medicine specimens based on a pre-constructed specimen collection platform, obtain corresponding specimen sample images, and record the collected specimen metadata; Store it in a pre-constructed specimen database; Step 2: Extract features from the collected specimen sample images, and construct corresponding specimen feature maps in combination with the constructed specimen database; Step 3: Identify the specimen attributes of the obtained specimen feature maps, and construct a corresponding informatization model of traditional Chinese medicine specimens based on the specimen attribute identification results; Step 4: Perform multi-dimensional feature matching processing on the constructed informatization model of traditional Chinese medicine specimens to obtain corresponding specimen identification results; Step 5: Simulate the growth environment of corresponding traditional Chinese medicine specimens respectively based on the specimen identification results and the collected environmental information, and evaluate the accuracy of the corresponding specimen identification results based on the growth environment simulation results to obtain corresponding credibility indicators; Step 6: Conduct multi-dimensional comprehensive evaluation on the credibility indicators of the specimen identification results, and integrate the evaluation results with the traditional Chinese medicine specimen information to obtain corresponding informatized traditional Chinese medicine specimen data archives.

2. The method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning according to claim 1, wherein The process of performing multi-dimensional data collection on traditional Chinese medicine specimens includes: Construct a specimen collection platform; the image acquisition terminal performs image collection on the traditional Chinese medicine specimens to be collected with a black and white bottom plate in different fill light modes respectively based on pre-set collection requirements, and obtains corresponding specimen sample images and specimen sample videos; Synchronously collect geographical information based on a positioning terminal; record the metadata information corresponding to each specimen sample image, and perform image association with the collected sample image set to generate specimen metadata; upload and store the collected specimen metadata in a pre-constructed specimen database.

3. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 2, wherein, The fill light mode refers to the natural light mode and the fluorescence mode provided by the fill light terminal. The natural light mode refers to guiding the environmental light source to irradiate the specimen through the natural light guiding structure built in the fill light terminal; the fluorescence mode refers to using the built-in ultraviolet LED array with adjustable wavelength and multi-band infrared light source, so that the corresponding image acquisition terminal performs image collection on the traditional Chinese medicine specimens to be collected in multi-spectral bands based on ultraviolet light and infrared light respectively.

4. The method for collecting and informatizing traditional Chinese medicine specimens based on image machine learning according to claim 3, wherein The process of extracting features from the obtained specimen sample images includes: Obtain the collected specimen sample images, perform geometric correction on the corresponding specimen sample images based on the scale ruler on the corresponding black and white bottom plate during the image collection process, and perform polynomial color correction and white balance adjustment on the geometrically corrected specimen sample images; obtain corresponding initial sample images; Perform local mean filtering processing on the obtained initial sample images; after the local mean filtering is completed, obtain the specimen type corresponding to the corresponding traditional Chinese medicine specimen, and perform adaptive enhancement on the corresponding traditional Chinese medicine specimen based on it to obtain corresponding enhanced specimen images, and obtain corresponding standard specimen images based on them; Extract feature points in the corresponding standard specimen images based on the improved SIFT algorithm to obtain corresponding image feature points; Mark the feature points on the initial sample images after adaptive enhancement based on the obtained image feature points, and generate corresponding feature distribution maps; Based on the obtained characteristic distribution map, morphological features, texture features, and color features are extracted from the corresponding standard specimen images to obtain corresponding morphological feature vectors, texture feature vectors, and color feature vectors. At the same time, a 3D convolutional neural network is used to process the collected specimen sample videos to capture the change characteristics of traditional Chinese medicine specimens in the time dimension and obtain corresponding video dynamic features, such as the fluorescence decay process or the appearance change of specimens at different angles. The morphological feature vectors, texture feature vectors, color feature vectors, and video dynamic features obtained under different spectral bands are fused with the collected geographical location information to obtain corresponding comprehensive feature vectors.

5. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 4, wherein The process of extracting feature points in the corresponding standard specimen image based on the improved SIFT algorithm includes: Performing Gaussian convolution processing on the standard specimen image based on Gaussian convolution kernels at different scales to obtain Gaussian convolution images at different scales; performing image difference calculation on the Gaussian convolution images at adjacent scales to obtain corresponding difference images; obtaining the pixel values corresponding to the pixel points in the corresponding difference images and comparing them with the pixel values corresponding to all pixel points in the neighborhood pixel set, and determining whether the pixel value corresponding to the corresponding pixel point is a local extreme point based on the comparison result. If not, no other operations are performed. If so, the corresponding pixel point is marked as a candidate key point. Furthermore, obtaining the curvature coefficient corresponding to each candidate key point and comparing it with a preset curvature threshold. If the curvature coefficient is not greater than the curvature threshold, the corresponding candidate key point is marked as an image feature point. If the curvature coefficient is greater than the curvature threshold, the corresponding candidate key point is discarded.

6. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 5, characterized in that, The process of identifying the specimen attributes of the obtained specimen characteristic map and constructing a corresponding information model of traditional Chinese medicine specimens based on the specimen attribute identification results includes: Based on the existing traditional Chinese medicine knowledge base, obtaining a traditional Chinese medicine attribute data set related to traditional Chinese medicine; extracting attribute features of the corresponding traditional Chinese medicine specimens based on the obtained specimen characteristic map to obtain corresponding specimen attribute features. Associating the obtained specimen attribute features with the traditional Chinese medicine attribute data set and constructing a corresponding classification framework based on it; integrating the classification framework with the obtained specimen characteristic map to obtain a corresponding information model of traditional Chinese medicine specimens. The information model of traditional Chinese medicine specimens includes the characteristic data, attribute association, and classification structure of traditional Chinese medicine specimens, as well as relevant traditional Chinese medicine knowledge.

7. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 6, wherein, The process of constructing the classification framework is: visualizing the association relationship between the key attribute information in the traditional Chinese medicine attribute data set and the specimen attribute features as an attribute association network, and marking the association strength and direction between each key attribute information and the specimen attribute features. Constructing a corresponding classification framework based on the attribute association network.

8. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 7, wherein, The process of obtaining the specimen identification result includes: Matching the characteristic information of the information database of traditional Chinese medicine specimens corresponding to the corresponding traditional Chinese medicine specimens with known traditional Chinese medicine specimens to generate corresponding initial matching results, and the initial matching results include multiple sample matching items. Sort the known traditional Chinese medicine specimens based on the comprehensive similarity coefficient to obtain the corresponding similarity ranking sequence; pre-screen the known traditional Chinese medicine specimens in the similarity ranking sequence based on a preset similarity threshold, and use the known traditional Chinese medicine specimens with a comprehensive similarity coefficient higher than the similarity threshold as candidate specimens; summarize all the obtained candidate specimens to obtain the corresponding specimen identification result, and the specimen identification result includes information such as the specimen ID, name, and similarity data of the known traditional Chinese medicine specimen; the similarity data includes the similarity calculation results and the comprehensive similarity coefficient in multiple dimensions.

9. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 8, characterized in that, The process of obtaining the credibility index includes: Extract parameters from the traditional Chinese medicine specimen information database to obtain the ecological environment data related to the corresponding traditional Chinese medicine specimen; and based on it, extract the range of environmental parameters suitable for the growth of the traditional Chinese medicine specimen to obtain the corresponding specimen growth environment requirement parameter set; Based on the pre-constructed ecological adaptability model, obtain the adaptability of the corresponding traditional Chinese medicine specimen to different environmental factors to obtain the corresponding ecological adaptability index; Based on the ecological adaptability index, and in combination with the specimen growth environment requirement parameter set, simulate the growth state and performance of the corresponding traditional Chinese medicine specimen under different environmental factors, and record the key data points during the simulation process; generate a complete specimen ecological adaptability data set; Obtain the metadata of the collected specimens, based on it, obtain the environmental data recorded during the collection of the corresponding traditional Chinese medicine specimen, and perform standardization processing on it to obtain the corresponding environmental condition parameter set, and based on it, obtain the corresponding environmental specimen ecological data set; Obtain the environmental specimen ecological data set and the specimen ecological adaptability data set obtained, and evaluate the matching degree between the two to obtain the corresponding environmental matching degree index; Integrate the obtained environmental matching degree index to obtain the corresponding specimen ecological environment evaluation data; Based on the sample identification result, obtain the ecological growth requirements corresponding to each candidate specimen, and compare them with the obtained specimen ecological environment evaluation data for consistency; If the two are inconsistent, discard the corresponding candidate specimen; if the two are consistent, obtain the matching degree between the environmental parameters and the ecological growth requirements corresponding to each candidate specimen obtained from the corresponding sample identification result to obtain the corresponding environmental verification result; Based on the environmental verification result, correct the sample identification result corresponding to the corresponding candidate specimen. After the correction is completed, based on the traditional Chinese medicine expert knowledge base, obtain the classification rules and identification key points of each type of traditional Chinese medicine, and based on it, obtain the evaluation factors of the corresponding specimen identification result, and perform weighted summation on the obtained evaluation factors to obtain the corresponding credibility index.

10. The method for collecting and informatization processing of traditional Chinese medicine specimens based on image machine learning according to claim 9, wherein, The process of constructing the information-based traditional Chinese medicine specimen data file includes: Define the identification standard of the credibility index, and based on it, obtain the credibility level corresponding to the corresponding specimen identification result; obtain the traditional Chinese medicine specimen information model corresponding to the candidate specimen in the corresponding sample identification result, and compare it with the traditional Chinese medicine specimen information model corresponding to the traditional Chinese medicine specimen to be collected to obtain the corresponding data coverage rate and accuracy rate, and generate the corresponding data integrity evaluation value based on it; Integrate the obtained credibility level and data integrity evaluation value, and synchronously integrate the relevant information involved in the corresponding traditional Chinese medicine specimens to generate a digital file of informatized traditional Chinese medicine specimens.

Images